Adapt

or Perish: The Governance of Climate Change in Africa Triggers of Institutional Adaptation to Meet the New Challenge of Rising Hydro-Meteorological Disasters in Africa, 1995-2010 Comparative Insights from Mozambique, Uganda and Senegal

Arame TALL PhD Candidate African Studies Johns Hopkins University School of Advanced International Studies April 2012

TABLE OF CONTENTS
AKNOWLEDGEMENTS
PREFACE

3 4

INTRODUCTION PART I: BACKGROUND CHAPTER 2: TRIGGERS OF INSTITUTIONAL CHANGE CHAPTER 3: WHY CLIMATE CHANGE REQUIRES NEW OR ADAPTED INSTITUTIONS? PART II: METHODS CHAPTER 4: METHODS PART III: RESULTS CHAPTER 5: CROSS-COUNTRY CLASSIFICATION RESULTS CHAPTER 6: SENEGAL CHAPTER 7: UGANDA CHAPTER 8: MOZAMBIQUE PART IV: DISCUSSION 31 33 59 104 105 133 134 144 200 254 312 7

CHAPTER 9: CROSS-COUNTRY ANALYSIS 313 CHAPTER 10: CONCLUSION, TOWARDS A THEORY OF INSTITUTIONAL ADAPTATION TO MEET THE NEW CHALLENGE OF CLIMATE CHANGE 331 BIBLIOGRAPHY 347

AKNOWLEDGEMENTS This thesis was a long odyssey. It would have never come to completion without the support and constant encouragement of the following individuals and institutions. I stand before them in deep gratitude. Mame Boury and Mamadou Tall, without whom I would have never come this far; I am grateful to the assistance of Ms. Anita Twesigomwe in the search, compilation and archiving of the newspaper clips that underpinned my events analysis for Uganda. I am thankful for the assistance of Lucia Jorge in the search, compilation and archiving of the newspaper clips that underpinned my events analysis for Mozambique. In Senegal, archivist Sammy Laris was of invaluable support to unearth, compile and catalogue newspaper clips on government responses to climate-related disasters between 1995-2010. I am indebted to Filipe Lucio of the World Meteorological Organization for his insightful comments on the Mozambique chapter, and to Alessandra Giannini of Columbia University for her time reading through the Climate Change chapter. UNDP-GEF and UNDP Mozambique country Program on Climate Change Adaptation supported and funded my fieldwork in Mozambique from March-May 2011, as part of the project document development for the Mozambique Coastal Adaptation project. The SAIS PhD Committee provided the bursary that enabled research in Senegal and Uganda. I am indebted to the Senegalese Red Cross National Society for providing the logistical support and manpower that greatly facilitated the Climate VCAs in Senegal. Last but not least, I am thankful to all of my PhD committee members, who have taken the time to read and comment on the various versions of this manuscript; I thank them for their time and intellectual generosity. 3

PREFACE On August 28 2011, following two days of heavy uninterrupted rainfall, sudden landslides on the slopes of Mount Elgon in center-east Uganda buried twenty-six people alive at once in the district of Bulambuli. Ten more lives were claimed in the landslides of the days after. These followed the national landslides of just the year prior, which killed over three hundred people, in a country that had only seldom known landslides in the past. In Mozambique, Indian Ocean cyclones and tidal incursions, increasingly frequent in recent years, have displaced entire coastal communities along Mozambiques two thousand five hundred (2,500) kilometer-long oceanic interface, forcing them inland and depriving them of a much-needed livelihood, which the ocean had provided them with hitherto, compelling them to adapt to a new life as agriculturalists. West of the continent, in Senegal, a formerly drought-ridden country of the Sahel until the end 1980s, flooding has been progressively claiming more lives since the mid-1990s, creating damage only faintly estimated by official reports. All across Africa, climate-related disastershereafter referred to as Hydro- Meteorological Disasters (HMDs) have been on the rise since the 1990s; and their impacts have begun to wreck havoc in communities at the frontlines in Africas rural areas and urban centers alike1. Many attribute this increase in the frequency and magnitude of climate-related disasters in part to global anthropogenic Climate Change2, and contend that with more climate changes to come a result of the inertia in the climate system (mostly

1 United Nations Development Programme (UNDP), Human Development Report 2007/2008, Fighting climate

change: Human solidarity in a divided world (New York: McMillan, 2007), 75-88.
2 Intergovernmental Panel on Climate Change (IPCC), Climate Change 2007: Impacts, Adaptation and

Vulnerability, Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, ed. M.L. Parry et al. (Cambridge: Cambridge University Press, 2007).

resident in the oceans), which will continue to react for many more years to current- day atmospheric carbon stocks we are already committed to more HMDs are expected to occur, with sizeable impacts over Africa3. In the continent least prepared to address the impacts of climate-related changes, these new threats are a major source of concern. How are African countries governing the new challenge of a fast changing climate? Which institutions and policy frameworks have countries endogenously devised to confront rising HMDs? How have relationships within society, between governed and governors in Africa, changed or been reshaped as a result? How are they likely to change in response to continued climate fluctuations? Fundamentally, what makes African countries take Climate Change seriously, and place the issue high on the political agenda? The present dissertation seeks to answer this last essential question. We already know a lot on what needs to be done and have plentiful advice readily available for governments regarding the design of institutions effective for Climate Change Adaptation (CCA) and Disaster Risk Reduction (DRR), including establishing early warning systems, embedding institutions mandated to tackle climate risks under powerful ministries and so forth (see World Resources Report 2010-11)4. However the other side of the causality arrow- what MAKES governments adopt these effective institutions, uptake all of this advice and engage in institutional change to better address and deliver on the new demand for effective CCA-DRR policies that is where a critical gap in understanding lies, one which this dissertation seeks to address. As the issue of Climate Change moves to the fore of national, African and international agendas, a framework is urgently needed to understand the complex
3 Intergovernmental Panel on Climate Change (IPCC), Managing the Risks of Extreme Events and Disasters to

Advance Climate Change adaptation (SREX), Special Report of the Intergovernmental Panel on Climate Change, ed. Field et al. (Cambridge: Cambridge University Press, 2012), 8-9. 4 World Resources Institute (WRI) in collaboration with United Nations Development Programme, United Nations Environment Programme, and World Bank. World Resources 20102011: Decision Making in a Changing ClimateAdaptation Challenges and Choices, ed. P. S. Angell et al. (Washington, DC: WRI, 2011), 70-79.

interactions between climate and societies in Africa, and attempt to predict how precisely African societies, livelihoods and governance structures will respond to increasing climate variability. To answer this question, we focused on rising HMDs as a proxy for Climate Change impacts in Africa, and investigated national governments policy responses to HMDs during the period 1995-2010, when HMD frequency began to increase over Africa. In each of the country cases where we conducted research, Mozambique, Uganda and Senegal, we then asked: what prompted governments to adopt the policy and institutional frameworks that they did to address rising climate risks? Positing disaster management as a public good, we analyze why some countries across the continent have provided the public good of disaster prevention, towards the fulfillment of their constitutional mandate of protecting citizens against all crises, including climate ones; whereas others have not. What this dissertation unravels is that in many instances, institutional adaptation to meet the new challenge of Climate Change is an issue eminently path-dependent. In the countries of Africa where high-frequency HMDs in recent years have wrecked havoc (accidents of history), therein governments have had no choice but to address the issue. When the imperative of adaptation met convinced leaders and trusting donors willing to support the countrys newfound desire for effective national policies, plans and management of climate risks, therein the governance of Climate Change was improved. Thus the desire for change has to come from within, stemming from a strong national demand for improved CCA-DRR, which in turn led to the development of endogenous institutions to seriously confront rising climate risks and manage new climate opportunities. Under these circumstances, effective CCA-DRR institutions have obtained. Understanding these political underpinnings of policy and institutional change is an important pre-requisite for all the activists and programs promoting Climate 6

Change adaptation on the continent, and expecting governments to uptake these programs. Climate Change is one of the defining issues of our time. In the words of the UNDP, no issue merits more urgent attention or more immediate action5. This is so even more for developing regions such as Africa, projected to bear the brunt of its impacts in the 21st century. A framework is urgently needed to understand how Climate Change will impact the human, socio-economic and political systems of developing regions, as well as incentive structures within them for institutional change. This dissertation contributes an important stone to that edifice.
5 UNDP, Human Development Report 2007/2008, 1

Chapter 1: INTRODUCTION
When the cyclones come, all my cassava harvest is destroyed. I cant go fishing because the cyclone comes from the sea. I rebuild the roof of my house with coconut tree leaves, but if the cyclone persists for days, the roof flies off again and rains flood into my house. What am I to do against all these calamities? I would like to open a small shop selling pastries in front of my house, to have the means to buy a zinc roof to put over the heads of my family, but means to start up are limited. God, please help us. - Cyclone afflicted resident of Macuacuane Village, Pebane, central Mozambique

1.1 Problem Statement 1.1.1 Climate Change & Rising Disasters in Africa

Climate Change will very likely lead to an increase in the frequency of heavy rainfall events worldwide6, with medium confidence that anthropogenic, or man-made, influences have contributed to the intensification of extreme precipitation on the global scale7. We also now know with quasi-certainty that Climate Change will bring over most land areas a warming of extreme daily minimum and maximum temperatures on the global scale8, with warmer and fewer cold days/nights, and warmer and more frequent hot days/nights. Warm spells and heat waves will very likely be more frequent as well, with impacts on all sectors of human activity, from agriculture to water resources, human health, industry and tourism9. More intense tropical cyclones are also likely10. Finally, it is likely that currently observed extreme coastal high water and rise in mean sea level are a result of anthropogenic Climate Change, and will continue to rise as Climate Change ensues. Climate Change impacts related to changes in patterns of extreme events, based on projections to the mid- to late 21st century, not withstanding any changes or new
6 IPCC, Fourth Assessment Report, 8-12 7 IPCC, SREX, 13 8 Ibidem 9 Ibidem 10 Ibidem

developments in adaptive capacity, are hence significant and non negligible11. Though the attribution of single extreme events to anthropogenic climate change versus natural climate variability is challenging, it can be affirmed that a changing climate leads to changes in the frequency, intensity, spatial extent, duration, and timing of extreme weather and climate events, and can result in unprecedented extreme weather and climate events12. Thus with Climate Change, a new era of more erratic fluctuations in weather and climate patterns is to be expected, including on the continent of Africa, where impacts of such likely changes in climate and weather are most feared, due to the continents limited means to withstand them, as well as the high dependence of a large majority of its population on natural climate factors temperature and rainfall chiefly for livelihood and development. Indeed, in Africa impacts of a changing climate on local communities dependent on climate for their livelihoods and sustenance are large sources of concern. An average 57% of Africas active population is employed in a rain-fed agricultural sector that is highly sensitive to rainfall and temperature fluctuations. Only 6.8% of arable land in Africa is irrigated13. Furthermore, increasing population densities in ill-planned settlements at the peripheries of Africas urban centers, in Dakar, Accra, Lagos, Nairobi and Johannesburg, are directly exposed to the vagaries of changing rainfall patterns14. When climate-related hazards strike in Africa, they generate human development setbacks that have monumental social ramifications15. They jeopardize progress towards achievement of the Millennium Development Goals and generally force poor households to forfeit meager assets in the process of coping with climate
12 IPCC, SREX, 11 13 Food and Agricultural Organization of the United Nations, FAO Statistical Yearbook 2009, (Rome: FAO, 2009),

11 IPCC, Fourth Assessment Report, 17-18

accessed April 04, 2012. http://www.fao.org/economic/the-statistics-division-ess/publications- studies/statistical-yearbook/fao-statistical-yearbook-2009/en/ 14 Pelling, Mark and Ben Wisner: Disaster Risk Reduction: cases from Urban Africa. London: Earthscan, 2009 15 Tall, Arame, Climate forecasting to serve communities in West Africa, Procedia Environmental Sciences 1 (2010): 421-431, accessed online December 1, 2011. http://www.sciencedirect.com/science/journal/18780296

related disasters16. This has prompted UNDP to declare that Climate Change is hampering efforts to deliver the MDG promise17. The projection that, with Climate Change, Africa will be faced with more frequent extreme weather and climate events is a worrisome prospect. In light of all of these reasons, Africa has been identified by the Inter-governmental Panel on Climate Change as the second region most vulnerable to Climate Change impacts in the world, immediately after the Arctic polar zone, and before small island nation states and Asian/African mega-deltas; Africa, because of the continents low adaptive capacity and projected climate change impacts18. More frequent climate and weather extremes are not just a future scenario in Africa however- many African countries have already begun facing them in recent years. Over the decade of the 1990s alone, Africa as a whole has experienced a surge in the number of heavy rainfall events and floods hitting the continent19. This constitutes a sharp contrast with the prolonged droughts of the 1970s-80s, leading many scientists to observe that a greening is occurring in areas such as the Sahel, which suffered from severe droughts for two decades and now enjoys plentiful rainfall on average. This return of the rains has been far from a being a welcome godsend however. When the rains returned, they came back with full force, carting along a plethora of flash floods, water-borne epidemics and dry spells scattered between the bouts of plentiful but ill-distributed rainfall. Recent empirical evidence emanating from across Africa confirms this assertion: between 2007-2009, disaster interventions across Africa have increased by a factor
16 As noted by Emmanuel Skoufias, poor households are typically less equipped to deal with shocks [] In the

absence of an effective public safety net system [as is the case throughout most of Sub-Saharan Africa], poorer households may use coping strategies that ultimately prevent them from ever escaping from poverty [] for example, selling their productive assets, such as draft animals, or taking their children out of school. World Bank, World Development Report 2003: Sustainable Development in a Dynamic World (Washington D.C.: Oxford University Press, 2003), 1088 17 UNDP, Human Development Report 2007-08, 28 18 IPCC, Fourth Assessment Report, 13 19 International Federation of the Red Cross/Red Crescent Societies (IFRC), World Disasters Report 2009: Focus on Early Warning, Early Action (Geneva: IFRC, 2009), 172

10

of 820. In 2007 alone, the continent experienced close to 30% of the worlds hydro- meteorological disasters the majority of which were flood and flash flood disasters, causing about 2 million victims in East and Central Africa during the January floods, and more than 2.6 million victims between West and East Africa during the July-August floods. In 2007, floods in Sudan, where damages were thoroughly documented, caused more than US$ 300 million in damages, whereas in Madagascar cyclone Indhala caused over $240 million worth of destruction21. In 2008, floods destroyed homes, infrastructure and crops, killing over 300 people in West Africa22. Droughts, floods, pest infestations, rising sea levels, storms and cyclones are increasingly reducing opportunity and wrecking havoc in communities all across the continent23. Whether a partial manifestation of anthropogenic Climate Change as suggested by a growing number of climate scientists24, or a consequence of higher localized human vulnerability caused by higher exposure (for instance increasing human densities in low elevation coastal zones in Africa; see McGranahan et al.)25, weaker safety nets and social protection26 and declining agrarian fortunes27, or yet the mere spurious outcome of better disaster reporting on the part of African countries; an upward trend in climate-related disasters is noticeable across the continent, as evidenced by the significant rise in the number of reported climate-related disasters since the mid-1990s (see fig. 1.1).

20 IFRC, World Disasters Report 2009, 172 21 The Center for Research on the Epidemiology of Disasters (CRED), Annual Disaster Statistical Review, The

numbers and trends 2007 (Brussels: CRED, 2008)

22 Braman, L., Early Warning, Early Action, an Evaluation, Report to the IFRC/Red Cross Red Crescent Climate

Centre (2009), Accessed April 4, 2012 http://www.climatecentre.org/downloads/File/ewea_an_evaluation_of_ifrc_west_and_central_africa.pdf 23 UNDP, Human Development Report 2007-08, 73 24 See IPCC, SREX; and IPCC, Fourth Assessment Report, 13 25 McGranahan, G., D. Balk and B. Anderson, The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones, Environ Urban 19 (2007): 1737 26 Ellis, Frank, Stephen Devereux and Philip White, Social Protection in Africa, Northampton: Edward Elgar Publishing, 2009 27 Bryceson, D. F., The Scramble in Africa: Reorienting Rural Livelihoods, World Development (2002): 725739

11

Fig. 1.1: Number of climate-related disasters reported in Africa since the 20th century: A sharp increase from the mid-1990s (based on EM-DAT data; source: author). Note: Are classified as climate-related disasters: floods, pest infestations, droughts and storms/cyclones.28

These Hydro-Meteorological Disastershereafter referred to as HMDsare the principal object of this dissertation. Using rising HMDs as a proxy for projected CC impacts in Africa, we ask: how are African countries currently addressing the challenge of a changing climate? Which governments are doing what in the face of rising HMDs, and what policies and institutional frameworks have they endogenously devised to address the new challenge? Which factors have prompted them to adopt the policies that they did? In instances where policy change has occurred, what has triggered it? In general, what factors prompt a country to become a climate disaster averter, versus a disaster responder or a firefighter?
Vulnerability. Hazard * Vulnerability
28 We follow here the definition of a Disaster as the conjugation of a naturally driven Hazard and human-induced

DISASTER= Capacity The character and severity of impacts from climate extremes depend not only on the extremes themselves but also on exposure and vulnerability. Exposure and vulnerability are key determinants of disaster risk and of impacts when risk is realized (IPCC, SREX, 2012). A disaster can thus be conceived of as a serious disruption of the functioning of a community or a society causing widespread human, material, economic or environmental losses, which exceed the ability of the affected community or society to cope using its own resources. It results from the combination of hazards, conditions of vulnerability and insufficient capacity or measures to reduce the potential negative consequences of risk (United Nations International Strategy for Disaster Reduction, 2009)

12

We circumscribe our analysis to HMDs that have occurred between 1995 and 2010, period when HMDs have begun to rise in Africa (see fig. 1.1) and over which more reliable data is available. We analyze what African countries have done hitherto to address these disasters, as a means to understand their response patterns and surmise how they may continue to manage HMDs into the future and rise up to the new challenge of Climate Change, as impacts become more pronounced over the 21st century. Against the context of changing climate risks and rising extreme weather events, projected to only keep rising in the 21st century, understanding the policies in place to address impacts of currently occurring HMDs in Africa is a pressing issue, important to appraise national responses to the impacts these disasters are already exerting on local livelihoods, societies and political systems, and to attempt to predict outcomes if climate disasters continue to rise. On the ground, across Africa, communities and countries have already begun to contend with climate changes we are committed to, while bracing for those yet to come. Therein lies the relevance of studying how one of the worlds most vulnerable regions to climate change and variability is already impacted and will continue to be impacted by global Climate Change, and what endogenous institutional solutions are being attempted to thwart its most immediate and damaging impacts on vulnerable communities. 1.1.2. CC Governance: Climate Shock Management & Constitutional Design Despite the magnitude of the Climate Change challenge, and its implications for human development in Africa, it is unfortunate to note that scholarly work on national climate change adaptation policies has been severely wanting. Not enough political scientists have addressed the issue of Climate Change adaptation, which has been by and large left to climate scientists, geographers and physical science 13

researchers; whereas if the latters prediction of steadily increasing climate extreme events turns to be accurate, we may be witnessing over the next few decades some of the largest reversals in human development ever witnessed29. This is even more surprising in light of the findings of the World Banks global track study, which concludes that by 2020, the annual costs of adaptation for developing countries will range from $75 billion to $100 billion/year. Of this amount, the average annual costs for Africa would approximate $18 billion/year30. Just as when the Human Immuno-Deficiency Virus (HIV) threat emerged on the continental scene in the decade of the 1990s, the 21st century is witness to the new Climate Change threat, and we do not know what to make of its projected impacts and opportunities, nor how to analyze and forecast its consequences on already fragile African political and social systems. A framework is thus urgently needed to understand the impacts of Climate Change on Africas vulnerable communities and countries, and devise preventive policy solutions realistic in the context of existing governance structures. In this thesis, we bring Climate Change back into the realm of policy scrutiny. We delve into the systemic factors that spur governments to take Climate Change seriously, and confront it effectively to climate proof their development gains and engage on a track of climate-compatible development. Our approach to test our framework is to look into how already occurring climate changes impacting African countries are currently managed, as a way of understanding patterns of response, and predicting how they will likely impact and be managed into the future. Looking into the past to predict how the future will likely unfold.

2010), 19-25.

29 UNDP, Human Development Report 2007/08, 1. 30 World Bank, Economics of Adapting to Climate Change, Synthesis Report (Washington DC: The World Bank,

14

We focus on one manifestation of Climate Change to do this: climate-related extreme events; and hone in specifically on how countries at risk are currently addressing these, as a means of inferring how they will respond to future climate hazards, assuming a perfect line of continuity in the disaster management (DM) policies in place. In the cases where the perfect line of continuity is broken, we look into the systemic reasons that led to a change in DM policy, in order to derive insights into what factors prompt governments to adopt better policies to improve their management of climate shocks. Two strategic reasons prompt us to focus on the governance of climate-related hazards as a proxy to analyze the governance of Climate Change in Africa: 1) A confident projection for Africa that future Climate Change will be an exacerbation of current climate variability. As such, gaining a better understanding of how governments currently respond when climate and weather fluctuations occur (i.e: climate-related hazards) allows us an introspection into how governments will most likely respond to future hazards again assuming a line of continuity in governance of these climate risks. 2) A reasoned possibility that Climate Change will lead to an increase in the frequency of extreme weather events. There remains a lot of uncertainty regarding whether climate-related threats will increase with CC; current scientific evidence suggests this (IPCC 2007). The death toll and social, economic and psychological damages generated by climate-related or hydro- meteorological disasters at their current frequency levels are enough however to put their governance of as a serious object of study, even if future scientific breakthroughs do not demonstrate/project an increase in their frequency.

15

What we already know is that Africa is characterized by institutional and legal frameworks that are, in some cases, insufficient to deal with environmental degradation and disaster risks (see Sokona and Denton, 2001; Beg et al., 2001)31. What we have yet to find however is how much degradation and risks will countries in Africa absorb before they engage in meaningful institutional reform? What are the general triggers of such institutional reforms? A fundamental related question of interest is what can be done to curtail the impacts of CC on African countries, by these countries themselves, and ensure effective governance of Climate Change risks and opportunities, in order to thwart their negative impacts on the development endeavor. A preliminary answer seems to be: build resilience to current climate-related disasters, since they will likely only go crescendo with CC32. As we will find soon, our preliminary findings however indicate that only a very small minority of Africas fifty-countries is doing so. Most importantly however, we are interested in knowing why countries in Africa adopt the climate disaster risk management policies that they do. What prompts or deters states from adopting socially beneficial climate disaster risk reduction (CCA- DRR) policies? Which factors determine African governments response to Climate Change? Is it limited knowledge or technical expertise on national vulnerability to climate risks? No additional financial resources to tackle climate risks and opportunities? Or simply no political buy-in to govern climate change and avoid its detrimental consequences? We get to these reasons and detail them in our country findings chapters.
31 In IPCC, Fourth Assessment Report, 453-4.

32

United Nations International Strategy for Disaster Reduction (UNISDR), Hyogo Framework for Action 2005-

2015: Building the resilience of nations and communities to disasters (Geneva: United Nations, 2007). Accessed March 29, 2011. http://www.preventionweb.net/files/1037_hyogoframeworkforactionenglish.pdf

16

1.2. Research Question and Hypotheses 1.2.1. Research Questions My dissertation aims to answer the fundamental question: How are countries in Africa governing rising Hydro-Meteorological Disasters (proxy for Climate Change), and what factors determine African governments response to climate shocks? The goal of the present research is to test whether X (hypothesized Independent Variables) causes Y (adoption of socially beneficial institutions to meet the new challenge of CC). Is GDP/capita, disaster frequency or an accident of history (hypothesized Independent Variables) the main driver of the observed differences in climate risk management institutional frameworks across Africa, or do alternative hypotheses hold more explanatory power? What makes African countries take Climate Change and its associated risks seriously, and place it high on the political agenda? Extensive literature exists on what needs to be done, with widespread policy advice, guiding interested governments in the adequate design of institutions effective for Climate Change Adaptation (CCA) and Disaster Risk Reduction (DRR), including establishing early warning systems, embedding institutions mandated to tackle climate risks under powerful ministries and so forth (see World Resources Report 2010-11)33. However we have yet to understand what MAKES governments adopt these effective institutions, uptake all of this advice and engage in institutional change to better deliver on the new demand for effective CCA-DRR.
33 World Resources Institute (WRI) in collaboration with United Nations Development Programme, United

Nations Environment Programme, and World Bank. World Resources 20102011: Decision Making in a Changing ClimateAdaptation Challenges and Choices, ed. P. S. Angell et al. (Washington, DC: WRI, 2011), 70-79.

17

Our general question has two sub-components: 1) How do countries in Africa confront rising HMDs? What policies have African governments experimented with to confront rising climate-related disasters between 1995-2010? What is the current institutional architecture in place to govern climate-related disasters at national-level? Where is climate finance coming from and flowing to? 2) What factors prompt governments in Africa to effectively govern climate- related risks? What factors prompt/deter a State from adopting socially beneficial institutions to address the new demand for climate change adaptation disaster risk reduction (CCA-DRR)? What incentives make countries adopt policies that prevent HMD impacts and climate disaster-proof development gains? What are enabling and limiting conditions? Under which institutional, political and socio-economic circumstances does a country adopt CCA-DRR measures? What have been the pathways for pressure/change? More generally, what makes governments adopt socially beneficial policies and institutions? What are obstacles to achieving socially beneficial institutions in the context of Africas governance? These two subsets serve as the research questions I will be specifically investigating. 18

1.2.2. Hypotheses
RQ1: Which policy responses have African countries provided to govern rising HMDs in 1995-2010?

H1.1 : Most African countries satisfy with ex-post response when they are hit by HMDs, because they do not have or apply pre-defined procedures for disaster management.

H1.2: Current national climate adaptation/disaster management policies are not aligned with priority local adaptation needs on the ground, because a disconnect exists between national policy-making on CCA and local realities.

RQ2: What factors prompt States to establish institutions for effective

CCA-DRR?

What incentives prompt States to adopt the policies and institutional frameworks that they do in response to rising climate-related risks? Here, we probe the incentives and pressure channels for effective CCA-DRR in the context of Africas governance. An incentives analysis will enable such a probing, with a deep look into the incentives systems that modulate the reality of African governance of CC. Our working hypotheses are as follows. H2.1: Climate-related disasters reshape political relationships at the sub- national-level, because they create new channels of pressure from the base that make the government pay attention to the needs of local citizens. 19

 H2.2: Severe past disaster(s) prompt(s) governments to engage in ex-ante preparedness for HMDs, since wherever there have been multiple severe disasters, national leaders have reformed national climate-disaster management policies/institutional frameworks. H2.3: Donor pressure drives policy change in national climate disaster management, through increased funding earmarked for CCA-DRR.

H2.4: Government capacity matters, because poor countries are not able to carry out ex-ante disaster preparation.
This last hypothesis runs as follows: State capacity (fiscal) determines the type of disaster management policy in place in the country. Wealthier, more capacitated countries engage in ex-ante disaster preparation and generate better disaster outcomes, whereas poorer, weaker countries only respond to disasters and manage disasters poorly. H2.4-a: Government Capacity is positively related with Disaster management policy (yes, capacity does matter) H2.4-b: Capacity does not matter as much as Institutions do (strength of relationship between Institutions and Disaster Policy more significant than strength of relationship between Capacity and Disaster Policy).

The present thesis is dedicated to responding to these specific research questions and testing the above-outlined hypotheses. Figure 1.2 summarizes our main hypotheses.

20

Cyclone Flood Drought Storm Landslide

Competing Independent Variables (cause): Dependent Variable (effect):

HMD frequency/ magnitude Donor pressure for improved CCA-DRR Climate Finance: Aid earmarked for CCA-DRR State capacity: GDP/ capita Dedicated leadership Key political bastion at risk Extensive Media coverage of disaster impacts Institutional change in the management of climate risks: adoption of socially benevicial institutions to address the new challenge of CCA-DRR

Climate Hazard Hits

Figure 1.2: Hypothesized Independent Variables, and pathway of cause and effect.

1.3. Methods In order to test our hypotheses, we focus on one analyzable sub-set of projected Climate Change impacts for Africa: risks of extreme weather events, which if un- buffered, translate into large-scale disasters. Our hypothesis-testing is rooted in empirical evidence emerging from across Africa, with cases from Senegal, Uganda and Mozambique, where we analyze the performance of current governance schemes at the national, sub-national and local levels to manage climate risks and address the adaptation needs of communities at the frontlines of Climate Change impacts, as a proxy for assessing how well these systems will manage exacerbated climate risks in the future, assuming perfect policy continuity.

21

We circumscribe our analysis to the period beginning in 1995, when a consistent rise in HMDs began to be observed in Africa (see fig. 1.1), up until 2010, when enough reliable data is available in each country. A three-phase research and analysis plan enabled us to answer our research question: phase 1 of this research is dedicated to identifying policies in place to confront HMDs across Africa, while phase 2 zooms in at community-level to assess the adequacy of national policies to address local adaption needs, and phase 3 delves into an incentives analysis of why a given policy in each country came to be. We used a mixed design (of both quantitative and qualitative methods), involving two units of analysis: the household and the nation-state. We avoided the ecological fallacy by drawing conclusions in each of the parts of our analysis that only apply to the units of analysis under consideration in that part. For instance, the goal of phase 1 is to identify the various policies in place across Africa to address climate-related disasters; the conclusion from that first part were then used to select cases illustrating each type of policy. The goal of this endeavor differs from that of phase 2s, which seeks to assess which of the policies categorized in phase 1 is most aligned with local communities adaptation needs, zooming in on households within communities that were severely hit by disasters (hence the change in units of analysis). The insights reached in this section were only applied to the communities under consideration. Finally, we returned to the nation-state in our final analysis phase to identify what factors in the context of Africas governance reality could incentivize governments across Africa to adopt the latter policy. The different goals of each research phase justify the different units of analysis. 1.3.1. A note on Sampling Selection of country case studies: Country cases have been selected based on a continent-wide classification of the different climate-related disaster management policies in place at the national-level. 22

For the purposes of this research, we have hypothesized three distinct climate disaster management policy types across Africa: 1) The Policy of ex-post response to disasters (the Firefighters): countries choosing not to, or unable, to address disasters until they occur. Countries in this category solely respond to disasters after they have occurred, mobilizing any personnel available and sending them to disaster sites. For this reason we label them: the Firefighters. Official emergency response in the countries of this category is characterized by amateurism and improvisation, and there are no pre-established well-rehearsed procedures to follow. 2) The Policy of ex-ante preparedness for disasters (the Prepared Firefighters): Countries in this category have experienced a paradigm shift in their strategy to address disasters and have transitioned from responding to disasters only after they occur, to preparing for them before they occur (ex-ante preparation) so as to ensure more efficiency and swiftness in relief operations. 3) The Policy of pro-active disaster prevention allying ex-ante preparedness for disasters AND Disaster Risk Reduction (the Disaster Preventers): Finally countries in the third category have shifted the focus away from the disaster events towards the prevention of disasters. They focus on pre-empting disasters by reducing disaster risk/vulnerability so that hazards may not turn into disasters, using weather forecasts, early warning systems, risk/vulnerability mapping, and other related tools for climate risk management. We delve in-depth in the methods and results our cross-continental classification of African countries by climate disaster management type in chapters 4 and 5. Following our classification of all of Africas fifty-three countries, we selected a country case study from each policy type and use that country to make inferences for the entire policy group. This approach is inherently prescriptive: we am trying to arrive at a policy that can be recommended to African governments, and others 23

across the developing world scrambling to find strategies to address the challenge of climate adaptation, with an extensive explanation of the particular circumstances under which the latter policy generally arises and is implemented in a country. As one may notice, we worry about case selection. Our preliminary cross- continental classification of countries by climate-disaster management policy type in Africa serves as an attempt to use objective criterions to select the cases studies we consider in this research. As cautioned by Adam Przeworski in his contribution to the Symposium on the Role of Theory in Comparative Politics34:
We are often told to find cases that are as similar as possible, in as many aspects as possible, and then find a crucial difference that can explain what one wants to explain. In so doing however, we rig our research to reach the results we want to reach. There is a need to worry about selection, and try our ultimate best to eliminate selection bias when selecting cases and making inferences in a world that is often not exogenous [...] Is the mechanism by which our observations are produced independent of what we are trying to explain or not? Unless we pass that test, we will be making biased inferences. Our outcomes will be due to selection, not treatment. Use objective case selection criterions.

The independent mechanism that we devised to generate our country cases is encapsulated in the first part of our research: an identification of all the policies in place across Africa to address mounting hydro-meteorological disasters. The typology of policies we defined guided our country case selection process, and one country was selected to illustrate each policy type. Once selected, we delved into our country cases and used them as case studies. The only biases that were introduced when selecting between different country options within the same policy group were personal preferences on ability to conduct research in certain countries relative to others (more in Methods chapter); as well as a concern to keep selected countries as comparable as possible (similar on at least n variables) across policy types to rule out a number of competing independent variables that could explain differing institutional outcomes.

34 Kohli, Atul, Peter Evans, Peter J. Katzenstein, Adam Przeworski, Susanne HoeberRudolph, James C. Scott and

Theda Skocpol, The Role of Theory in Comparative Politics: A Symposium, World Politics 48-1 (1995), 1-49, 17.

24

Thus I got to my three principal country cases: Senegal (representing the climate disaster responders policy group), Uganda (to infer for the Disaster Prepared Firefighters group) and Mozambique (representing the effective climate disaster averters). On this basis, we conducted our cross-country comparison of triggers of institutional adaptation to meet the challenge of rising climate-related disasters. Then, in each country we asked: how were climate related shocks governed in 1995- 2010? Which national policy was in place, what was the institutional/governance landscape: how was management of climate-related hazards organized? How well did the national policy in place address local adaptation needs? What are constraints/enabling conditions prompted the government to implement the climate disaster management policy solution that it did? Selection of community case studies: In order to fully understand the impacts of disasters on communities, we also carried out extended fieldwork in the most severely impacted communities within my principal country cases. Random selection was used to select amongst different climate disaster hotspots communities most severely impacted by HMDs in each country case. Within the selected communities, a significant sample of the population was surveyed (an average of 30% of all households in each target community), following a community-based Vulnerability and Capacity Assessment (VCA) methodology toolbox that uses both a standard questionnaire gathering data about my key variables under study, as well as extended focus group interviews with male and female members of the community. Again households were selected using random sampling (every 6th household on our way, walking). We get more in depth into the VCA methodology and results of these community- level data mining in the Methods chapter as well as in country chapters. Then in each community we asked: what are local needs for adaptation to rising HMDs, in excess of local capacity to cope with their impacts? How relevant and 25

adequate are local and national plans/policies in place to meet these local needs for adaptation? Based on answers to these questions, we reach a final assessment of adequacy of current governance schemes to manage climate related risks, one that is both quantitative and qualitative. 1.4. Preliminary Findings What we have found is that communities at the frontline across Africa are already bearing the burden of coping with increased climate-related shocks, such as floods, drought and cyclones, which currently exert a heavy economic and human toll on already marginalized communities. Extreme-weather events, fruit of both intrinsic and anthropogenic disruptions in the continents highly variable climates, are an immediate threat to development in these communities and countries, and are likely to become so for many others as climate variability increases and unpreparedness, as well as dependence on climate factors, persists. What this dissertation reveals is that institutional adaptation to meet the new challenge of Climate Change in Africa is often an issue eminently path-dependent. Only when accidents of history occured in this case, high-impact HMDs that have hit countries of Africa with increasing frequency in recent years, wrecking havoc and leaving governments with no choice but to adapt or perish has institutional change ensued to deliver more effective governance of the Climate Change challenge, setting the country on a new track and institutional pathway. More generally, this thesis offers us a theory for why governments adopt societally beneficial policies and establish coherent institutional frameworks. The path forward for effective mangement of climate risks is political incentives that ensure that DRR-CCA is placed high on the national agenda, and institutionalized within a coherent national framework for Climate Change adaptation in order to give a real chance for Climate Change risks to be effectly governed in Africa. These incentives have to come from within, however, stemming from a strong national demand for 26

improved CCA-DRR, which in turn leads to the development of endogenous institutions to seriously confront rising climate risks and manage new climate opportunities. As our country cases will evidence, the national demand for effective national institutions to confront the CC challenge can only be country-driven. Once this national desire is formed, donor support in the development of coherent national policies and institutional mechanisms are needed. However, national driving of this process is an essential needed pre-requisite. Understanding these political underpinnings of policy and institutional change is an important pre-requisite for all the activists and programs promoting Climate Change adaptation on the continent, and expecting governments to uptake these programs. 1.5. Plan The present thesis is organized in four parts. Part 1 introduces the conceptual foundations of this research, part II details our methodology and experimental design, part III exposes results and country findings and the final part brings together comparative insights and conclusions emanating from cross-country analysis. In part I, we introduce concepts, and attempt to build a common understanding of the implications of Climate Change for Africa, and why it is an urgent problem, increasingly gaining prominence on national agendas across the continent. We first conduct in chapter 2 a literature review of the hypotheses that explain institutional change in general, and institutional change in response to climate shocks in particular, and strongly pin our research against the extensive theory of institutional change in Africa; before then going in-depth into the scientific basis of Climate Change (its causes, key concepts, consequences, projected impacts over Africa and uncertainties) in chapter 3. 27

Part II offers a detailed exposition of the experimental design and methods we have used to conduct our comparative analysis of climate-related disaster management policies across Africa. Chapter 4 thus exposes in detail our research design and methodology. In this part, we delve into the methods utilized to conduct our cross- country comparison of climate-related disaster policies across Africa, and present the results of this classification, basis of my country case selection. We will see a map of Africa shaded in three different colors: red for the large majority of countries that are solely disaster responders; yellow for the handful of countries that are on the track of disaster preparedness for effective response to HMDs when these occur (the Prepared Firefighters); and finally, green for the six countries that qualify as effective disaster preventers and have shifted focus away from disaster events to combating the underlying factors of disaster risk and vulnerability. Part III, the crux of this research, presents our results and encapsulates country chapters. Chapter 5 displays the results from our Africa-wide country classification by disaster management policy. Chapter 6 presents the findings from the RED policy group- Senegal serving as the principal case study to illustrate this policy type. Chapter 7 is dedicated to the YELLOW policy group, principally centered on findings from Uganda, whereas chapter 8 presents the findings from the GREEN policy group, represented by Mozambique. These country chapters summarize the insights and findings emanating from each country case, with inferences made for the entire policy group regarding the incentives that generated the specific policy type in place. Finally, part IV, the closing section of this research, draws the conclusions emanating from our cross-country comparison of disaster management policies to address rising HMDs in Africa. Chapter 9 comparatively presents the insights coming from our country cases, and outlines what the most optimal policy to address rising HMDs in Africa appears to be. We end with a closing chapter that draws the conclusions and perspectives that our comparative research offers for 28

better governance of Climate Change risks and opportunities on the continent, and for the curtailing of its socio-economic impacts. All in all, our dissertation will have generated: 1) A continental assessment of what Climate Change brings and implicates for African policy-makers and ordinary citizens; 2) An innovative methodology to assess local communities adaptation needs, in excess of their local capacities to cope the climate adapted version of the Red Cross Vulnerability and Capacity Assessments (VCAs); 3) A cross-continental picture of the different policies in place to address rising climate related risks, as well as an assessment of these policies based on their ability to respond to community adaptation needs; 4) Policy solution(s) for African countries to address the risks and opportunities created by Climate Change, as well as recommendations on where to invest the new influx of climate finance and optimal institutional arrangements to put in place in order to effectively govern climate risks. These solutions are rooted in the successful experiences of other African countries and in the reality of Africas governance incentives; 5) A deeper, more-contextualized understanding of how the projected socio- economic, political and cultural impacts of Climate Change will concretely play out in African countries, through which pathways of cause and effect, and with what implications for development prospects on the continent of Africa. A final note on conceptual definitions: All throughout this thesis, we will define and redefine. This is an attempt to build common ground and understanding, in order to bridge two very different epistemic communities that each have their own research agendas, approaches and conceptual frameworks as well as research customs, yet which my topic brings together: the classic development/social science (political science, anthropology and economics) community and the global climate/environmental change natural scientists. These two research communities 29

seldom work jointly. Despite intensifying efforts to bridge them, they remain two separate communities of practice. Thus, we will be painstakingly throughout this thesis attempt to clarify each concept we use, as a way forward to build a common understanding of the topic of climate change national institutions and institutional adaptation.

30

 PART I:

Theories of Institutional Change: New Institutionalism Revisited Towards a Theory of Institutional Adaptation to Meet the New Challenge of Climate Change Governance in Africa
31

In this first part, we conduct a literature review of the hypotheses that explain institutional change in general, and institutional change or adaptation for effective Climate Change governance in particular, and strongly pin our research against the theory of institutional change in Africa. We then go in-depth into the conceptual linkages between Climate Change Adaptation (identifying and preparing for likely future changes in climate/weather patterns), Disaster Risk Reduction (concerned with managing the root causes of disasters), the Development endeavor in Africa (dedicated to poverty alleviation and building community resilience) what we coin as the Triple bottom line of CCA- DRR-Development and Governance (sets of actors/networks, rule making systems and formal and informal rules that steer societies towards collective goals). Here, we attempt to understand why the challenge of Climate Change Adaptation requires new or adapted institutions to meet the triple bottom line of DRR-CCA-Development in Africa. This thesis addresses the fundamental question of how policies/institutions get changed. Much literature exists on institutional change, notably with multiple case studies over Africa. In our thesis, we test institutional change theory on the new quandary of institutional adaptation to meet the climate change challenge, in all of the case countries where we conducted research. What brought about change in institutions in place to address the new issue of Climate Change in these countries? Does this process of institutional change confirm existing theory on sources of institutional change? Do the theoretical hypotheses hold? Firstly, we review the literature on institutional change however and its founding precepts, before we move on to apply institutional change theory to our analysis of Climate Change institutions in Africa. 32

CHAPTER 2: THEORIES OF INSTITUTIONAL CHANGE: INSTITUTIONALISM REVISITED 2.1 Institutionalism vs. New Historical Institutionalism A focus on institutions as explanatory variables for societal outcomes is not new. Indeed, the study of institutions or Institutionalism dates back to the early 20th century after World War I, when American political scientists began systematically studying European political systems, comparing concepts such as: presidential vs. parliamentarian systems, federal vs. unitary, the structure of political parties, the origins, nature and impact of democratic, socialist and fascist regimes35. The focus of that body of literature however was formal legalistic, and did not factor in informal and dynamic variables as: interest groups, public opinion, political parties, process variables, input functions, decision-making and the processes of change (see Wiarda in Rustow and Erickson)36. When the world changed again after WWII, and the newly independent nations of Africa and Asia emerged onto the world scene, a new framework was needed to account for processes of political change in the new nations of the third-world, as compared to the development process. The field of political development was born as a response to this need. For many decades after that, as deep-seated ideological battles raged between Modernization theorists (see Almond and Verba37, Lipset38 and Rostow39) and Dependencia neo-marxists (see Samir Amin40, as well as Huntingtons 1968 critique
35 Rustow, Dankwart A., and Kenneth P. Erickson (eds.), Comparative Political Dynamics (New York: Harper

Collins, 1991)
36 Ibidem 37 Almond, Gabriel A., and Sydney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations

(Princeton: Princeton University Press, 1963)

38 Lipset, Seymour M., "Some Social Requisites of Democracy", American Political Science Review 53 no. 1 (1959):

69-105

39 Rustow, Dankwart A., Transitions to Democracy: Toward a Dynamic Model, Comparative Politics 2-3 (1970):

337-363
40 Amin, Samir, Imperialism and Unequal Development (New York: Monthly Review, 1977)

33

of developmentalism41, Gunder Frank42 and Cardoso43), the focus on institutions was lost. However, institutions were again brought back onto the table in the decade of the 1980s, as a response to neo-classical economic theory, which was perceived as a-historical. This literature that reenacted institutionalism was coined the New Institutionalism, and came to the fore in the late 1980s with dynamic proponents such as Douglas North and Mancur Olson. The new school of thought placed very strong focus on historical processes as explanations for institutional outcomes, which have bought the body of literature its name of Historical Institutionalism. New historical institutionalism is concerned with explaining through case studies and process tracing why a particular historical outcome occurs, then generalizing from it. North44 and Olson45 highlight the importance of institutions, among others. The work of Engerman and Sokoloff46 emphasizes the historical roots of institutional differences (often tracing it back to colonization). Two competing explanations however need to be distinguished within Historical Institutionalism: structural explanations of institutional change, and agent-based explanations of institutional change. Both approaches explain through case studies and process tracing why a particular historical outcome (social or political) ensues. Fundamental epistemological differences exist however in the assumptions and methods of these two approaches. Table 2.1 summarizes the differences between the two different explanations of sources of institutional change.
Table 2.1: Agent-based vs. Structuralist Institutional Change Theories under Historical Institutionalism

41 Huntington, Samuel P., Political Order in Changing Societies (New Haven : Yale University Press, 1968) 42 G. Frank, Andre, Latin America: Underdevelopment and Revolution (New York: Monthly Review Press, 1970) 43 Cardoso, Fernando H., and Faletto Enzo, Dependency and Development in Latin America, trans. by Marjory M.

Urguidi (Berkeley : University of California Press, 1979)

44 North, Douglass, Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge 45 Olson, Mancur, Dictatorship, Democracy, and Development, American Political Science Review 87 (1993):

University Press, 1990)

46

567-76 Stanley L. Engerman, and Kenneth L. Sokoloff, "Factor Endowments, Inequality, and Paths of Development among New World Economies," Economia 3 1 (2002): 41-101

34

Explanation of institutional change

AGENT-BASED Institutional Change Theory (Quantitative equilibria for mutual profit-maximizing among rational players explain institutional outcomes) Institutional outcome is most often the result of rational purposeful behavior by players History as the outcome of rational and purposeful behavior based on the idea of equilibrium, not a result of historically pre-determined outcomes Institutional change ensues from: - Agents desire for improved societal performance (or campaigning on this platform), aligns private interests with public ones, leading to socially efficient outcomes - Local demands that determine new political opportunities/constraints at the local level forcing national leaders to put local demands on the national agenda

STRUCTURAL Institutional Change Theory (Path dependence, and accidents /randomness explain institutional outcomes)

Main theorists

Arguments

Most often systemic, structural and accidental factors determine institutional outcomes Accidents of history and randomness (conjunctures), as well as path dependence, lead to particular institutional configurations Examples of conjunctures (accidents of history) are: Exogenous conjunctures: - Economic crises - Agendas of international funding agencies promoting democracy - Colonial domination/arrival - War / revolution - Natural disaster (North) Endogenous conjunctures: - Weak military capacity of a regime (Theda Skocpol) - Existence of a well-capacitated government body to manage issue - Factor endowments / countrys geography (Engerman/Sokoloff) or being an area of low population density (Herbst) or low mortality (Acemoglu et al.) - Informal behavior (e.g.: gossip in Scotts Weapons of the Weak) Catherine Boone Barrington Moore, Theda Skocpol, Sam Grindle Huntington: the precursors To a certain extent Acemoglu & North, Geddes, Herbst, Englebert, Mancur Jonson also Olson: the main advocates NB: Different from the old formal legalistic Institutionalism, focused on the comparative analysis of legal systems and structures Institutional invention is Institutions (rules of the game) constrain the possible, and has been enacted behavior of organizations (players in the in countries of Latin America. game), both politically and economically. In Individual agency matters turn, institutions are humanly devised. This (mostly when politicians have apparent tension between agency and campaigned on a platform of determinism is resolved with the theory of societal performance path dependence. improvement) and it needs to If the institutional setting and the state are be understood within the prism bad for the economy and for societal of behavioral science, and not performance at large, why don't the rational choice alone (Merilee individuals in society just change the Grindle) institutions for the better? The answer to this problem from North is that institutional change In Africa, local level is path dependent. Path dependence is based configurations of power and on the idea that it is problematic and interest matter and determine "costly" to change paths. Path dependence institutional outcomes (Boone) explains why economies get stuck in an Regimes choose the institutional framework that is not efficient institutional-building strategies or does not induce growth and at large that maximize their advantage improved societal performance. in particular (local) political 35 Institutional Change is overwhelmingly contexts. The result is differing incremental and Slow, as institutional institutional strategies of rural outcomes are in large part historically incorporation, or neglect, determined (path dependence). within even a same country. Within institutional framework, State building strategies differ

Arguments

Institutional invention is possible, and has been enacted in countries of Latin America. Individual agency matters (mostly when politicians have campaigned on a platform of societal performance improvement) and it needs to be understood within the prism of behavioral science, and not rational choice alone (Merilee Grindle) In Africa, local level configurations of power and interest matter and determine institutional outcomes (Boone) Regimes choose the institutional-building strategies that maximize their advantage in particular (local) political contexts. The result is differing institutional strategies of rural incorporation, or neglect, within even a same country. State building strategies differ because rulers face different challenges and opportunities (Boone, 2003). Thus, one of the critical ways to guarantee socially efficient institutional outcomes is through adequate political institutions which link the City to the countryside. As she contends, one of Africas top development priorities is to reform and strengthen the political institutions that link city and countryside. Local demands and political configurations will have to play a significant role in determining institutional change, substantive democracy in Africa will have to be bottom driven (Ake).

Institutions (rules of the game) constrain the behavior of organizations (players in the game), both politically and economically. In turn, institutions are humanly devised. This apparent tension between agency and determinism is resolved with the theory of path dependence. If the institutional setting and the state are bad for the economy and for societal performance at large, why don't the individuals in society just change the institutions for the better? The answer to this problem from North is that institutional change is path dependent. Path dependence is based on the idea that it is problematic and "costly" to change paths. Path dependence explains why economies get stuck in an institutional framework that is not efficient or does not induce growth and at large improved societal performance. Institutional Change is overwhelmingly incremental and Slow, as institutional outcomes are in large part historically determined (path dependence). Within institutional framework, organizations are afforded opportunities to enact incremental changes to the framework if they perceive higher payouts from doing so. Institutions modulate new opportunities that players seize to shape institutions Thus, Institutions are not necessarily or even usually created to be socially efficient; rather they, or at least the formal rules, are created to serve the interests of those with the bargaining power to devise new rules. In a zero- transaction-cost world, bargaining strength does not affect the efficiency of outcomes, but in a world of positive transaction costs it does (North, 1990) If economies realize the gains of trade by creating relatively efficient institutions, it is because under certain circumstances (very rare) the private objective of those with the bargaining strength to alter institutions produce institutional solutions that turn out to be or evolve into socially efficient ones Overly historically deterministic. Though Norths theory does leave agency to organizations, who can enact incremental changes within the institutional framework in so far as it maximizes their payoffs, it is constrained within larger framework of path dependence.

Limitations Overly case-specific, with limited external validity: How replicable are identified case-specific causalities to other cases?

36

In the next sections we go in-depth into the theoretical approaches of each sub-set of the New institutionalism body of literature, and the various factors they hypothesize constitute sources of institutional change, notably in Africa. We will then utilize these hypotheses in our analysis of sources of institutional change to explain institutional adaptation to meet the new CC challenge in African countries. 2.2 Definitions & relevance of Institutional theory Let us begin with a few definitions however to clarify our understanding of the institutional change literature. Rawls in his Theory of Justice defines a political institution as a common understanding about the rules of the game that define a political system acceptance of these rules entails that all parties subject themselves to these rules each knows that the other knows the rules all will abide by the rules because they know the others know the rules and will make him/her abide by them he/she was to stray away from them47. The latter part of this definition is particularly important. Douglas North, key pioneer of the body of literature on institutions, echoes Rawls definition in many regards, but taking an economic property rights approach to institutions. North posits that institutions indeed define the rules of the game, both politically and economically; they are the humanly devised constraints that shape human interaction48. These constraints can be both formal (rules, laws) and informal (behavior codes, cultural norms), and serve to reduce uncertainty and facilitate exchange in the presence of transaction costs. Transaction costs are in North's own words: "the costs of defining, protecting, and enforcing property rights"49. These transaction costs and institutions are the key to explaining economic growth or the lack thereof, and why growth has been so absent, rather than present, in the economies of the world, argues North. 47 Rawls, John, Theory of Justice (Cambridge: Belknap Press of Harvard University Press, 1971), 47. 48 North, Institutions, Institutional Change and Economic Performance, 3
49 Ibidem, 28

37

Applied to the issue of Climate Change, this definition of institutions makes us envisage the presence or absence of an institutional framework to address CCA as an explanation for societies effectiveness, or lack thereof, in governing CC and its critical impacts on vulnerable segments of society. This lack of common rules of the game to achieve the common purpose of addressing CC in this context defines societies ability or inability to effectively address the new risks and opportunities raised by CC. Finally, North distinguishes "institutions", which are the rules of the game, from "organizations", which are the actors, or players in the game. Hence, according to North, "it is the interaction between institutions and organizations that shapes the institutional evolution of an economy"50, and the achievement of societal goals at large one could add. Institutional change takes place in modifications of and changes in contracting, eventually leading to new "rules". This all happens "because individuals perceive that they could do better by reconstructing exchanges (political or economic)". Thus, institutions are the outcomes of the actions of organizations (players), the ones holding more bargaining power eventually establishing institutions (political and economic arrangements or rules of the game) that advance the interests of their creators51. In this sense, North asserts, institutions are not established to be socially efficient but rather to further the interests of those with more bargaining power52; and since institutions are path dependent (once established, they are very costly to change), they remain and perpetuate themselves indefinitely. This idea of path dependence is critical to the institutional change literature focused on structural determinants of institutional change, as we shall see in the next section devoted to the structural explanations for institutional change.
51 Ibidem, 73-82 52 Ibidem, 92-106 50 North, Institutions, Institutional Change and Economic Performance, 6

38

Many authors have elaborated on Norths property-rights, new institutional economics approach to grow and give it more depth (Mancur Olson53, North and Weingast54), corroborated by country examples (Acemoglu et al.)55 and analyzed its veracity when it comes to explaining institutional change (Barbara Geddes and Jeffrey Herbst /Pierre Englebert vs. Merilee Grindle). We shall turn to these authors to gain a better understanding of the role of institutions in generating, or hindering, societal performance. Why are institutions important? What evidence is there that they are critical? North posits that first and foremost institutions are important because they secure private property rights, reduce uncertainty and give individuals incentives to invest/produce in the presence of transaction costs (costs of enforcing property rights)56. North in this sense brings a needed complement to neo-classical theory by introducing transaction costs; rare if not inexistent are the cases in which the conditions necessary for the invisible hand obtain57. This is because transaction costs exist and property rights are not self-enforcing and costless. Indeed, since Adam Smith, economists have constructed their models on the firm bedrock of the gains from trade. Specialization and division of labor are the key to the wealth of Nations. In constructing their models, however, economists have ignored the costs arising from such specialization and division of labor (i.e: transaction costs). These transaction costs underlie the institutions determining the structure of political-economic systems. What the neo-classical economic model assumes is that perfectly specified and costlessly enforced property rights will give incentives for individuals to capture the returns to society of investment at all the
53

Olson, Dictatorship, Democracy, and Development, 567-76

54 North, Douglass C., and Barry R. Weingast, "Constitutions and Commitment: The Evolution of Institutions

55

Governing Public Choice in Seventeenth-Century England", Journal of Economic History 49 4 (1989): 803-832

Acemoglu, Daron, Simon Johnson, and James Robinson, The Colonial Origins of Comparative Development: An Empirical Investigation, American Economic Review 91 (2001): 1369-1401 56 North, Institutions, Institutional Change and Economic Performance, 6 57 Ibidem, 7

39

margins. However, such conditions have never obtained throughout history; many resources are closer to common property than exclusively owned. As a result the necessary conditions for achieving the equi-marginal efficient solution have never existedneither in the Roman Republic nor in the 20th century US or Soviet Union58. First, growth has been more exceptional than stagnation or decline, which suggests that efficient property rights are unusual in history. Secondly, transaction costs do exist; and a theory of institutions is required to fill the gaps in the neoclassical economic model. In this seminal work, Structure and Change in Economic History (1981), North proposes: a. A theory of property rights that describes the individual and group incentives in the system; b. A theory of the state since it is the state that specifies and enforces property rights; c. A theory of ideology that explains how different perceptions of reality affect the reaction of individuals to the changing of objective situations59. The example of 17th century Britain he gives in his 1989 piece with Wiengast60 is illustrative of Norths theory and drives home what he understands by property rights are not self-enforcing and costless. Indeed, in 1668 in Great Britain, wealth owners in Parliament, exceedingly losing patience with the Crowns encroaching on their wealth and private property and its arbitrary expropriation of their wealth whenever it needed added revenue (reign of the Stuarts under King James II at the time), led a revolution to change the power bargain between Parliament and the 58 North, Institutions, Institutional Change and Economic Performance, 8-9 59 North, Douglass C., Structure and Change in Economic History (New York: Norton, 1981) 60 North and Weingast, "Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England", 803-832 40

Crown of Great Britain. The result of this uprising, which came to be known as the Glorious Revolution, were new politico-economic arrangements (institutions) that ensured that: 1) The Crown could no longer expropriate wealth from private citizens; 2) Whenever it needed revenue the Crown would have to ask permission to Parliament and obtain clear, transparent and predictable tax revenue; 3) Politically independent courts were established to uphold the sanctity of private property rights and the rule of common law. Thus both Crown and Parliament committed themselves to set these institutions in stone, and never again change them. The result according to North and Weingast were phenomenal increases in the level of output and growth. Indeed by restoring confidence in the ability to invest without the risk of an arbitrary autocrat coming in to seize ones produce/output, Englishmen had an incentive to invest and to save their monies without fear that their pounds would be expropriated. This in turn led to the emergence of private debt markets to collect savings and redistribute them throughout the economy as loans, which led to money creation and overall to thriving of private markets and financial flows. The authors conclude that the departure in the historic record of British growth levels that resulted from the Glorious Revolution was even more significant than the growth hikes that occurred beginning in 1750 implying that more secure property rights and safeguards from arbitrary expropriation could have perhaps given business owners incentives to invest in new technology, and thus led to the Industrial Revolution. That is a theory that is still unconfirmed, but what North/Wiengasts study demonstrates is that good institutions that protect the private property of citizens have the powerful ability to incentivize investment and lead to growth. Similarly, applying this argument to the Climate Change issue, good institutions that clarify the rules of the game and roles of various organizations to address CCA, and sanctify these through enforcement by designated entities such as courts and legally mandated organizations to take leadership, potentially have 41

the powerful ability to incentivize the effective management of CCA and promote additional investments on the issue. Mancur Olson in his 1993 article Dictatorship, Democracy, Development, and to a certain extent in his early work on The Rise and Decline of Nations further elaborates on this point, giving a luminary game-theoretical illustration of how in practice the protection of property rights and clear definition of rules of the game does incentivize production. Olson, one of the pioneers of institutional economics, leads us to ponder upon the differences in productions levels in the presence of a roving bandit (such as the many bandit groups that roamed through ancient China in search of wealth to plunder) and in the presence of stationary bandit (a bandit who settles down, puts on a crown, maintains a monopoly over violence in his new domain and gets revenue by taxing his citizens). Olson maintains that in the presence of a stationary bandit, citizens have an incentive to produce because they know the bandit will protect them from the theft and violence of other bandits, and will only expropriate from them a fixed amount of tax revenue. In turn, in order to maximize his tax revenue, the stationary bandit has an incentive to invest in basic public goods for his citizenry to ensure that they produce as much as possible so he can derive even more tax revenue. This theory holds, and explains why there were no major rebellions and so much production has occurred under stationary bandits, from the early pharaohs up until Louis XIV before the 1789 French Revolution61. Olson states however that autocracy/stationary bandits/dictatorships are only second best options however; the reason being that the autocrat also has a rival incentive: that of extracting as much revenue as possible in the short time period he is guaranteed to be in power (short time horizon). When the time horizon of the autocrat is short, he will coerce his citizens, expropriating from them ever- unpredictable levels of tax revenue and giving them disincentives for investment and production. Although the autocrat does not maximize his revenue through this 61 Olson, Dictatorship, Democracy, and Development, American Political Science Review 87 (1993): 567-76 42

arrangement/institution, it satisfies his second short-term rationality and as a result growth is hampered. For this reason Olson concludes that DEMOCRACIES are the single best political arrangements because by ensuring that no one group becomes an autocrat (by definition what a democracy does), they guarantee less capricious and agreed-upon levels of taxation/state expropriation and provide incentives for investment, production and growth. Thus Democracy has not only moral appeals, but also under appreciated economic appeals concludes Olson62. This functional appeal of democracy is also an important aspect to keep in mind when deliberating on the effective governance of CC. What particular implication does this have? This assertion that good institutions matter for growth holds tremendous implications. Though originally written through the lens of economic growth, it applies to general questions of societal performance across all sectors, and begs the larger question: How then does institutional change occur to ensure that good institutions are implemented to favor growth and overall good societal performance? Many authors have addressed this question, and their explanations, relevant to our understanding of the circumstances under which institutional change in the management of CC obtains, can be characterized as STRUCTURAL approaches to institutional change and AGENT-BASED approaches to institutional change (refer to table 2.1). The first group of structural thinkers builds on Norths idea of path dependence. They assert that institutions are the fruit of past arrangements (wrought out by 62 Olson, Dictatorship, Democracy, and Development, 574-5 43

whomever had most bargaining power) and once instated, these institutions are costly and difficult to change. The second group places emphasis on agent-based processes of institutional change. 2.3 Sources of Institutional Change 2.3.1 The Structural or Path Dependent Explanations of Institutional Change The father of new institutionalism, and its most prolific advocate, is by far Douglas North. For North: History matters. It matters not just because we can learn from the past, but because the present and the future are connected to the past by the continuity of a societys institutions. Todays and tomorrows choices are shaped by the past. And the past can only be made intelligible as a story of institutional evolution.63 Through this phrase that opens his seminal 1990 work Institutions, Institutional Change and Economic Performance, North recognizes the profound effect previous institutions have on the ones of today. At times North comes close to providing a deterministic account on human behavior, where human behavior is constrained by institutions, yet at the same time these institutions are "humanly devised"64. This apparent tension is solved within and leads to the theory of path dependence. An undeniable merit in North's theory in relation to neo-classical economics is that he has come to point to the need for a theory of the state and bargaining power. Indeed, he was one of the first to forcibly point that economic performance, and societal performance in general, depends on the workings of the state, a novel statement in the early 1990s. Through this statement, he has built into neo-classical theory the realization that bargaining power, ideology and the state are critical to 63 North, Institutions, Institutional Change and Economic Performance, vii 64 Ibidem, viii

44

economic performance, and to societal performance more generally. He argues that: "Institutions are not necessarily or even usually created to be socially efficient; rather they are created to serve the interests of those with the bargaining power to create new rules"65. As such, the absence of economic growth is no longer a mystery, because: "It is polity that defines and enforces property rights and in consequence it is not surprising that efficient markets are so exceptional"66. Those with political power might not choose good institutions because they do not necessarily maximize their revenues. But, if the institutional setting and the state are bad for the economy, why don't the individuals in society just change the institutions for the better? The answer to this problem from North is that institutional change is path dependent. Path dependence is based on the idea that it is problematic and "costly" to change paths. This is what North has in mind when he starts his institutional treatise with the credo: "history matters". Therein lies the crux of this entire literature of Institutions. Bad institutions (politico-economic arrangements in society) lead to poor societal performance. But because they are so costly and path-dependent, bad institutions survive and continuously hamper growth, and societal performance. In North's theoretical framework, path dependence explains why economies get stuck in an institutional framework that is not efficient or does not induce economic growth. It is the story of how past events and culture influences the present. In North's own words:
Institutions determine the opportunities in a society. Organizations are created to take advantage of those opportunities... The resultant path of institutional change is shaped by: (1) the lock-in that comes from the symbiotic relationship between institutions and the organizations that have evolved as a consequence of the incentive structure provided by those

65 North, Institutions, Institutional Change and Economic Performance, 7 66 North, Structure and Change in Economic History,

45

institutions and (2) the feedback process by which human beings perceive and react to changes in the opportunity set.67

Thus the only sources of institutional change in Norths theory of institutions are as follows: 1) Organizations (players) seizing opportunities for change within the existing historically determined institutional framework to enact small changes at the margins, incrementally. The actors of this change are the organizations (see definition above). In their pursuit of wealth and utility maximization, organizations incrementally alter the institutional structure. The firm explores the most efficient governance structure and organization within existing institutional constraints. Such maximizing activity by the firm results from learning by doing and investing in the kinds of skills and knowledge that will pay off (within the existing institutional framework). But an alternative is to devote resources to changing the institutional constraints. Which direction the firm or economic organization takes depends upon its subjective perception of the payoffs.68 Thus for North the interplay between self-interested organizations seeking to maximize the gains and objectives of their creators & institutional constraints (both formal and informal) that set boundaries and condition the actions of players/organizations constitute the factors that shape the potential wealth- maximizing opportunities of entrepreneurs and determine the outcome of Institutional change.69 67 North, Institutions, Institutional Change and Economic Performance, 7 68 Ibidem, 140 69 Ibidem

46

Again, institutional outcomes that result are not necessarily efficient, but rather reflect the interests of the organizations with most bargaining power. Institutional change is the result of individual entrepreneurs responding to incentives for self- interest maximization, within the existing system. This change though according to North remains overwhelmingly incremental. Though Norths theory is cast under an analysis of economic performance, it can be generalized to explain societal performance, or the lack thereof, as a result of static inefficient institutions that were created by the most influential of the day, with no particular regard to efficiency and coherence, but rather on the basis of self-interest. 2) North mentions also the possibility for institutional change to occur abruptly as a result of accidents of history, such as war, revolution or natural disaster: War, revolution, conquest and natural disasters are sources of discontinuous institutional change.70 However these, in Norths theory, remain marginal cases. Institutional change remains overwhelmingly an incremental and a continuous process. Finally, North acknowledges that institutional change is a complex story however: in part due to the persistence of informal constraints in many settings (Africa being a prime example) where informal constraints at sometimes at antipodes of formal constraints. This renders it difficult to predict the constraints-modulated opportunities organizations will be seizing and the outcomes of their interventions. Outside of North, but still linked to the idea of path dependence, other sources of institutional change are identified. These explanations delve further into the historical accidents or conjunctures that also determine historical outcomes.

70 North, Institutions, Institutional Change and Economic Performance, 103

47

Engerman and Sokoloff first look at this idea in their Factor Endowment, Inequality and Paths of Growth in New World economies in which they assert that institutions result from the factor endowments/geography of countries. Looking at New World economies, they observe that colonies in which colonizers brought in slaves to work on labor-intensive sugar plantations led to the establishment of unequal institutions upon independence (Caribbean/Central America) whereas in the colonies of North America (US, Canada) where no slaves were brought in, society was more equal, leading to the widespread establishment of egalitarian institutions that protected private property for all citizens. Thus when industrialization took off in 1750 the latter were able to gain from the opportunities from trade71. On another continent, Jeffrey Herbst confirms these structural causes for the difficulty of institutional changes. Looking at Africa, Herbst poses the following question: if institutions are so detrimental for growth, why do people simply not change them? His response is that because of the particular political geography of Africa, leaders are made to govern an unintegrated peasantry that lives in the hinterland difficult to reach because of the lack of access and roads. As a result politicians have a strong incentives to cater only to urban interests and pressure, which in turn dissuades them from making any needed investments in the agricultural sector of the hinterland where the majority of the population lives72. The result again is that economic growth is hindered. Pierre Englebert further corroborates this assertion by stating that because leaders in Africa are illegitimate (in the sense that they do not derive vertical legitimacy from the broad pool of their citizenry to whom they are not accountable) they engage in patterns of behavior that are politically rational but economically irrational73. Englebert finds that in

71 Engerman, and Sokoloff, "Factor Endowments, Inequality, and Paths of Development among New World

Economies", 41-101

72 Herbst, J. I., States and power in Africa: comparative lessons in authority and control (Princeton: Princeton

University Press, 2000)

73 Englebert, Pierre, State Legitimacy and Development in Africa (Boulder: Lynne Rienner Publishers, 2000)

48

countries where leaders are found to be legitimate, economic growth is higher by 2% than in their non-legitimate counterparts74. Acemoglu, Robinson and Johnson in their Colonial Origins of Contemporary Development agree with Engerman and Sokoloff. They look at a similar source of institutions: colonial settlements as determined by white settler mortality. They conclude that wherever in their African, Asian and American colonies colonialists were able to settle thanks to low disease mortality, they established neo-Europes and property rights institutions that upheld the sanctity of private property; whereas where they could not settle due to high disease mortality, they established extractive institutions that coerced local citizens and led to disincentives for growth for example, in King Leopolds Congo. These extractive institutions in turn survived after Independence, leading to the present state of nill growth in many parts of the world75. Confirming Norths theory of path dependence, Acemoglu and Johnson in Political losers further assert that once institutions in place, institutional change that favors technological innovation will be blocked by politicians who fear they may lose power from the new institutional arrangements (the thrust of arguments about the political economy of reform and the difficulty of insulating economic decision- making from political pressures)76. Barbara Geddes, in her seminal Politicians Dilemma, confirms Norths political economy / rational choice analysis of institutional change by positing, in her study of Brazil, that politicians have two clear constituencies: the broad public their narrow political base.
74 Englebert, State Legitimacy and Development in Africa

75

Acemoglu, Daron, Simon Johnson, and James Robinson, The Colonial Origins of Comparative Development: An Empirical Investigation, American Economic Review 91 (2001): 1369-1401 76 Acemoglu, Daron and James Robinson, Political Losers As a Barrier to Economic Development AEA Papers and Proceedings 90 (2000): 126-130

49

Satisfying one comes at the detriment of the other. Often however pressures on them force politicians to satisfy their political base (eg: by appointing someone not necessarily the best for a position), which hampers economic growth. Thus economic growth and change are difficult in these circumstances77. Cases from across Africa documenting prebendal or neopatrimonial political systems confirm this situation, whereby leaders are faced with both a broad public and a narrower, ethnical, clan or otherwise-based constituency to which they are accountable (see Ekeh 197578 and Mamdani 199679). These unaligned bases of accountability, and one could surmise of institutions, have hampered growth as well as societal performance (see Ake 1996)80. Focusing on a study of the efficiency of markets in Africa, Bates confirms the dichotomy between State agents public interests and private gains as the main driver of economic stagnation in Africa. A continuation of colonial state structures (path dependence), rural development agencies in much of sub-Saharan Africa can be understood as institutions of rural political control and taxation according to Bates81. Thus all of the above reviewed authors from Engerman and Sokoloff, to Geddes, Herbst, Englebert and Bates) confirm North assertion that History matters because the present and the future are linked to the past through the continuity of societys institutions82. The main sources of institutional change for these structural authors are thus: Incremental changes, which are enacted by self-interested rational players seizing opportunities afforded to them within the existing historically determined institutional framework to increase their private gains (North and
77 Geddes, Barbara, Politician's Dilemma: Building State Capacity in Latin America (Berkeley: University of

California Press, 1994)

78 Ekeh, P., Colonialism and the two publics in Africa, Comparative Studies in Society and History 17 (1975): 91-

112
79 Mamdani, M., Citizen and subject: contemporary Africa and the legacy of late colonialism (Princeton: Princeton

University Press, 1996)

80 Ake, C., Democracy and development in Africa (Washington, D.C: Brookings Institute, 1996) 81 Bates, Robert, Markets and States in Tropical Africa: the political basis of agricultural policies (Berkeley:

82

University of California Press, 1981)

North, Institutions, Institutional Change and Economic Performance, vii

50

Geddes); these incremental changes are only socially efficient however in the rare circumstances that private gains align with societal ones; Conjuncture or accidents of history, whether endogenous such as factor endowments (Engerman and Sokoloff), low population density (Herbst), areas of low disease mortality (Acemoglu et al.) or exogenous (colonial institutions; Bates and Geddes). 2.3.2 The agent-based explanations of institutional change Another cluster of institutional theorists purport a competing theory of institutional change than the one driven by North and his cohort. Based principally on agent- based arguments, this group of scholar finds that institutional change is a result of: Individual agency and desire for societal improvement (not rational choice, behavioral science) Institutional outcomes not the fruit of historically determined process, but outcome of intense bargaining between national leaders and local elites (Boone) Merilee Grindle, chiefly, disagrees with Norths theory of path dependence. She agrees fully that institutions are important to generate better economic outcomes however in her magisterial Audacious Reforms: Institutional Reforms and Democracy in Latin America she proves that institutional invention is possible, and has been enacted in countries in Latin America (Venezuela which chartered its first local/state elections and allowed individual candidates in 1989; Bolivia which passed in 1993 its National Participation Law decentralizing the political system in effect and in Argentina where under the leadership of Raul Alfonsin and other intellectuals reforms began in 1994 to give Buenos Aires more political autonomy and introduce a balance of power in the political institutions). Thus, contradicting directly Geddes, Grindle proves that institutional change does happen but to understand how it happens one has to look beyond the prism of rational choice theory, and bring in other insights from behavioral science and sociology. As she 51

concludes Agency does matter and there is still space for agency in changing ones institutions and enacting change83. In the same vein, Catharine Boone does away with historical determinism. In her seminal Political Topographies of the African State in which she analyzes the institutional choices adopted by Africas newly independent states to build their nations, she concludes that in-country differences in centralization/decentralization of the state apparatus are not the product of exogenous factors e.g.: the import of the state apparatus or external funding agencies ideas for what constitutes good governance, but rather the product of internal variations in the configuration of political power and relationships within agrarian society. Institutional differences are determined endogenously: spatial variations in institutional design and the extent of core-periphery power-sharing are the products of political struggle and bargaining that goes on within African society between rulers, their rural allies and their provincial rivals. As a result, institution-building strategies aimed at incorporating rural societies into modern states have varied starkly not only across, but also within countries in Africa84. Boone continues on to condemn the over-emphasis on the predatory State. So compelling was the categorization of the State as a Leviathan, she writes, that little room was left for accounts describing marketing boards, official credit agencies, settling schemes and provincial administrations as politically neutral or benevolent initiatives of the state leading to striking generalizations on the exploitation and disempowerment of rural society85. In her account, agrarian society is not disempowered however, but the important scenes for internal battles among rural elites who also sought to control and wield power over the ordinary citizens, whom were their clients, followers, kinfolks and subjects. New regime politicians in the city

83 Grindle, Merillee S., Audacious reforms: Institutional invention and Democracy in Latin America (Baltimore:

Johns Hopkins University Press, 2000)

84 Boone, C., Political Topographies of the African State: Territorial Authority and Institutional Choice (Cambridge:

Cambridge University Press, 2003)

85 Boone, C., Political Topographies of the African State, 5-6

52

found these widely differing rural politicians in place who defined local political contexts, and designed their strategies for state-building according to the possibilities they had for collaborating with them or not. Though Boone shares with North a game-theoretical understanding of the interactions between self-interested rational organizations within a given institutional system, she does not agree that institutional outcomes are pre- conditioned by history. Rather she leaves full agency to the organizations within the system, and based on her findings for Africa, she conceives of institutional change as being the equilibrium outcome of fierce political struggles between different actors who all seek to wield as much control as possible, the result of which yields a mutual equilibrium in which all organizations find an increase in their payouts. This equilibrium again here does not indicate social efficiency, but rather is the wealth- maximizing optimum of the various actors / players who took part in the negotiations. New politicians in the city found themselves locked in negotiations and confrontation with rural elites over the distribution of power, political prerogative and authority, and rural wealth The intensity and nature of the rural political challenge to new African regimes varied by region, shaping and constraining the possibility for collaboration between regimes and rural notables. The result has been varying institutional strategies for rural integration.86 Boone thus calls our attention back to the importance of local level configurations of power and interest. In her words, these local configurations matter, and need to be taken into account when explaining institutional outcomes. Regimes choose the institutional-building strategies that maximize their advantage in particular (local) political contexts. The result is differing institutional strategies of rural incorporation, or neglect, within even a same country. State building strategies 86 Boone, C., Political Topographies of the African State, 35-36 53

differ because rulers face different challenges and opportunities87. Thus, one of the critical ways to guarantee socially efficient institutional outcomes is through adequate political institutions which link the City to the countryside. As she contends, one of Africas top development priorities is to reform and strengthen the political institutions that link city and countryside.88 In reforming and strengthening these political institutions, local demands and political configurations will have to play a significant role in determining institutional change, and in all likelihood, demand will have to be bottom-driven, rippling all the way up to the top of the State apparatus. This sentiment is echoed by Claude Ake in his Feasibility of Democracy in Africa in which he states that substantive democracy in Africa will have to be bottom-up89. Yet, it remains that institutional change is no easy task and it took a coalition in most cases shared from Latin America and Africa to enact change. However given an increasing consensus on the importance of institutions to growth and societal performance, the question of how institutional change occurs is urgent and will require greater inter-disciplinary work between political scientists, economists, drawing on fields as wide as rational choice theory but also behavioral sciences to understand determinants of institutional change. Overall, from this group of agent-based explanations for institutional change emanates that the main sources of institutional change are: An agents desire for improved societal performance (or campaigned on platform for this), bringing private interests in line with public ones. As a result, socially efficient outcome ensues.

87 Boone, C., Political Topographies of the African State, 35-36 88 Ibidem, 5-6 89 Ake, C.

54

 Local demands determine new political opportunities/constraints at the local level forcing national leaders to put local demands on the national agenda (Boone and Ake). The limitation of this agent-based model is however that it is overly case-specific, drawing on Latin America and African cases. Its external validity has to be tested, against the more conceptually broad idea of path dependence as a determinant of institutional change. 2.4 Towards a Theory of Institutional Adaptation to meet the CC Challenge How does our new understanding of the body of literature on Institutional change help us to better conceive of institutional change or adaptation to meet the new challenge of CC in our case countries across Africa? First and foremost, the institutional change theory above-reviewed offers us a rich theoretically-grounded set of potential independent variables (IVs) to use as input into our investigation of the enabling conditions of institutional innovation in Africa to meet the new challenge of Climate Change. As such, the theory can be easily applied to our analysis, in which we will test the hypotheses put forth in the theory (both agent-based and structural hypotheses), on the new case of climate institutional change. Applying institutional change theory to analysis of climate change institutions, we will thus be asking: how have countries adapted institutions to meet the new challenge of CC? Such an analysis will enable us to: 1) Gain better understanding of the determinants of institutional change in light of the new Climate challenge: what prompts governments across Africa to adapt institutions or devise new ones to meet the new CC challenge? 2) Conduct hypothesis testing of Norths theory of institutional change on a new issue, that of climate institutional change, based on empirical evidence from 55

across Africa. This will enable us to verify whether it holds that institutional change is slow and incremental, and mainly only takes hold as the fruit of sustained incremental efforts by wealth-maximizing organizations; or on the contrary, whether the agent-based explanations of institutional change are better fitted to explain CC institutional adaptation.

56

 What prompts a country to engage in institutional change/adaptation to meet the new CC challenge? I. Structural hypotheses of institutional change a. Institutional change is slow and incremental: Self-interested rational organizations bring about incremental change in historically- determined institutional framework to increase their private gains (North and Geddes); under rare circumstances, the private interests of rational organizations (seeking to maximize their payouts and further the objectives of their creators) align with socially beneficial goals, in these rare instances, socially efficient institutions (new rules) obtain b. Conjuncture or accidents of history i. National Disaster, War or Revolution (North) ii. Factor endowments/geography (Engerman/Sokoloff) iii. Colonial legacy (Bates) iv. Population density (Herbst) II. Agent-based hypotheses of institutional change: c. An agents desire for improved societal performance (or campaign on platform for this) brings private interests in line with public ones => socially efficient outcome ensue (Grindle) d. New political opportunities/constraints at the local level force national leaders to put local demands on the national agenda (Boone) e. Strong endogenously-developed institutions (coherently integrated with local institutions and downwardly accountable)
Box 2.1: List of theoretical hypotheses to be tested on the new CC case

Part III will be devoted to explaining the methods we utilize to test the hypotheses of the institutional change literature on the new CC case. Before we proceed to explain our methodology however, it is important to revisit the issue of CC. Why does the new CC challenge require new or adapted institutional frameworks at the 57

national and regional levels to meet its imperatives? What is the triple bottom line of DRR-CCA-Development and why are strong institutions so essential to achieve it? We end this part with a clarification of the conceptual roots of Climate Change institutional adaptation, towards a theory of institutional change to meet the new challenge of CC. 58

CHAPTER 3: CONCEPTUAL FRAMEWORK: INSTITUTIONAL ADAPTATION FOR EFFECTIVE GOVERNANCE OF CCA IN AFRICA Why are new or adapted institutions required to effectively govern the new challenge of Climate Change Adaptation in Africa? With Climate Change, unprecedented reversals in human development are expected in communities most dependent on climate-sensitive ecosystem services for their livelihoods (UNDP 2007-08). On the ground, across Africa, communities and countries have already begun to contend with climate changes we are committed to, while bracing for those yet to come. Indeed, communities at the frontlines of a changing climates impacts across Africa already carry out spontaneous adaptations individually or collectively without institutional support or coordination from national or local governments (see Agarwal 200890; also UNDP 2007-08 and Benaouda et al. 2008 for examples in Morocco91). In this sense, Climate Change lays bare the key governance and institutional challenges of African countries. Climate Change adds an additional stressor to the multiple ones that already encroach on local livelihoods (e.g: market price fluctuations, weak social safety nets, etc.), which institutions, from local to national levels, have to contend with and address. Most of all, faced with the sheer magnitude of the challenges likely to be occasioned by Climate Change, countries across Africa, and institutions within them, will have only one of two options: to Adapt in order to effectively deliver on the CCA imperative, or to Perish or at least suffer severe loss.
90 Agrawal, A., and Nicolas Perrin, Climate Adaptation, Local Institutions, and Rural Livelihoods, IFPRI working 91

paper # W08I-6 (2008), Accessed April 04, 2012. http://sitemaker.umich.edu/ifri/files/w08i6_agrawal_perrin.pdf Benaouda H., A. El Ouali and A. Saloui, Rapport final du projet CRDI/INRA : Mcanismes dadaptation aux changements climatiques des communauts rurales dans deux cosystmes contrasts en plaine et montagne du Maroc (Settat: Institut National de la Recherche Agronomique, 2008)

59

 In order to understand the magnitude and cross-ranging impacts that Climate Change is likely to bring in Africa, and the imperative of institutional adaptation to meet these new challenges, we review the projected impacts of Climate Change in Africa. How is Africa likely to be impacted by changing climate? What does most current scientific knowledge project? What are limitations in our understanding of how Africa will be specifically impacted by long-term Climate Change? We investigate these questions in the next sections. 3.1 Climate Change, The Scientific Basis: Causes, Concepts, Implications 3.1.1 What is Climate Change (CC)? Before we proceed to expose the projected impacts of Climate Change for Africa, we must first explain what is Climate Change- shorthanded as CC from on onwards. For the purposes of this research, we follow the United Nations Framework Convention on Climate Change (UNFCCC) definition of Climate Change (CC) as a change of climate which is attributed directly or indirectly to human activity that alters the composition of the global atmosphere and which is in addition to natural climate variability observed over comparable time periods92. This definition focuses on man-made, or anthropogenic interferences with the climate system, which may lead to dangerous disturbances of the global climate systems equilibrium. 3.1.2 Change vs. Variability Climate Change (anthropogenic) is different from (natural) Climate variability. On one level, the two concepts can be differentiated on the grounds that one is anthropogenic whereas the other is naturally driven. On another level however, the consequences of CC and those of Climate variability manifest themselves on 92 United Nations, United Nations Framework Convention on Climate Change Fccc/Informal/84 Ge.05-62220 (E)
200705 (1992), accessed December 1, 2011. http://unfccc.int/resource/docs/convkp/conveng.pdf

60

different timescales. Indeed, Climate Change refers to the long-term changes in atmospheric conditions that occur at timescales of decades, centuries and millennia, whereas climate variability refers to shorter-term fluctuations that occur at the timescale of a season, months or a few years (see fig. 1). Fluctuations in atmospheric conditions are only ascertained to be the result of Climate Change (CC) when the latter occur consistently over long enough periods of time, as evidenced by data demonstrating a statistically significant departure from the mean value of a climate variable e.g.: a change in average rainfall amounts over a 30-year period. Only then can we speak of Climate Change or changes in long-term climate patterns and trends. All else generally only represents a manifestation of climate variability or natural weather changes.
Fig. 3.1: Atmospheric change, a question of timescales. Source: IRI

3.1.3 What causes Climate Change? Climate Change is caused by increased concentrations of Greenhouse Gases (GHG) in the atmosphere resulting from human activities. The main GHG responsible for CC is carbon dioxide. Methane, nitrous oxide and sulphur hexafluoride are other significant contributors. Sources of carbon dioxide include exhaust fumes from industrial production and car use, while methane emanates from cows and intensive cattle breeding, and nitrous oxide from agricultural land/animal manure 61

management. Altogether combined, these gases are responsible for the GHG- induced changes to our atmosphere, through a process called the Greenhouse effect. Figure 3.2 illustrates schematically how the Greenhouse effect operates. Increased atmospheric concentrations of carbon dioxide mainly, as well as other GHGs, form a canopy of GHGs that traps outgoing radiation, and thwarts planet Blue from emitting thermal infared radiation back out to space. Instead, this heat is deflected back to Earth. As a result, the Earths energy balance is set off equilibrium; the amount of incoming energy (solar radiation) no longer equates to the amount of energy emitted back out to the atmosphere (back radiation). This excess heat is returned to Earth, making Earths global average temperature rise a phenomenon better known as Global Warming.
Fig. 3.2: The Greenhouse Effect: Greenhouse Gases trapping outgoing terrestrial radiation (red arrow), the cause of global Climate Change. Sources: Pearson Prentice Hall, 2004; IPCC, 2001.

Between the decade of the 1870s (corresponding to the beginning of the Industrial revolution) and the early 2000s alone, Earths average temperature has risen by 0.7 62

93. The rise of Earths global average temperature has multiple consequences, the

most prominent of which being the melting of polar ice caps, which in turn cause the rise in sea level a phenomenon better known as Sea Level Rise, as the condensed polar ice sheets sitting on land thaw to liquid water and flow to the ocean, raising seawater levels worldwide. It is very unlikely that the 20th-century warming can be explained by natural causes alone94. Many scientists today argue that to maintain life on Earth as we know it and achieve significant curtailing of global warming, carbon concentrations in the atmosphere should be maintained at 350 ppm, with a cutback on current atmospheric CO2 concentrations levels that stand at 380 ppm (see Hansen et al. 200895; please note that IPCC emission reduction targets are more conservative). However the point remains that to avoid dangerous human interferences with the climate systems equilibrium, GHG emissions need to be curbed.

Table 3.1: Dangerous Climate Impacts and the Kyoto Protocol. Note: Sea-level rise from the disintegration of the West Antartic Ice sheet at 550ppm (a 2C increase in global average temperature), would be devastating to the worlds coastal cities and population centers. The shutdown of the Gulf Stream circulation and the loss of its warming influence could lead to a sudden ice age in Northern Europe. Source: ONeill and M. Oppenheimer, 200296.

A subsidiary consequence of Global Warming not related to ice cap melting, is changes in global weather patterns and climate trends. As the Earth has an energy
94 IPCC, Fourth Assessment Report 95 Hansen, J., et al., Target atmospheric CO2: Where should humanity aim? (2008) 96 ONeill, B.C, and M. Oppenheimer, Dangerous Climate Impacts and the Kyoto Protocol, Science 296 (2002):

93 IPCC, Climate Change 2001: Third Assessment Report

1972

63

surplus and strives for new equilibrium to balance it energy equation, this drives changes in wind patterns, ocean-land teleconnections and hosts of other atmospheric changes difficult to predict, resulting in more erratic weather and climate patterns. The projected consequences of GHG-induced CC can thus be summarized as follows: A rise in mean global temperature: Global warming (very confident) Melting Polar Ice Caps & Sea Level Rise (SLR) (very confident) Shifting global climate/weather patterns as Earths climate system adapts to

energy imbalance and searches for a new equilibrium (exact parameters and consequences unknown) 3.1.4 Is Climate Change New? No, the Earths climate system has always been in constant motion. Using various paleo-climate techniques to reconstruct past millennia climate data (essentially based on readings of ice cores, paleo-fire/ocean, lakes, boreholes, caves, tree rings, plant macros, insect pollen and historical data), scientists were able to obtain proxy climate data with which to compare more recent data emanating from modern measurement instruments (thermometers, precipitation gauges, etc.). Their findings revealed that natural climate changes have always existed, and were driven by phenomena as varied as changes in the Earths orbit, leading to fluctuations in incoming solar radiation, volcanic eruptions or asteroid impacts. What is new however is the magnitude and rapidity of change observed in contemporary times. Indeed, current climate changes dwarf past millennia of changes. Fig. 3.3 hints to how climate changes experienced in more recent times on the geological time occur at an order of magnitude and rate of change that is different from anything experienced in the past 2.5 million years.

64


Fig. 3.3: Current climate changes dwarf past millennia of climate changes (read graph from right to left). Source: Pearson Prentice Hall, 2004.

The IPCC Fourth Assessment Report further corroborates the significance of the observed increase in rates of change. Global atmospheric concentrations of CO2, methane (CH4) and nitrous oxide (N2O) have increased markedly as a result of human activities since 1750 and now far exceed pre-industrial values determined from ice cores spanning many thousands of years. Atmospheric concentrations of CO2 (379ppm) and CH4 (1774ppb) in 2005 exceed by far the natural range over the last 650,000 years97.

3.1.5 Evidence of Climate Change Empirical evidence supports the assertion that increased GHG have made their way to the atmosphere, with detrimental impacts on Earths climate system (see figures 3.4-6).

97 IPCC, Fourth Assessment Report

65


Fig. 3.4: The Keeling Curve : Atmospheric CO2 measured at Mauna Loa, Hawaii monthly since 1960. Source: NOAA Earth System Research Laboratory, January 2012

Fig. 3.5: Abnormal increases in Northern Hemisphere surface temperatures since the mid-1970s. Source : IPCC, 2001

Fig. 3.6: Surface Melt on Greenland. 70m thinning of the Greenland melt area was observed between 1992 and 2002 ; the satellite-era record melt of 2002 was exceeded in 2005. Sources: Roger Braithwaite, University of Manchester (UK) and Waleed Abdalati, Goddard Space Flight Center.

66

3.1.6 Implications for Africa: Adapting to the Inevitable A final prominent feature of the Climate Change crisis is that it is irreversible. Indeed, due to past century atmospheric carbon pile-ups we are already committed to, further changes in climate are unavoidable. Nations must prepare for them. Indeed, the combined force of inertia and cumulative outcomes of climate change are important features of the CC problem. Major parts of the climate system display time lags between action and reaction, which makes them respond slowly to changes in GHG concentrations in the atmosphere. This inertia built into the climate system, and mainly inherent in the globes oceans that have a long memory means that the climate system will still continue to change as it adapts to the increased emissions from past decades. This means that even if all nations were to agree to instantly halt carbon emissions (i.e.: perfectly successful MITIGATION), we would still be experiencing changes in the climate system caused by the carbon emissions deposited into the atmosphere during the previous century, to which we are already committed to and have to adapt, regardless of what ensues from climate mitigation international negotiations. Therein lies the imperative of ADAPTATION. Climate Change thus ought to be mitigated, in order to thwart continued carbon emissions from reaching a tipping point of catastrophic climate change (the 350ppm target) beyond which the Earth reaches a disequilibrium so severe the consequences of which no-one is to predict especially if we were to get to the point where ice sheets over Antarctica and Greenland give way. However it cannot be stalled. As a consequence, for the first half of the 21st century, the world in general, and the worlds poor in particular, will have to live with Climate Change to which we are already committed. 67

 As eloquently stated by the UNDP Human Development Report 2007/08, without urgent mitigation action the world cannot avoid dangerous climate change. But even the most stringent mitigation will be insufficient to avoid major human development setbacks. The world is already committed to further warming because of the inertia built into climate systems and the delay between mitigation and outcome. For the first half of the 21st century, there is no alternative to adaptation to climate change98. As the IPCC further confirms: Anthropogenic warming and sea-level rise will continue for centuries due to the timescales associated with climate processes and feedbacks (notably thermal expansion and response of ice sheets to warming), even if greenhouse gas concentrations were to be stabilized. Thus both past and future anthropogenic carbon dioxide emissions will continue to contribute to warming and sea level rise for more than a millennium99. As such, there are double imperatives: one of Adaptation to the inevitable to Climate Change we are already committed to due to accumulated atmospheric carbon concentrations, and one of Mitigation to thwart catastrophic Climate Change by curbing further carbon buildups. As nicely summarized by the UNDP special report on Climate Change, at the international level, doing nothing offers a guaranteed route to a further build-up of greenhouse gases, and to mutually assured destruction of human development potential. This is still a preventable crisis though, the world still has a few more years before we reach the tipping point. But if serious carbon setback efforts are not engendered, by the end of the 21st century, the specter of catastrophic ecological impacts could have moved from the bounds of the possible to the probable100.
99
98 UNDP, Human Development Report 2007-08

IPCC, Fourth Assessment Report 68

100 UNDP, Human Development Report 2007-08

At the national level as well, doing nothing will condemn citizens to a fate of reduced opportunity and major human development setbacks; and scholars will be writing the chronicle of a predicted catastrophe. The imperative of Adaptation to Climate Change is most immediate for the worlds poorest regions where climate fluctuations will have profound impacts on local livelihoods; some contend that these impacts have already begun. Africa, one of the most vulnerable continents where impacts of CC are feared, is only left with two choices: to Adapt or to Perish or at least suffer very severe losses. 3.2 Projected Climate Change Impacts over Africa Large uncertainties remain regarding the exact impacts CC will have on the continent, in part due to the limited yet growing understanding of the drivers of African climate variability. In even larger part however, our understanding of CC impacts in Africa is crippled by our limited knowledge on how human and political systems will react to future stresses imposed by a changing climate. In the next sections, we lay out these uncertainties, as it is important to understand them, in order to appreciate what we know and what still remains unknown about what CC will bring for Africa. 3.2.1 Understanding Africas Climates Africas climates are both varied and varying101. Indeed, Africa, the second largest continent, has a wide range of climate regimes ranging from very humid equatorial regimes giving birth to the great equatorial rainforests of the continent, to the arid tropical climates of the African Sahel.102
102 Nkomo et al, 2006 101 Hulme M., R., Doherty, T. Ngara, and M. New, Global Warming and African Climate Change: a Reassessment,

In Climate Change and Africa, ed. Pak Sum Low, Cambridge: Cambridge University Press, 2005

69

 Africas climates are also characterized by high multi-decadal, inter-decadal and inter-annual variability, most notably in rainfall. They exhibit differing degrees of temporal variability, particularly with regard to rainfall103. A case in point illustrating African rainfall variability is the Sahel drought of the 1970s, which represents a pendulum swing in the evolution of this region of Africas climate. The Sahel drought was a sustained period of decreasing rainfall from the 1970s to the 1990s, followed by pickup in rainfall (see figure 3.7), which eloquently illustrates the high variability inherent in Africas climate systems. In West Africa, this 30-year decline in rainfall caused a 25-35 km southward shift of the Sahelian, Sudanese and Guinean ecological zones in the second half of the 20th century. This has resulted in a loss of forested area over grassland and acacia, the loss of flora/fauna, and shifting sand- dunes in the Sahel (see Gonzalez 2001, ECF and Potsdam Institute 2004 in IPCC 2007)104.

Fig. 3.7: Rainfall anomalies in the African Sahel from 1901 to 1997. Positive anomalies equate to above normal rainfall, while negative values refer to below normal rainfall, or periods of drought (from Nicholson et al., 2000)

103 Hulme et al., Global Warming and African Climate Change: a Reassessment 104 IPCC, Fourth Assessment Report, 443-44 70

3.2.2 Drivers of African climate changes: CC vs. Variability The climate of the continent is controlled by complex mechanisms105. The two most important drivers of African climate variability remain nonetheless: 1) Ocean forcing. El Nino Southern Oscillation (ENSO) events in the eastern Pacific Ocean remain the most important driver of inter-annual variability, with warm ENSO events correlating with high rainfall in equatorial eastern Africa (correlation strongest during the short rains September-November rainy season) and low rainfall in Southern Africa (correlation strongest during December-February) as well as in the Sahel (see Giannini et al. 2003106; Janicot et al 2001107), all the way across to Ethiopia (Korecha and Barnston, 2007108 ); 2) Land-atmosphere interactions (land cover changes and their feedbacks to the climate system). 3.2.2.1 Better understanding exists today of the complex mechanisms responsible for rainfall variability in Africa. Firstly are sea-surface temperature (SST) anomalies, both near (e.g.: SSTs in Gulf of guinea for Sahel) and far, through oceanic- atmospheric bridges and tele-connections. For instance, it is increasingly recognized today that the Sahel drought of the 1970s was associated with the warm SST trend at the time. Giannini et al. confirm this assertion based on the observed record of sea surface temperature from 1930 to 2000, they suggest that variability of rainfall in 105 IPCC, Fourth Assessment Report, 435 106 Giannini, A., R. Saravanan, and P. Chang, Oceanic Forcing of Sahel Rainfall on Interannual to Interdecadal
Time Scales, Science 302 7 (2003)
107 Janicot, S., S. Trzaska, and I. Poccard, Summer Sahel-ENSO teleconnection and decadal time scale SST

Rainfall changes

108

variations, Climate Dynamics 18 3-4 (2001): 303-32

Korecha, D. and A. G. Barnston, Predictability of June-September rainfall in Ethiopia, Mon. Wea. Rev. 135 (2007): 628-650 71

the Sahel results from the response of the African summer monsoon to oceanic forcing, amplified by land-atmosphere interaction. The recent drying trend in the semi-arid Sahel is attributed to warmer-than-average low-latitude waters around Africa, which, by favoring the establishment of deep convection over the ocean, weaken the continental convergence associated with the monsoon and engender widespread drought from Senegal to Ethiopia109. Secondly are ENSO processes, a significant influence on rainfall variability at inter- annual scales. ENSO has been associated in recent years with severe droughts in Southern Africa for example. Last are land use/land cover changes. Indeed, several studies have highlighted the importance of terrestrial vegetation cover and the associated dynamic feedbacks on the physical climate. A study by Bounoua et al. has demonstrated for instance that an increase in vegetation density has been suggested could result in a year-round cooling of 0.8C in the tropics, including tropical areas of Africa (see Bounoua et al. 2000 and Christenssen et al. 2007, in IPCC 2007)110. Also, complex feedback mechanisms between land cover changes/deforestation, human migration to the cities compounded by changes in atmospheric dust loadings played a role in the persistence of the Sahel drought and its surroundings. Complex interactions between land cover/land use changes, human agency and feedback loops onto the physical climate characterize land use influences on the climate variability; the complexity of these interactions precludes simple interpretations111. The largest source of uncertainty in the planets radiative balance remains however the influence of aerosols, a case in point being Saharan atmospheric dust loadings.

109 Giannini et al., Oceanic Forcing of Sahel Rainfall on Interannual to Interdecadal Time Scales 110 IPCC, Fourth Assessment Report, 457 111 Ibidem 72

Other mechanisms that appear to influence rainfall variability to a lesser degree are the localization and intensity of the African Easterly Jet (AEJ) and the Tropical Easterly Jet (TEJ)112, and the warmth of the Mediterranean Sea (giving abundant rainfall). The way these mechanisms influence regional weather patterns are being demonstrated through correlations here and there, but uncertainties remain large113. The attribution question remains nonetheless. How much of these changes in rainfall are naturally driven, and how much can be attributed to increased green house gases? These are the answers that climate scientists worldwide are currently attempting to elicit. Many researchers make the case that GHG-induced CC is the culprit for the many observed climate changes in Africa however. The first change that this group of researchers highlights is the increased climate variability observed across the continent in recent years. Researchers report on observations of increased extreme weather events and exacerbated climate variability. Increased inter-annual variability has, been observed in the post-1970 period, with higher rainfall anomalies and more intense and widespread droughts reported (for examples see Richard et al. 2001; Fauchereau et al., 2003114). In different parts of southern Africa (e.g.: Angola, Namibia, Mozambique, Malawi, Zambia), a significant increase in heavy rainfall events has also been observed (Usman and Reason, 2004115), including evidence for changes in seasonality and weather extremes (Tadross et al., 2005; New et al., 2006116). During recent decades,
113 114
112 The atmospheric circulation has to come to a different equilibrium, given the oceanic forcing: it does so by

shifting the strength or location of these features, consistently with moisture convergence.

IPCC, Fourth Assessment Report, 457 IPCC, Fourth Assessment Report 115 Ibidem 116 Ibidem

73

eastern Africa has been experiencing an intensifying dipole rainfall pattern on the decadal time-scale, characterized by increasing rainfall over the northern sector and declining amounts over the southern sector (see Schreck and Semazzi 2004 in IPCC 2007117). This observation is consistent with the recovery of the rainfall observed in the Sahel in recent years, also coined greening of the Sahel118. Secondly, on the climate change timescale, they point to the substantial inter-annual and multi-decadal rainfall variations observed since the second half of the 20th century, a case in point being the near continent-wide droughts in 1983 and 1984, with dramatic impacts on economies and environments. Oceanic SST-driven tele-connections are not enough anymore to explain changes in rainfall variability. It is clear that the oceans, especially the Pacific Ocean are important modulators of both inter-annual and inter-decadal climate variability in Africa. But the oceans alone do not hold all the clues... work examining the variability of African climate, mostly rainfall, has to be set in the wider context of emerging understanding of human influences on the larger global scale climate (see IPCC 2007; also Williams & Funk, 2011)119. However, the extent to which these rainfall variations are related to anthropogenic GHG-induced global warming remains undetermined120. Although many scientists are beginning to piece together the attribution question indirectly, through the relative influences of GHGs and sulfate aerosols on the oceans and patterns of SSTs relevant to Africa (see Ting et al. 2009121; also Chang, Chiang et al. 2011), the

117 IPCC, Fourth Assessment Report

118 Giannini, A., M. Biasutti and M. M. Verstraete, A Climate Model-Based Review of Drought In The Sahel:

Desertication, The Re-Greening And Climate Change, Global and Planetary Change 64 (2008): 119-12
119 Ting, Mingfang, Yochanan Kushnir, Richard Seager, And Cuihua Li, Forced and Internal Twentieth-Century

SST Trends in the North Atlantic, Journal of Climate (2009) 1469-81

120 Hulme et al., Global Warming and African Climate Change: a Reassessment 121 A. Park Williams & Chris Funk, A westward extension of the warm pool leads to a westward extension of the

Walker circulation, drying eastern Africa, Climate Dynamics (2011)

74

attribution question does not have an easy one-on-one answer, and uncertainties remain large. 3.2.2.2 Attributions for temperature changes are easier. There is a consensus on the finding that temperatures in Africa are on the rise. The continent of Africa is warmer than it was 100 years ago with an average rate of warming of 0.5C/century through the 20th century. The six warmest years in Africa have all occurred since 1987, with 1998 being the warmest year122. These changes are consistent with global trends. Africas rate of warming is not dissimilar to that experienced worldwide and the periods of most rapid warming the 1910s-1930s and the post 1970s occur simultaneously in Africa and the world123. Model based predictions of future GHG induced CC for the continent clearly suggest that this warming will continue, and in most scenarios, accelerate so that the continent on average could be between 2 and 6 C warmer in 100 years time. While these predictions of future warming may be robust, there is much less confidence about the direction and even magnitude of change when it comes to rainfall projections124. Will we witness desiccation and a return to the droughts of the 1980s, or should we expect more rainfall, as the current greening of the Sahel seems to suggest? Low confidence in current climate model projections precludes any definitive trend about the direction or magnitude of change in rainfall125. Processes such as ENSO, land cover-atmosphere interactions/changes and biomass aerosols impacts should first be accounted for and included in GCM predictions before rainfall changes can be predicted adequately126. 122 Hulme et al., Global Warming and African Climate Change: a Reassessment 123 Ibidem 124 Ibidem 125 IUNDP, Human Development Report 2007-08 126 Hulme et al., Global Warming and African Climate Change: a Reassessment 75 Temperature changes

Discernible human influences extend beyond average temperature to other aspects of climate. Human influences have: very likely contributed to sea level rise during the latter half of the 20th century, likely contributed to changes in wind patterns, affecting extra-tropical storm tracks and temperature patterns likely increased temperatures of extreme hot nights, cold nights and cold days; and more likely than not increased risk of heat waves, area affected by drought since the 1970s and frequency of heavy precipitation events127. 3.2.2.3 As we have demonstrated, distinguishing between the effects of GHG-induced Climate Change and natural climate variability is an exercise prone to multiple uncertainties. However, whether a result of exogenous anthropogenic CC or naturally driven variability in the continents climate systems, communities and countries all across Africa are already contending with ongoing climate changes and their associated environmental, socio-economic, cultural, political and psychological impacts. It is important to realize that climate changes pose a problem to Africas societies whose magnitude they may not be well equipped to face. Social scientists at the interface of climate and societal impacts are trying to scope out the magnitude of the problem that enhanced greenhouse effect may pose for African climate and for African resource managers128. However this is an arduous challenge due to the large uncertainties that remain associated with the projection of climate impacts in Africa. Africa is not aided in this by the relative absence of weather stations on the continent (see fig. 3.8). Conclusion

127 IPCC, Fourth Assessment Report 128 Hulme et al., Global Warming and African Climate Change: a Reassessment 76

Relative Density of Weather Stations in Select African countries, with Africa average. Source: UNPD 2007/08129

Despite these large uncertainties, climate scientists, working under the auspices of the IPCC Fourth Assessment Report, have ventured to state what they do know, based on their model projections, and with which levels of confidence. Table 3.2 summarizes the state of the knowledge on impacts projected to hit Africa as a result of CC, by variable of interest and in the short-, mid and long-terms.

129 UNDP, Human Development Report 2007-08 77

3.2.2.4
Projection Variable
Temperature

Projected Climate impacts in Africa

Projection Confidence level

Rainfall

Sea Level

Extreme events

Scenarios suggest warming across Africa ranging from >0.2 to 0.5C per decade, in excess of natural est. temperature increases Africa could be on average between 2 and 6C warmer in 100 years time (Hulme et al., 2001) Climate Change very likely to impose net annual costs, which will increase over time as global temperatures increase. CC likely to bring an exacerbation of current rainfall variability Models however do not agree as to the magnitude and even direction of change in rainfall over the continent (large uncertainties for Sahel) Consequences of rainfall changes on African economies highly dependent on rain-fed agriculture are enormous. Implications for all sectors of human activity, notably health and agriculture. Global expected sea level rise: 88cm by year 2100 (worst case scenario) and a mean of 48cm (IPCC, 2001), with large local differences due to tides, winds and atmospheric pressure patterns. For Africa, towards end of the 21st century, projected sea level rise will affect low-lying coastal areas with large populations. Cost of adaptation could amount to at least 5 to 10% of GDP, depending on costs of coastal protection versus costs of land-use relocation; potential for movement of populations and infrastructure Large consequences on human coastal settlements and infrastructure, notably on tourism (Niang-Diop et al., 2005) Possible impacts on other sectors: changes in water supply, agricultural productivity loss and human migration (according to the local dynamics of human demography); - Salinization of irrigation, estuaries and fresh-water systems - Possible decimation of mangroves, coral reefs and other natural buffer ecosystems - Increased risk of deaths and injuries by drowning in floods Very likely contribution of SRL to coastal extreme high sea level/tides, storm surges (SREX, 2011) and tropical cyclones CC will bring with quasi certainty over most land areas warmer and fewer cold days and nights, as well as warmer more frequent hot days and nights, with the following impacts: - Increased yields in colder environments; decreased yields in warmer environments; increased insect outbreaks - Reduced energy demand for heating; increased demand for cooling Hot extremes, warm spells and heat waves very likely to be more frequent, with: - Reduced yields in warmer regions due to heat stress - Increased water demand; water quality problems, e.g. algal blooms - Increased risk of heat-related mortality, especially for the elderly, chronically sick, very young and socially isolated - Reduction in quality of life for people in warm areas without appropriate housing; impacts on the elderly, very young and poor Very likely increase in the frequency of heavy rainfall events, notably in East Africa (SREX, 2011) with: - Damage to crops; soil erosion, water-logging of soils - Increased risk of deaths, injuries & infectious respiratory/skin diseases - Disruption of settlements, commerce, transport and societies due to flooding: pressures on urban and rural infrastructures; loss of property - Adverse effects on quality of surface and groundwater Likely increase in areas affected by drought, with: - Land degradation; lower yields/crop damage and failure; livestock deaths; increased risk of wildfire; more widespread water stress

Temperature projections for Africa remain robust

Future changes in precipitation are much less confident and less well defined, but trend is towards more erratic rainfall. Accelerated Sea Level Rise is one of the most probable consequences of CC. Sea level rise under warming is inevitable. Projection limitations include: weakness in SLR predictions over Africa, as well as in predictions of how human systems will react.

There is now higher confidence in projected patterns of warming and other regional-scale features, including changes in wind patterns, precipitation and some aspects of extremes.

78

Health

Water availability

Agriculture

- Increased risk of food and water shortage; increased risk of malnutrition; increased risk of water- and food- borne diseases - Water shortage for settlements, industry and societies; reduced hydropower generation potentials; potential for population migration Likely increase in the intensity of tropical cyclones, with: - Crop damage; windthrow/uprooting of trees; damage to coral reefs - Power outages causing disruption of public water supply - Increased risk of deaths, injuries; post-traumatic stress disorders - Disruption by flood and high winds; withdrawal of risk coverage in vulnerable areas by private insurers; potential for loss of property Under all scenarios, creation of new Climate Change refugees Shifting disease patterns projected for malaria, meningitis&cholera: Possible extension/contraction, depending on location, of areas suitable for malaria by 2020, 2050 and 2080, with a 5-7% potential increase in malaria distribution (mainly in the highlands) Consensus among all scenarios that by 2100, the highlands of eastern Africa and areas of southern Africa likely to become more suitable for malaria transmission Possible contraction of the malaria risk zone by 2050 continuing into 2080, in large parts of Western Sahel and southern central Africa, likely to become unsuitable for malaria transmission Possible increases in floods will increase the propensity of water- borne diseases such as malaria and cholera Migration-related health effects to be expected with increased SLR Meningitis and cholera also likely to change patterns with exacerbated climate variability and change. High confidence that many semi-arid areas (e.g. southern Africa) will suffer a decrease in water resources due to climate change The population at risk of increased water stress in Africa, for the full range of SRES scenarios, is projected to be 75-250 million by the 2020s and 350-600 million people by the 2050s However, the impact of CC on water resources is not uniform across the continent. Likely increase in the number of people who could experience water stress by 2055 in northern and southern Africa; in contrast, more people in east and west Africa will be likely to experience a reduction rather than an increase in water stress. By 2020, yields from rain-fed agriculture could be reduced in some countries by up to 50%, with decreased crop growth periods By the 2080s, significant decreases in suitable rain-fed land extent and production potential for cereals are estimated Wheat production likely to disappear from Africa by 2080s Notable reductions in maize production likely in Southern Africa under possible increased ENSO conditions By 2080, a 5 to 8% increase in arid and semi-arid land in Africa projected under a range of climate scenarios (est. 60-90 million ha) By 2100, parts of the Sahara likely to show agricultural losses of between 2 to 7% of GDP. Western and central Africa also vulnerable, with impacts ranging from 2-4% of GDP. Not all changes in climate and climate variability will be negative Agriculture and the growing seasons in certain areas (e.g., parts of the Ethiopian highlands and southern Africa such as Mozambique) may lengthen under climate change due to a combination of increased temperature and rainfall changes Mild climate scenarios project further benefits across African croplands for irrigated and, especially, dryland farms Even under these favorable scenarios however, populated regions of the Mediterranean coastline, central Africa, western and southern Africa expected to be adversely affected.

Less confidence however in the global decrease of tropical cyclone numbers Owing to vigorous debates among climate and health scientists, robust knowledge on future behavior and trends of climate-sensitive diseases. Socio-economic impacts of changing malaria, meningitis epidemic outbreak patterns are difficult to estimate. Robust projections for impacts on water availability. But uncertainties for water access remain large: Current assessments do not fully capture multiple future water uses and water stress Assessing future trends in agricultural production in Africa remains exceedingly difficult.

Table 3.2: State of Knowledge on Projected CC impacts in Africa. Source: IPCC, 2007 (unless otherwise noted)

79

Table 3.2 reveals that projected Climate Change impacts for Africa span across the whole continent, with implications for all sectors of human activity (agriculture, industry, tourism, water availability and ecosystem services). Many of these projections have already begun to be observed. As Table 3.2 also highlights, Climate Change will not bring only loss and risks however; it will also open new opportunities for gain in areas where there were none (for instance, new temperate areas suitable for agriculture and winter tourism) and reduce risks in areas where these were being suffered (e.g.: contraction of the malaria risk zone in the Western Sahel, but expansion into southern Africa and the highlands of East Africa; possible increase in water stress for millions in northern and southern Africa, but decrease in water stress for west and east Africans). The impacts of Climate Change will not be homogeneous across Africa. The shifting risks and new opportunities that will be brought by Climate Change are two sides of the same coin. Albeit unevenly distributed across the continent, within countries and within communities, there will be both losers and winners from Climate Change. The potential for gain from Climate Change is there nonetheless, provided households, communities and countries in Africa are able to seize the new opportunities and adapt quickly to new climate equilibria. Finally, it also appears clearly from table 3.2 that the lines of causality between changes in climate variables and societal impacts are not linear. Climate variables and human dimensions interact in complex ways to buffer or compound the impacts of Climate Change on society, with multiple lines of causality and feedback loops from one to the other. Climate Change impacts on the health offer a prime example of the complexity of climatehuman systems interactions. Likely feedback loops from changing disease patterns to other vulnerability factors or background

80

stresses130, for instance conflicts, migrations and HIV/AIDS with which they will interact, are likely to render populations more susceptible to other diseases (e.g.: cholera) and malnutrition. Human dimensions that mediate CC impacts in poor settings demonstrate the manifold non-linear loops of interaction between climate and human variables, and the layers of complexity. These multiple layers of causality cannot be reduced to a simple cause and effect relationship.

Large uncertainties remain as to the outcome of each climate variable change- human dimension interaction. How households, communities and entire nations will respond to changes in their climate variables remains the single largest determinant of the outcome of climate stresses. Indeed, despite multiple educated guesses, knowledge on the likely socio-economic impacts of Climate Change in Africa is not robust and uncertainties remain large on how societies will respond to the impacts of Climate Change. Noteworthy is the paucity of literature on how the projected climate change impacts projected will be borne by poor countries, notably in the least-capacitated area of the world to address them i.e: Africa. It is widely recognized though that Africa is one of the regions most vulnerable to the projected impacts of CC and variability. Given the continents high socio- economic vulnerability and low adaptive capacity, projected climate changes have potentially explosive impacts on Africas fragile political and economic systems. However how precisely these impacts are likely to play out in African countries remains a black box labeled adaptive capacity to multiple stresses. Africas vulnerability to climate change largely depends on its current and future adaptive capacities. Multiple stresses heighten vulnerability to climate stress in Africa.

130 IPCC, Fourth Assessment Report 81

Several of these stresses are likely to be compounded by climate change and climate variability in the future131. As argued, CC impacts may further constrain development and the attainment of the MDGs in Africa. Adaptive capacity and adaptation thus emerge as critical areas for consideration on the continent132. Which exact adaptive capacity will be needed on the continent to face what particular climate impacts are the crux of this dissertation, which focuses on the governance requisites of effective climate change adaptation. How are the already fragile political and economic systems of African countries to confront the far-reaching and multi-dimensional projected impacts of CC? CCA raises key challenges for the African State, challenges that get to the heart of age-old governance and institutional questions that have haunted the African State since the winds of Independence blew over the continent in the 1960s. In the next and final section of this part, we attempt to make sense of these multiple imperatives faced by the African States under a changing climate, to be delivered as one under the Development endeavor, and propose a conceptual framework bringing together the manifold concepts involved in the governance of CCA in Africa, and the institutional underpinnings of such governance.

131 IPCC, Fourth Assessment Report 132 Ibidem 82

3.3 Governing CC in Africa: Definitions and Conceptual Framework Only in recent years has CC started to be perceived as a central development challenge, that could serve as a potential significant limiting factor to development (see UNDP, 2007-08133 and World Bank World Development Report, 2010134). Today scholars and practitioners alike are beginning to appreciate the implications of the massive projected impacts of Climate change (reviewed in the section above) for development. Still no framework exists to help us structure our understanding of CC impacts on development at the national and sub-national levels in Africa. In large part this situation is due to the still sizeable uncertainties regarding how CC will exactly manifest itself on the continent. In even larger part, however, this situation is owed to our limited understanding of how human systems and societies in Africa will react to, anticipate, buffer and adapt to projected CC impacts; and how CCA, DDR and governance are linked as one under the Development agenda. Therein lies the relevance of analyzing how CCA, DRR, and Governance in Africa are, within the wider framework of the Development endeavor. We develop these conceptual linkages in the next sections, beginning with the concept of Climate Change Adaptation (CCA), and building on it one concept at a time. 3.3.1 Meeting the CCA imperative: Context & Challenges 3.3.1.1 Defining what constitutes Adaptation is a challenge. In human systems, it is the process of adjustment to actual or expected climate and its effects, in order to 133 UNDP, Human Development Report 2007-08 134 The World Bank, World Development Report 2010: Sustainable Development in a Dynamic World (Washington
D.C.: Oxford University Press, 2010).

Adaptation vs. Development?

83

moderate harm or exploit beneficial opportunities135 . The goal of adaptation is to build the capacity of a country or society to adapt to new climate risks and equilibriums a term coined as adaptive capacity. However adaptive capacity is intimately connected to social and economic development but is unevenly distributed across and within societies136. The IPCC further notes that: in several sectors, climate response options can be implemented to realize synergies and avoid conflicts with other dimensions of sustainable development. Decisions about macroeconomic and other non-climate policies can significantly affect emissions, adaptive capacity and vulnerability137. Thus dissociating Adaptation from Development, and preventing negative externalities from one to the other, is a difficult question. We offer our conception of the question in the next section, one of Climate Change Adaptation and Sustainable Development not as trade offs but rather of the former as a subset of the latter. How to adapt however is a difficult tricky question. A wide array of adaptation options is available, but more extensive adaptation than is currently occurring is required to reduce vulnerability to CC. There are barriers, limits and costs, which are not fully understood138. Examples of planned adaptation are displayed in Table 3.3.

135 IPCC, SREX 136 IPCC, Fourth Assessment Report 137 Ibidem 138 Ibidem

84

Table 3.3: Examples of planned adaptation. Source: IPCC, 2007

3.3.1.2 For developing countries, the looming implications of footing the adaptation bill are also cause for concern. The World Banks global track study reveals that by 2020, the annual costs of adaptation for developing countries will range from $75 billion to $100 billion per year; of this amount, the average annual costs for Africa would be about $18 billion per year139.
139 World Bank, Economics of Adapting to Climate Change, Synthesis Report (Washington DC: World Bank, 2010)

Financing Adaptation in Africa

85

 One further point in the Climate Change debate, subject to much contention during the global negotiations on who pays for Adaptation, is its irony. It is the poorest who did not and still are not contributing significantly to greenhouse gas emissions that are the most vulnerable140. Indeed, Africas contribution to global carbon emissions amounts to a mere 3% of total global emissions, yet it is projected to suffer its most severe impacts.
Oceania 2% Africa 3%

CO2 emissions by region

LA 1%

Europe 16%

North America 27%

Asia 50%
(including China, NICs and India)

Fig. 3.9: The irony of CC. (adapted from CARMA.org 2010 data)

Thus, future generations are not the only constituency that will have to cope with a problem they did not create. The worlds poor will suffer the earliest and most damaging impacts of Climate Change141. Responding to climate change will require the integration of adaptation into all aspects of policy development and planning for poverty reduction (UNDP 2007/08). As finally recommended, success [in the development endeavor] will have to involve a great deal of adaptation, because climate change is still going to affect the poorest countries significantly even if serious efforts to reduce emissions
140 Steiner & Dervis in UNDP, Human Development Report 2007-08 141 UNDP, Human Development Report 2007-08

86

start immediately. Countries will need to develop their own adaptation plans but the international community will need to assist them142. Some noteworthy international efforts to support adaptation in Africa have already taken place. Recognizing the imperative to adapt to past-century emissions we are committed to and the need to assist developing countries to plan for development with climate adaptation in mind, the UNFCCC at COP-7 commended all party countries to establish National Adaptation Plans of Action (NAPAs). In this context, Least Developed Countries benefited from LDCF funding to the amount $195,000 to develop their National Action Plan for Adaptation to Climate Change (NAPAs) documents, identifying each countrys most urgent and priority needs for adaptation and evaluating its vulnerability in the face of climate change143. All three (3) country cases treated in this research, Senegal, Uganda and Mozambique, received LDCF support, and developed their NAPAs in the early 2000s. Furthermore, at the Conference of the Parties (COP15) held in December 2009 in Copenhagen developed countries pledged to provide new and additional resources, including forestry and investments, approaching USD 30 billion for the period 2010- 2012 and with balanced allocation between mitigation and adaptation. This collective commitment has come to be known as "Fast-start Finance" (FSF). Following up on this pledge, the COP in Cancun, in December 2010, took note of this collective commitment by developed country Parties and reaffirmed that funding for adaptation will be prioritized for the most vulnerable developing countries, such as the least developed countries, Small Island Developing States and Africa. As such, Africa is expected to receive a total Fast-start finance of up to 30billion USD by 2020. What a review of the NAPAs reveals however is that the endeavor of coping with the impacts of climate change in African countries stands against a myriad of severe 142 UNDP, Human Development Report 2007-08
143 NAPA Senegal, 2006

87

constraints. Archbishop Desmund Tutu of South Africa gives an eloquent glimpse at what some of the challenges of enabling Adaptation in the context of Africas governance are. In a world of adaptation apartheid he states, the adaptation challenge of developing countries has to be met by governments operating under severe economic, social and governance constraints144. A preliminary survey of the status of Climate Change Adaptation in African countries reveals that chief among the number of obstacles thwarting its serious tackling are: lack of political will to take on the issue of CCA seriously; the lack of integration of Climate Change adaptation as an item in Poverty Reduction Strategic Papers (PRSPs); paucity of reliable information and knowledge on climate risks and projected impacts (meteorological information an imperative for adaptation in countries where high dependence on rain-fed agriculture is still a reality, such as those in Africa), Uncertainties that remain large in what future climate will resemble (large range of possibilities, flexible adaptation pathways needed), the difficulty of putting a price tag to adaptation (Adaptation being so akin to Development) and finally international support for Adaptation, though this is slowly changing since the Copenhagen accord with more money streaming into Adaptation, notably in Africa. This is a non-exhaustive list, but it sheds light on the constraints to enabling effective CCA on the continent. 3.3.2 CCA & Development: where do they intersect? A necessary starting point to any conceptual understanding of Climate change linkages with the development endeavor, we purport, is a holistic view onto the Climate Change challenge. Indeed, CC cannot be analyzed sectorally, but holistically, as a challenge to overall development that requires innovations and adapted solutions mainstreamed into national plans and across all levels of public decision- making (both geographically and sectorally). UNDP proposes further that the most 144 UNDP, Human Development Report 2007-08 88

optimal pathway to address the climate change challenge through the correct holistic lens is via the development of climate-resilient, low-carbon development strategies, which address both the adaptation and mitigation imperatives of climate change at the national-level, and exhort countries to think holistically (see UNDP 2011145; also UNFCCC NAMAs and NAPs, UNFCCC 2011)146 . At a more conceptual level, my approach is much akin to that of Amartya Sens discourse on poverty as deprivation, in which he depicts poverty (or underdevelopment) as a state of vulnerability to exogenous shocks (e.g.: international market price fluctuations, disease, inflation, death of breadwinner). Hence, under this conception, the Development endeavor consists of building the resilience of poor persons and communities to all exogenous shocks, transforming their former vulnerabilities into capacities, safeguarding their assets, towards the goal of enabling people to exert their full capacities and lift themselves out of poverty. My conception of Climate Change fits neatly within this conception of development as protection from exogenous shocks. Under this conception, climatic shocks are treated as just one of many other exogenous shocks (economic, sanitary, political) that threaten to send vulnerable persons and communities reeling further down into poverty. Thus protection from climate shocks is a needed requisite for development, just as protection from plummeting cash crop export prices on the international market, skyrocketing food prices, or sudden death of a breadwinner are needed. In this conception, Climate Change Adaptation (CCA), or the process of safeguarding oneself, ones community or ones nation from the impacts of exogenous climate shocks, current and future, is thus an important sub-set of the Development endeavor. This conception of adaptation as a sub-set of development stands at antipodes with a number of mainstream conceptions, which depict CCA as
145 UNDP, Climate-resilient low-carbon growth (2011) 146 UNFCCC, NAMAs and NAPs (2011)

89

an endeavor that overlaps with development in many regards, but still sits outside of mainstream development, at the same level as Mitigation (see figure 3.10).

Fig. 3.10: Where CC meets Development: Old and new conceptions of Climate Change Adaptation (left: old conception147; Right: new proposed and increasingly accepted notion of CCA as a subset of the Development endeavor)

Climate Risks:

Increased climate hazards / Variability

Adaptation Activities:

Climate Risk Management Resilience building Vulnerability reduction

Definition: Timescale of Action:

COPING Near-term (years)

NEW CLIMATE EQUILIBRIUM Inter-decadal Variability Longer-term CC and Change Building codes Process change Retrofitting Paradigmatic shifts Example: change in Keeping agricultural systems (from development rain-fed to climate on track in the independent) face of Business unusual evolving risks New development paradigm CLIMATE PROOFING TRANSFORMATIONAL CHANGE Near to mid-term Mid- to long term (years to decades) (decades to century)

Fig. 3.11: The different timescales of Climate Change Adaptation. Source: author.

147 Refer to Climate and Development Network (CDKN) Framework, www.cdkn.org 90

3.3.3 CCA Development & Governance This leads us to the issue of the Governance of CCA. Given the increase in climate risks projected with Climate change, the State is called on even more insistently to take responsibility for safeguarding its citizens in light of the new threats. This responsibility is explicit in most national constitutions across Africa, where is explicitly stated that the Republic shall be responsible to protect the physical integrity and ensure security of its citizens. The Constitution of Senegal for instance, in its Act No. 2001-03 of 22 January 2001, part II on civil liberties and the human person, economic and social rights, and collective rights, Article 7, states that The human person is sacred. H/she is inviolable. The State has the obligation to respect and protect him/her. Every individual has the right to life, to liberty, to security (emphasis added by author), to free development of his/her personality, to physical integrity including to protection against all physical mutilation148. That of Uganda mandates the State to ensure that efficient machinery is in place to address all hazards and risks that may lead to calamitous disasters for a community149. An interpretation of these texts, in light of a national climate shocks, brings to light the States primal responsibility to ensure security, well-being and integrity of all of its citizens in peace time and in time of crisis. Both these texts make it a constitutional responsibility of the State to provide relief to victims in times of national disasters, or otherwise address disasters, and posit disaster management as a public good to be delivered to all citizens, regardless of their sex, origin or birth place. In light of Climate Change, with projected impacts high, the States constitutional responsibility to protect and ensure the security of its citizens during times of climate crisis becomes even more primordial.
148 Constitution of Senegal, 2001 149 In Uganda National Policy for Preparedness and Management of Disasters, 2010

91

In this dissertation, we posit disaster prevention as a public good, at the same rank as education or health. We will look into the systemic reasons that prevent governments from delivering on this public good. The insights we derive from our cross-country investigation are valuable for all the scholars attempting to understand why CCA-DRR policies, and public goods delivery in general, is not forthcoming from a score of governments in Africa. More fundamentally, Climate Change lays bare the key governance, accountability and institutional challenges which need to be addressed by African countries. CC adds an additional stressor to the multiple ones that already encroach on local livelihoods (e.g: market price fluctuations, weak social safety nets, etc.), which institutions, linking local and national levels, have to contend with and address. Ribot confirms this notion of Adaptation as one sub-set of development endeavor, an attempt to reduce one vulnerability source out of multiple sources of vulnerability communities face: loss of a breadwinner, market prices, climate vulnerability leading to a loss of assets150. Furthermore, since Climate impacts are felt most acutely at the local level, local institutions matter immensely in supporting communities to adapt. Institutions at the local level matter to define rules to address CC impacts and serve as custodians of knowledge to address new risks and opportunities linked to CC (Denton et al., forthcoming). Institutions are key to govern the issue of Climate Change, and an institutional framework or regime is key to effectively govern the new risks and opportunities posed by a changing climate.
150 Ribot, Jesse, Vulnerability does not Fall from the Sky: Towards Multi-Scale, Pro-Poor Climate Policy, In

Social Dimensions of Climate Change: Equity and Vulnerability in a Warming World, Mearns, Robin and Andrew Norton (eds.), 2010.

92

Yet Africas institutions to manage development challenges and build in channels for downward accountability remain wanting in many regards (see UNECA Governance Report 2005151; also Tall 2009152). Efforts towards effective power decentralization and democratic rules of the game that ensure electoral turnover of legitimate leaders across the continent remain patchy at best. Faced with the sheer magnitude of the challenges to be brought about by Climate Change, countries across Africa, and institutions within them, will have only one of two options: to Adapt or to Perish, in metaphorical terms. Therein lies a phenomenal opportunity for countries to reform their governance structures, and innovate new ways of ensuring efficient institutional coordination, spanning local to national, in order to meet local development priorities and effectively deliver on the CCA imperative. 3.3.4 Disaster Risk Reduction This responsibility becomes even more primordial given the projection of rising HMDs. In light of the looming prospect of likely shifts in global climate and weather patterns, that would potentially increase intensity and frequency of extreme weather events in Africa153 , managing the causal factors of disasters, an approach known as Disaster Risk Reduction (DRR), becomes once again an urgent priority. UNISDR defines DRR as the concept and practice of reducing disaster risks through systematic efforts to analyze and manage the causal factors of disasters, including through reduced exposure to hazards, lessened vulnerability of people and property, wise management of land and the environment, and improved preparedness for adverse events154.
151 United Nations Economic Commission for Africa, Governance Report (2005) 152 Tall, A., Learning from the Pre-colonial Past (2009) 153 IPCC, SREX 154 UNISDR, 2009

93

The projection for rising climate-related disasters puts DRR back on the agenda as an urgent development priority. DRR and CCA are hence interwoven as CCA-DRR, an essential joint attempt to build the resilience, coping capability and adaptive capacity of communities vulnerable to exogenous climate shocks. If we do not reduce vulnerability to existing climate risks, many of the MDGs do not stand a chance of being met155. In the context of Africas fragile states however, already stretched thin with many public responsibilities that they are not fulfilling, how precisely will this responsibility be carried out? Will it be at the expense of other primordial state duties, with a diversion of the states development funds to the more urgent crisis response every time a climate crisis will strike? Or will the issue not be addressed at all, passed on to international donors and NGOs to address it? The age-old dilemma between state capacity and state responsibility is reenacted (see World Bank 1992 World Development Report on the Role of the State)156. Where should responsibility for climate change adaptation reside? If States in Africa are not capable of taking the CCA challenge fully on, due to various constraints, should private mechanisms be taking on the responsibility of safeguarding persons, communities and national assets in light of the projected CC risks? Private insurance has risen in recent years as a response to this fundamental question. Multiple examples of agricultural weather-index based insurance schemes can thus be cited across the context (see Suarez and Linnerooth-Bayer 2010157; also Patt et al., 2009158). Even these insurance pilots had the State play major roles, co- convening dialogue, building capacity, managing weather stations used for index
157
155 UNDP, Human Development Report 2007-08 156 World Bank, World Development Report 1992

Climate Change 1 2 (2010): 271-278

Suarez, P., and Linnerooth-Bayer, J., Microinsurance for local adaptation, Wiley Interdisciplinary Reviews:

158 Patt, A.G., Peterson, N, Carter. M, Velez, M., Hess, U. and Suarez, P., Making index insurance attractive to

farmers, Mitigation and Adaptation Strategies for Global Change 14 (2009): 737-757

94

measurements, etc. In the context of Africas poorest constituencies, the promise of insurance for all is not a durable solution (Patt et al., 2010159). Furthermore, what of the areas of life affected by climate risks beyond livelihoods, such as lives (climate disasters high death toll can be cited in this instance), heritage and national infrastructure? These lingering questions seem to point that much responsibility remains with States to take on a part, large or limited, of meeting the CCA imperative. 3.3.5 The Triple Bottom Line: Governing CCA-DRR-Development Thus with CC, the African State finds itself with more work to do. In addition to its traditional development mandate, it now has to meet the triple bottom line of DRR-CCA-Development, delivering on all three as one within the development endeavor. The new climate change challenge brings the imperatives of eliminating African poverty, climate dependency and disaster vulnerability all under the banner of the development endeavor. As such, addressing CC will have to be done within the prism of achieving the three different goals, delivering on them as one under the Development agenda, and meeting the triple bottom line of CCA-DRR-Development. Efficient and socially beneficial institutions are needed however to deliver on this new triple bottom-line, and its new imperatives put on the table by a changing climate. Before a new issue can be governed, a country requires sets of actors/networks, rule making systems and formal and informal rules that steer societies towards pre-defined collective goals160. New or adapted institutions, or rules of the game, clusters of rights rules and decision-making procedures, will be needed to effectively confront the new issue of CC, and ensure that appropriate 159 Patt, A.G., Suarez, P. and Hess, U. How much do smallholder farmers understand and want insurance?
160

Evidence from Africa. Global Environmental Change 20 1 (2010): 153-161 Heike, Shroder, Social Dimensions of Global Environmental Change, IHDP, University of East Anglia

95

responses and adaptation measures are provided to new local/national CC risks and opportunities. In a fast changing world, the issue of institutional innovation to address this new challenge of the 21st century in Africa is an important one. What our research reveals is that in practice, countries have already begun experimenting with new climate institutions; whether or not these are efficient and socially beneficial is another question. 3.3.6 Why new/adapted institutions to meet the triple bottom line? Our scoping of African organizations addressing the new challenge of CC reveals that as of now, organizations that have traditionally existed to achieve the goals of each subset of the triple bottom line (CCA-DRR-Development) have been kept apart. Institutional innovation has not taken place by and large. Development planning bodies and agencies are different from the hubs working on CC, which are housed under a ministry different from the one that addresses disasters, with weak cross-institutional linkages between the three different sectors kept by and large separate. The imperative of CCA however brings all three sectors together. As impacts of rising climate-related disasters are felt by and large by communities at the frontlines, the latter do not distinguish between whether it was lack of CCA, DRR or development which brought about the severe CC impacts they experience. All three are needed components to buffer an exacerbation of climate impacts at community level. Thus, there is an urgent need to bring the three institutional silos and approaches in line to build community and national resilience in light of projected increase in HMDs. 96

How in practice though should this be achieved? Much variation exists already within African countries innovations address the triple bottom line. A look into African countries endogenous institutions solutions as theyve begun to confront more frequent HMDs is an important endeavor to derive examples of solutions. As our research demonstrates, such solutions can be found in Africa. Following that, we will delve into the crux of our enquiry: what made countries embark on a path of institutional innovation to address the new challenge of the 21St century in Africa, whereas others still lag behind? Without much further ado let us expose the methods used to test our hypotheses of institutional change. 3.4 Glossary of Concepts, Acronyms And Key Terms This section lays down the definitions of all key terms we will be drawing on throughout this thesis, and the meaning in which we will be using them. Adaptation: Adaptation in human systems is the process of adjustment to actual or expected climate and its effects, in order to moderate harm or exploit beneficial opportunities In natural systems, the process of adjustment to actual climate and its effects; human intervention may facilitate adjustment to expected climate161. It is the capacity of a country or society to adapt to new climate risks and equilibriums. Adaptive Capacity:
161 IPCC, SREX

97

The goal of adaptation is to build the capacity of a country or society to adapt to new climate risks and equilibriums a term coined as adaptive capacity. However adaptive capacity is intimately connected to social and economic development but is unevenly distributed across and within societies162. CC: Climate Change A change of climate, which is attributed directly or indirectly to human activity that alters the composition of the global atmosphere and which is in addition to natural climate variability observed over comparable time periods163. CCA: Climate Change Adaptation Process of safeguarding oneself, ones community or ones nation from the impacts of exogenous climate shocks, current and projected Climate Variability Climate variability refers to shorter-term fluctuations that occur at the timescale of a season, months or a few years (refer back to fig. 3.1). Cyclone: An atmospheric system characterized by the rapid inward circulation of air masses about a low-pressure center, usually accompanied by stormy, often destructive weather. Cyclones circulate counterclockwise in the Northern Hemisphere and clockwise in the Southern Hemisphere. It is a violent tropical storm, especially one originating in the southwestern Pacific Ocean or Indian Ocean. Development Endeavor: Building the resilience of poor persons and communities to all exogenous shocks, transforming their former vulnerabilities into capacities, safeguarding their assets,
162 IPCC, Fourth Assessment Report 163 United Nations, UNFCCC, 1992

98

towards the goal of enabling people to exert their full capacities and lift themselves out of poverty. Disaster: A disaster is a severe alteration in the normal functioning of a community or a society due to hazardous physical events interacting with vulnerable social conditions, leading to widespread adverse human, material, economic, or environmental effects that require immediate emergency response to satisfy critical human needs and that may require external support for recovery164. We follow here the definition of a Disaster as the conjugation of a naturally driven Hazard and human-induced Vulnerability. We understand by this that the multiple hazards that make landfall in Mozambique for instance turn into large-scale disasters due to the acute socio-economic vulnerability they meet on the ground.
Hazard * Vulnerability

Disaster Risk = ___________________________ Capacity

A hazard also only becomes disaster if it overwhelms local capacities to cope. Thus a disaster can also be understood a serious disruption of the functioning of a community or a society causing widespread human, material, economic or environmental losses which exceed the ability of the affected community or society to cope using its own resources. It results from the combination of hazards, conditions of vulnerability and insufficient capacity or measures to reduce the potential negative consequences of risk165. DRR: Disaster Risk Reduction, DRM: Disaster Risk Management, Disaster Preparedness UNISDR defines Disaster Risk Reduction (DRR) as the concept and practice of 164 IPCC, SREX 165 UNISDR, 2009 99

reducing disaster risks through systematic efforts to analyze and manage the causal factors of disasters, including through reduced exposure to hazards, lessened vulnerability of people and property, wise management of land and the environment, and improved preparedness for adverse events166. Disaster Risk Management (DRM) involves the systematic development and application of policies, strategies and practices to minimize vulnerabilities and disaster risks in order to avoid or limit the adverse impacts of hazards on lives, economic and social developments in a country. Sustainable development, poverty reduction, good governance and disaster risk reduction are mutually supportive objectives, which need to be delivered as one under the Development endeavor. In order to meet challenges in the implementation of the national development plan, accelerated efforts must be made to build the necessary capacities at community and national levels to manage and reduce risk. The main objective of disaster preparedness is to ensure that in times of disaster, appropriate administrative, legislative and technical measures including procedures and resources are in place to assist those afflicted by the disaster and enable them to cope. Disaster preparedness aims to minimize the adverse effects of a hazard through effective precautionary measures and to ensure timely, appropriate and efficient organization and delivery of emergency relief services167 . Drought: Drought is a disaster driven by a prolonged shortage of rainfall, often generated by dry weather conditions. Drought involves many other factors such as food distribution and market prices, the breakdown of which may lead to a famine. When human populations are dependent on rainfall or ill prepared for rainfall shortages, drought often ensues. 166 Ibidem 167 In Uganda National Policy for Preparedness and Management of Disasters, 2010 100

 Exposure: The presence of people, livelihoods, environmental services and resources, infrastructure, and economic, social, and cultural assets, in locations that could be adversely affected. Flooding: Flooding is a disaster driven by excessive or concentrated rainfall. It can be of two natures: river flooding along rivers (when a river rises out of its bed and spills into is floodable plain) or flash flooding following periods of intense rainfall or consecutive days of rainfall (wet spells). Most flood disasters in Africa are seasonal, and usually occur during periods of intense rainfall as well as on el-Nio years (for East and West Africa) and LA Nina years (for Southern and northern Africa). GHG: Greenhouse Gases (GHG) Grenhouse gases are long-life gases whose increased concentrations in the atmosphere mainly cause anthropogenic Climate Change. The main GHG responsible for CC is carbon dioxide. Methane, nitrous oxide and sulphur hexafluoride are other significant contributors. Altogether combined, these gases are responsible for the GHG-induced changes to the global atmosphere, through a process called the Greenhouse effect. Global Warming: Rise in the global average temperature that occurs when Earths energy balance is set off equilibrium; the amount of energy (shortwave solar radiation) absorbed by Earth becomes greater than the amount of energy emitted back out to the atmosphere (long-wave radiation). Excess outgoing heat trapped by GHG is deflected back to Earth, causing a rise in global average temperatures. Between the

101

decade of the 1970s and the early 2000s alone, Earths average temperature has risen by 0.7168. Governance: Set of actors, networks, rule making systems and formal and informal rules that steer societies towards pre-defined collective goals 169. HMD: Hydro-Meteorological Disaster= Climate shock= Climate hazard= Climate- related disaster= Extreme weather event or climate-related event

Used indiscriminately through this thesis to mean the same thing: the occurrence of a value of a weather or climate variable above (or below) a threshold value near the upper or lower ends (tails) of the range of observed values of the variable170. Institution: North posits that institutions indeed define the rules of the game, both politically and economically; they are the humanly devised constraints that shape human interaction171 . These constraints can be both formal (rules, laws) and informal (behavior codes, cultural norms), and serve to reduce uncertainty and facilitate exchange in the presence of transaction costs. Landslide: Landslides and mudslides are rapid movement of a large mass of mud, rocks, formed from loose soil and water. It usually follows heavy rainfall and high ground water flowing through cracked bed rocks and earth quakes and lead to soil movements. Landslides and mudslides are very difficult to predict but their frequency and extent can be estimated by use of information on the areas geology, 168 IPCC, Third Assessment Report, 2001
169 Heike, Shroder, Social Dimensions of Global Environmental Change, IHDP, University of East Anglia 170 IPCC, SREX

171

North, Institutions, Institutional Change and Economic Performance, 3

102

geomorphology, hydrology, climate and vegetation cover and traditional knowledge. Community settlement on steep slopes and other uncontrolled land use practices increase the likelihood of landslides and mudslides prevalence. Organization: North distinguishes "institutions", which are the rules of the game, from "organizations", which are the actors, or players in the game. According to North, "it is the interaction between institutions and organizations that shapes the institutional evolution of an economy"172, or a society at large. SLR: Sea Level Rise Sea-Level Rise is a rise in global ocean levels, directly resulting from Global Warming. Vulnerability: Vulnerability is the propensity or predisposition to be adversely affected.
172 North, Institutions, Institutional Change and Economic Performance, 6

103

 PART III:

METHODS

104

CHAPTER 4: INQUIRY INTO AFRICAS GOVERNANCE OF CLIMATE RISKS This chapter exposes in detail our research design and methodology. In this dissertation, we set out to answer the following Research Question: How is CC governed in Africa? What factors prompt African countries to adopt new socially beneficial institutions to rise up to the new challenge of CC? We have hypothesized many potential causal factors, or independent variables, chief among which, the role of self-interested rational organizations in bringing about incremental change within the historically-determined institutional framework in order to increase their private gains leading to new institutions or rules of the game in African countries to manage climate shocks when the establishment of these matched organizations private interests (structural institutional change theory). Other possible drivers of climate institutional adaptation could be: conjuncture or accidents of history, for instance, a severe national disaster (North), geography (Engerman/Sokoloff), institutional legacies (Bates); eg: State capacity or yet again, the presence of a strong media. More agent-based explanations of climate institutional change could also be individual agents desire for improved societal performance which brings private interests in line with public ones (Grindle; for instance a politician very dedicated to climate change resilience); new political configurations/constraints at the local level that force national leaders to put CCA local demands on the national agenda (Boone). Also, the role of strong endogenously-developed institutions, downwardly accountable and coherently integrated with local institutions could important. 105

In order to test all these hypotheses, we have devised an experimental design for our research that rigorously isolates spurious causal factors from competing independent variables, and attempts to take the measure of each potential independent variable to the extent possible in the natural, non-laboratory world, in order to fully elicit the pathways of cause and effect between climate impacts, institutional change, and ensuing political institutions. The following section describes our experimental design. 4.1 Research Design Research Question: What factors prompt African countries to govern new the challenge of CC through socially beneficial institutions? Main Hypotheses: Our main hypotheses are summarized in Fig. 4.1.

Cyclone Flood Drought Storm Landslide

Competing Independent Variables (cause):

HMD frequency/ magnitude Donor demand for improved CCA-DRR CLIMATE FINANCE: Aid invlowing to country earmarked for CCA-DRR State capacity / GDP/ capita Dedicated leadership Key political bastion at risk Extensive Media coverage of disaster impacts

Dependent Variable (effect):

Institutional change in the management of climate risks: adoption of socially benevicial institutions to address the new challenge of CCA-DRR

Climate Hazard Hits

Figure 4.1: Hypothesized Independent Variables, and pathway of cause and effect.

106

A two-phase research plan enabled us to answer our research question: phase one is dedicated to identifying policies in place to confront HMDs across Africa, basis for our country case selection; phase two proceeds with an analysis of the national-level policy to reduce HMD impacts in place in each our case studies (policy analysis), while phase three delves into an incentives analysis of why a given policy in each country came to be. Each phase used its own mixed design (of both quantitative and qualitative methods) in order to answer its set research question, involving two units of analysis: the nation-state (for phases 1 and 3) and the community (for phase 3). 4.2 Phase 1: Identification of Climate Disaster Management Policies Across Africa In this section, we delve into the methods utilized to conduct our cross-continental identification of climate-related disaster policies across Africa, and present the results of this classification, basis of our country case selection. 4.2.1 Country Case Selection Protocol: The Spectrum of DRM Country cases have been selected based on a continent-wide classification of the different climate-related disaster management policies in place at the national-level. Currently, knowledge and data collection on the climate disaster management infrastructures and policy contexts of African countries is generally quite poor; against this paucity of knowledge on existing national disaster management policies, most programs supporting CCA-DRR assume a policy vacuum and are applied as standardized policy prescriptions across all African countries, with the concomitant expectation that all target countries will experience similar uptake of their CCA-DRR programs, provided similar inputs from international donors. 107

 In reality however, African countries display great heterogeneity in their disaster management landscapes and national policies, one that reflects the different vulnerability profiles, policy contexts and political incentives systems in place in Africas fifty-five countries. On the spectrum towards effective disaster management, they lay at different levels. Some are akin to Unprepared firefighters in their response to disasters, improvising and intervening only after calamities strike; while others are rather more similar to Prepared firefighters, with disaster preparedness plans adopted and rehearsed, and contingency plans in place to face all possible calamities; whereas a third, smaller group of countries are Disaster averters, who fulfill all the prepared firefighter conditions, but also invest in risk reduction. The latter countries experienced a paradigm shift in their approach to disasters, moving focus away from the hazard towards the underlying risk factors. In this dissertation, we introduce this three-tiered classification of African countries as: Unprepared firefighters, Prepared Firefighters and Disaster Averters. Ultimately, there is a need to improve our understanding of how and why countries move from being Unprepared firefighters (policy type I in our hypothesized classification of African countries climate disaster management policies), to Prepared Firefighters (policy type II) and Disaster Averters (policy type III). Our classification was conducted using indicators of ranked importance chosen on the basis of well-reasoned assumptions about what constitutes good, effective Disaster Risk Management (DRM; see table 4.1). It is important to note that we made every effort not to select outcome-driven indicators (e.g.: number left dead under DM policy type I) or indicators based on data availability; countries were ranked according to what we believed to be good, measurable indicators that reveal the actual policies currently devised by African governments to address climate- related disasters and their impacts. 108

Summarized in table 4.1 is the hypothesized typology of disaster management policies we used to examine and classify African countries disaster management effectiveness.
POLICY Ex-post response ONLY TYPE Policy Indicators INDICATORS: - No mandated disaster - National mandated disaster - All of type II policy indicators + - Use of scientific knowledge/climate information to anticipate disasters informed by climate information - Functioning Early Warning Systems exist - Rigorous risk and vulnerability assessments/ cartography are conducted to inform long-term risk-proofing and disaster prevention - Disaster prevention is mentioned in National PRSP and streamlined into management (DM) unit exists; management (DM) unit exists or if it does: it does not effectively coordinate across government institutions to ensure good disaster response - No dedicated budget line for disaster - No rehearsed established procedures/responsibilities for disaster response - No national DM strategy in for disaster response - No drills - Track record of delayed and ineffective response (relief >3days after disaster event) - It has a budget/dedicated budget line; relief supplies in stock - DM unit is at cabinet level office); effectively coordinates across institutions - Established procedures/responsibilities for disaster response exist; - National DM strategy/law is disaster response responsibilities - Disaster response drills are conducted - Contingency planning is a Policy Type I: Unprepared Firefighters Ex-post response + Ex-ante preparedness for disasters Policy Type II: Prepared Firefighters Ex-post response + Ex-ante preparedness for disasters + Use of scientific knowledge to inform Disaster planning and prevention Policy Type III: Disaster Averters

(directly under PM/Presidents - Contingency planning

place outlining responsibilities in place, clearly outlines

109

[Bad, delayed, ineffective responders]

regular exercise - Track record of good response (1-3 days within disaster event) - No prevention or use of scientific knowledge/ climate information to anticipate climate-related disasters [Good responders]

development plans - Resources are allocated for disaster recovery/post disaster reconstruction (DRR included) [Disaster Averters]

Table 4.1: Policy typology: Firefighters vs. Prepared firefighters vs. Disaster Averters

In brief, countries with Policy type I, II or III are characterized by the following attributes: 1) Ex-post response only (the Unprepared Firefighters policy): countries choosing not to, or unable, to address disasters until they occur. Countries in this category solely respond to disasters after they have occurred, mobilizing any personnel on duty and dispatching them to disaster sites. For this reason we label them: the Unprepared Firefighters! Official emergency response in the countries of this category is characterized by amateurism and improvisation, and there are no pre-established well-rehearsed procedures to follow. Also included in this category are countries with a dedicated agency for Disaster Management but one without capacity, ability for efficient inter-ministerial planning and coordination, or no budget line. 2) Ex-post response + ex-ante disaster preparedness (The Prepared Firefighters policy): Countries in this category have shifted from response (responding to disasters only after they occur), to preparedness for disasters before they occur, so as to ensure more efficiency and swiftness in disaster response and relief operations. They have the following attributes: a national Disaster Management Strategy or law adopted with clearly defined and appropriated procedures and responsibilities for disaster response; in the national budget a dedicated budget line exists for disaster preparedness and 110

emergency response; a mandated agency for disaster management exists, able to coordinate across ministries and administrations and effectively command disaster response activities during an Emergency. Generally, for all countries in this category an annual Contingency plan is elaborated, and delineates the procedures, actions and responsibilities to be carried out in the advent of all possible disasters; furthermore regular drills of these disaster response procedures take place. As a result, countries in this policy category take no more than three to deliver needed relief to disaster-affected citizens. 3) The Disaster Averters (combining effective disaster ex-post response, ex- ante preparedness AND Disaster Risk Reduction based on scientific knowledge and climate information): Finally countries in this third category have experienced a paradigm shift in their strategy to address disasters and have transitioned from disaster preparedness to prevention. They engage in pro-active Disaster Prevention and have shifted the focus away from the disaster event itself towards the underlying risk factors that create disasters. Countries in this category focus on pre-empting disasters by reducing disaster risk and vulnerability so that Hazards do not turn into disasters, bringing to bear weather forecasts, early warning systems, risk/vulnerability mapping, and other relevant tools/initiatives for climate risk management. Indeed in these countries, disasters are identified as a crosscutting, transversal threat to development, often inscribed in national Poverty Reduction Strategy Papers (PRSPs). Hence, there is strong political commitment to bring all of the nations forces/segments together in order to reduce the countrys vulnerability to disasters and minimize the consequences of disasters on vulnerable segments of the population. Some of these initiatives include, but are not limited to: the of climate and meteorological risk information to inform disaster planning, national/ sub- national contingency plan(s) based on seasonal climate/weather forecasts, functional multi-hazard Early Warning Early Action systems, elaboration 111

and enforcement of long-term disaster risk reduction plans based on sound scientific research (risk identification, resilient building codes, mapping of risk zones, legislative bans on building/human settlement in risk zones, etc.), dedicated and sufficient resources allocated for the implementation of DRR plans at all administrative levels. In brief countries in this category take the threat of disasters very seriously and invest in building the countrys resilience to disasters, managing disaster risk with all means at hand. 4.2.2 Methodological Framework: The Spectrum of Effective Disaster Risk Management in Africa Based on the above storyline of what constitutes an Unprepared Firefighter, a Prepared Firefighter and an effective Disaster Averter, we defined three correspondent policy types (I, II and III) and a series of twenty (20) measurable indicators to proxy and reveal the existence or not of the attributes of each policy type, as well as capture the spectrum of disaster management measures practiced in Africa.

112

 Fig 4.3: The Spectrum of Disaster Risk Management (DRM) Policies in Africa

Our methodological framework, depicted in figure 3, overlaps in many regards with standard theoretical frameworks on what constitutes effective disaster risk management. To use only the World Meteorological Organizations (fig. 4.4), exhorting countries to implement effective disaster risk reduction (DRR), we see that all proposed measures for DRR (Risk identification, Risk reduction and Risk transfer) are encapsulated in our Policy Type III: Disaster Averters, countries that effectively identify, reduce and transfer disaster risk, abiding by all the guidelines highlighted in the Hyogo Framework for Action.

113

 Fig 4.4: Framework for Effective Disaster Risk Management derived from the Hyogo Framework Action (HFA). Source: World Meteorological Organization (WMO)173

Our first two policy types on the other hand The Unprepared firefighters and Prepared firefighters (Red and yellow respectively in our classification) both would simply fall under the category of Ineffective disaster risk management in the standard DRR theoretical frameworks. We amended these however and introduced a tripartite differentiation in replacement of the standard binary DRR framework commonly used, in order to capture the full variation that one sees when analyzing African countries. Indeed, the default baseline in Africa is that of ex-post disaster response (whether effective or not). Most countries in Africa do not have any of the Disaster Risk Management measures called for by the standard international DRR frameworks. Some have some elements: the Prepared firefighters (policy type II), countries that are in practice quite effective disaster responders, building preparedness for disasters but not making use of existing scientific knowledge to inform their preparedness efforts. At the top end of our classification are African countries that bring in climate information and knowledge to reduce vulnerability to climate- related disasters and losses, making provisions for risk identification, financial protection, post-disaster risk reduction and Early Warning Systems; these countries
173 World Meteorological Organization (WMO), accessed March 10, 2011. www.wmo.int

114

are called the Disaster averters. This maps into a sequential spectrum where some countries have very little in place by way of Disaster Risk Management and others have much more in place; hence generating a continuum of Disaster Risk Management (RM) effectiveness ranging from the most unprepared of firefighters (with a score of 0) to the effective Disaster Averter (score of 20). This constituted a much more cogent way of analyzing African countries disaster management policies in our sense. As we shall see in the results section, the highest score attained in Africa was a 17, and the lowest was a 0, thus this spectrum allows us to effectively capture the variation that exists within Africa, while still calibrating it to global standards for what constitutes effective DRM, as well as leaving the space open for possible future regional comparisons. 4.2.3 Indicators, data sources & classification methodology The twenty (20) indicators we used to classify Africas fifty-five (55) countries into our three Disaster Management (DM) policy types are displayed in box 3. Each country was assessed on each of these twenty questions/indicators, and answers to each question were scored on a binary scale of 0-1, then totaled to compute a final DM score for each country.
1. DM unit exists/there is an institutional basis for implementation of DRR-CCA Source: RC, PfA1 + country events analysis 2. National disaster response plan exists clearly defining roles and responsibilities during emergency Source: PfA1 (1), country events analysis + IFRC WPNS 3. National disaster management law has been ratified, promulgating into law of the land the different roles and responsibilities for DM, legal backing for DM (year law was adopted) Source: PfA1 (1), country events analysis + IFRC WPNS 4. DM unit or platform is at cabinet-level/mandated to coordinate across-institutions Source: PfA 1 (4)

115

5.

DM unit has financial capacity: Budget line for DM in national budget/Dedicated and adequate resources for DM are available at all administrative levels Source: PfA5 (1); 2004 reports in preparation for WCDR

6. 7. 8.

DM unit with human capacity: enough staff? Well trained? Source: PfA5 (1) National Disaster Preparedness Contingency plan is elaborated and regularly updated Source: PfA5 (2) Regular training drills and tests are held/DM responsibilities are well rehearsed at all levels of intervention (national to community) Source: PfA5 (2)

9.

Track record of efficient response (government has strong ability to respond); relief provided within 2days of disaster occurrence Data source: RC

10. DM responsibilities decentralized + adequate resources devolved to local level for DRR-CCA Source: PfA1 (3) 11. Most vulnerable are targeted in disaster interventions Source: VFL indicator #6 + HFA 12. Risk assessments (national/local) conducted based on climate knowledge, hazard data and vulnerability mapping Source: PfA 2 (1) 13. Functional multi-hazard EWS exist (with outreach to communities) Source: PfA2 (3) 14. Trans-boundary risk assessments conducted Source: PfA2 (4) 15. Disaster Risks included in school curricula/relevant info on climate disasters is available nation-wide Source: PfA 3 (1,2,4) 16. Infrastructural Disaster Prevention initiatives exist: climate-proofing of infrastructure, cities Source: PfA 4 17. Legislative Disaster Prevention initiatives exist: building codes, bans on human settlement in risk zones, etc. Source: PfA 4 (4) 18. DRM is explicitly mentionned in PRSPs/national development strategy papers, impacts on development are rightfully seen Source: PfA4 (2 + see perspectives futures); World Bank national PRSPs 19. Recovery efforts include DRR measures and mechanisms to prevent future disasters

116

Source: PfA 4 (5) 20. Financial reserves for recovery exist / risk transfer mechanisms Source: PfA5 (3)
Table 4.2: 20 indicators of DRM policies in Africa and data sources for each (Red indicators correspond to policy type I; Yellow indicators correspond to policy type II and Green indicators correspond to policy type III) Key to data sources: PfA=Hyogo Framework Priority for Action: triangulation of responses from HFA Mid-Term Review and Views from the Frontline (VFL) 2009-2010 assessment surveys, where available; International Federation of the Red Coss/Red Crescent (IFRC) WPNS= Well-Prepared National Societies Questionnaire (2009- 2010); World Bank PRSP= Poverty Reduction Strategy Papers, 2010 or latest available; RC= Answers to Questionnaire to Red Cross National Societies Disaster Management Unit.

4.2.3.1 Data Sources: We resorted to a wide array of data sources to be able to get answers to each question/indicator selected in accordance with our typology of disaster management policies. We thus identified several data sources for each indicator in order to attain high confidence/robustness, for each of Africas countries. The paucity of data on national disaster management is a notable fact about Africa; thus setting out to gather data for our selected indicators was a formidable feat. Through qualitative data miningexpert interviews by email, phone and in-person and survey questionnaires emailed to all Red Cross National Societies, country events analysis and use of existing country-level data from previous assessments (mainly, the Hyogo Framework for Action-HFA mid-term review and Views from the Frontline-VFL national progress reports), we gathered data for each country on indicators/questions 1 through 20. A total of fifty-two respondents were interviewed on the DM policies in place in their countries. Additionally fifty news articles and sixty-six official documents (government reports, declarations and legislations) were consulted, for a total of 108 data sources, to arrive at a final score for each indicator (using a binary scoring: 1 when indicator is met, 0 when not). 117

 To generate survey data, climate change adaptation experts gathered at the Africa Climate Change Collective Action forum held in Dar-es-Salaam, Tanzania from June 25-30th 2010, were administered an initial round of an expert questionnaire protocol in order to gather their views on their countries use of scientific knowledge to inform national disaster preparedness plans; this led to the initial hypothesizing of the spectrum of DRM policies in Africa, and generated answers for indicators 12 and 13 (see survey protocol in Appendix). Additionally, Red Cross national disaster managers from Africas fifity-five countries, best poised to characterize the practice of disaster management in their respective countries, being auxiliaries to the State and often the first to be on site in the advent of a disaster, were contacted with three key questions: 1) How long does the government generally take to respond to climate-related disasters? How many days before or after the Red Cross has it taken them in past disasters to get to affected sites? 2) How would you qualify your government's response to disasters: are they firefighters (their response is late, ineffective and delayed)? are they prepared firefighters (effective disaster responders)? or rather do they have strategies in place to anticipate/prevent disasters using available scientific information (disaster averters)? 3) What institution is in charge of disaster management in your country? Is the Red Cross part of any national platform to better manage disasters? Answers to these three questions served to inform indicators 1 through 11. This questionnaire was administered online, translated in French, English or Portuguese depending on the countrys official language. Furthermore, reports of the International Federation of the Red Cross and Red Crescent National Societies (IFRC) Well-Prepared National Societies Questionnaire (2009-2010) were consulted to further confirm assessments of National Society disaster managers, and provide insights on indicators 2 and 3. 118

Moreover, ordinary citizens, government disaster managers and civil society representatives from each country, whenever available, were asked to confirm our assessment of their national DRM policies. Finally, insights from HFA mid-term review national reports, and their bottom-up counterpart, the VFL national assessment reports, where confronted in all the countries where they were available, providing information for indicators 12 through 20countries progress in implementing HFA-recommended DRR measures. All in all, this enabled us to conduct a through assessment and generate a complete cross-continental picture of the national DM policy currently in place in each country, one computed using at least two and up to five credible data sources for each country, confronted and triangulated to give a substantiated score on each indicator. On the basis of country scores on each of the 20 indicators, we computed an overall DM policy score for each country under review. Once all the data collected on each indicator for the fifty-six countries under study, we were able to classify countries using our classification method. 4.2.3.2 To be classified as an Unprepared firefighter, countries needed to satisfy indicators 1-4, in accordance with our storyline for what constitutes an unprepared firefighter, and/or obtain a total score of 4 or below. Prepared firefighters needed to satisfy indicators 5-11, and have an overall score between 5 and 11. Finally Disaster averters satisfied indicators 11-20 and scored above 11 points. Of the 20 indicators, we weighed four more heavily however: Financial and human capacity (indicators 5 and 6): Scoring a 1 on these was a pre-requisite needed to qualify as a type II country (Prepared Firefighters); 119 Classification Method & Weighing:

 Use of knowledge to conduct risk assessments and existence of effective early warning systems (indicators 12 and 13): Scoring a 1 on these was a pre-requisite to qualify as a type III country (Disaster Averters). We weighed these four indicators more heavily using a qualitative method. We maintaining countries in one policy category, until they scored positively on one of the threshold indicators (5-6 and 12-13), which became de facto cut-off points to graduate to the next policy type. Scoring positively on indicators 5-6 was thus needed to move up to policy type II; similarly scoring positively on indicators 12-13 was a requisite to move up to category III. A country such as Djibouti for instance had a total DM policy score of 7which ought to have immediately placed it among the Prepared Firefighters. However because it did not meet indicators 5-6 (its DM unit did not have the financial and human capacity to carry out its duties), it could not justifiably be ranked as a Prepared Firefighter. As a result it was maintained in the Unprepared Firefighter category. When all countries under study were rigorously assessed, on each indicator out of the 20 selected, then classified according to their total DM policy score and then weighed according to whether or not they met the indicators of most importance, it became easy to generate a continental map of African countries by national disaster management policy in place: Firefighters shown in red, Prepared Firefighters in yellow and Disaster Averters in green (see chapter 5).

120


Fig. 4.6: Results of Cross-continental classification: The Patchwork of Disaster Management policies across Africa (Left insert: un-weighted map; Right: weighted map)

4.3 Resulting Country case selection Following our classification of all of Africas countries according to their disaster management policy type, I selected a country case study from each policy group, and used that country to make inferences for the entire policy group. Indeed, our comparative analysis required only one country per policy group: one Firefighter case, one Prepared Firefighter case and one Disaster averter case. However, looking at the maps in fig. 4.6, we were still left with multiple options for country case selection, namely: thirty-three (33) country case options to infer for the countries shown in red, or Unprepared Firefighters; nine (9) for countries shown in yellow, the Prepared firefighters; and six (6) country options for countries shown in green, or the Disaster Averters. How to select among these to come up with three country cases, and proceed with our comparative analysis? The only subjectivity in the entire case selection was introduced at this stage. When selecting among different country options within each policy group, two main criteria were followed. First, were personal preferences and ability to conduct research safely and successfully in certain countries relative to others. Secondly, 121

came the concern to keep selected countries as comparable as possible and similar on at least a few variables (see country comparability matrix 4.5) in order to control for competing independent variables that could explain differing institutional change trajectories. Keeping in mind these determining factors, we narrowed our list of country case options to five countries: Senegal for the Unprepared firefighter country group (in red, selected due to existing field research connections and easier access to key informants); Mozambique to infer for the Disaster averter country group (in green, selected following the chance of a concurrent UNDP research program in Mozambique, which our research was tagged onto in with ease); and finally Uganda/Ethiopia/Morocco as options left for the Prepared firefighter country representative, displayed in yellow. Table 4.5 was used to adjudicate and come to a final decision on the Yellow country case.
N Senegal Moz- ambique Ugan- da? Ethio- pia? Morocco? Most comparable country? (frequency count) Ug (22) Eth (12) Mor (6) Uganda

Regime type (Source: Economist Democracy Index 2010) Country size (surface area, thousand km2) Human Development Index Ranking Population (m) Population growth (%)

Semi- Democrat ic (Hybrid regime) 196.7

Semi- Democrati c (Hybrid regime) 799.4

Semi- Authori Democratic tarian (Hybrid regime) 241.038 1,104.3

Authoritaria n

446.6

Uganda

Low (Rank 155) 12.5 2.6

Low (Rank 184) 22.89 2.3

Low (Rank Low 161) (Rank 174) 33.4 82.8 3.2 2.6

Medium (Rank 130) 32 1.2

Uganda Ethiopia Morocco Ethiopia

122

GDP (current 12.8 US$) (billions) GDP per capita 1,020 (current US$) GDP growth 2.2 (annual %) Income share 6.2 held by (2005 lowest 20% estimate) Births attended by skilled health staff (% of total births) Life exp. at birth (years) Mortality rate, under-5 (per 1,000) Literacy rate, youth female (% of females ages 15-24) Primary completion rate, total (% of relevant age group) Mobile cellular subscriptions (per 100 people) Agricultural land (% of land area) 52 (2005 estimate) 59 93

9.79 428

17.01 509

32 386

91.37 2,771

Uganda Uganda Ethiopia Uganda Uganda Morocco

6.4 5.2 (2008 estimate) 55 (2008 estimate)

5.6 5.8 (2009 estimate) 42 (2006 estimate)

8.8 9.3 (2005 estimat e) 6 (2005 estimat e) 59 104

4.9 6.5 (2006 estimate) -

Uganda

50 142

53 9

72 38

Uganda Ethiopia Uganda Ethiopia Uganda

63.7 56 57 57

6.5 (2009 estimate) 85

33.3

72

55

80

Ethiopia

55

26

38

79

Uganda

47.5 (2008 estimate)

62.0

70

34.5 (2008 estimat e) 51

67.2 (2008 estimate) 16

Uganda Ethiopia

Agriculture, value added 17 (% of GDP) Industry, value added 22 (% of GDP) Services, etc., 62 value added

31

24

29 24 45 25 50 11 39 55

Uganda Morocco Uganda

Uganda Ethiopia

123

(% of GDP) Exports of goods and services (% of GDP) Imports of goods and services (% of GDP) Coastline length area at risk from SLR % of population within 100km of the coast (Data source: Earthtrends, WRI) Total Number of HMDs (1960-2010 period; source:CRED, EM-DAT) % change in average number of HMDs (1960- 1994 vs. 1995-2010) Effective decentralizati on policy? Y/N Total annual influx of climate finance (m USD) Net official development assistance and official aid received (current US$)

29 24 25 24 11 39 44 700 km 44 2,470 km 34 0km (landlocked ) 0 29 0km (landloc ked) 0 1,835 km

Uganda

Uganda Ethiopia Morocco Morocco

83%

59%

65%

Morocco

26

58

29

63

36

+0.2

+1.1

+7.7 +1.3 (exponentia l rise in HMDs)

+0.8

Ethiopia (closer to Moz) Uganda (closer to Senegal) Ethiopia (closer to Moz)

1,018

2,013

2,934

3,820

912

Uganda Ethiopia

124

(millions) Total public spending on education (% of government expenditure) Ease of access to key informants / conducting research? Y/N 19 (2008 estimate) 21 (2006 estimate) 20.3 23.3 (2007 estimat e) 25.7 (2008 estimate) Uganda Ethiopia

Uganda

Table 4.5: Country comparability matrix: Side-by-Side country comparison - Comparing oranges & oranges (World Development Indicators, 2011 estimates unless otherwise noted)

4.3.1 Resulting Country case selection: Thus we arrived at our three principal country cases of Senegal (representing the Unprepared firefighter policy group), Uganda (to infer for the Prepared Firefighter group) and Mozambique (representing the Disaster averters). On this basis, we could proceed with our cross-country comparison of triggers of institutional adaptation to meet the new challenge of CCA. Thus, in each country we asked: what were the impacts of climate related shocks in 1995-2010 and how were they governed? How well did the national policy in place address local adaptation needs? What factors prompted national governments to implement the disaster management policy and institutional framework that they adopted? 4.4 Phase 2: Zooming In At Community-Level To Elicit Adequacy Of National- Level Climate DM Policies This research involved large amounts of fieldwork at both the national and community level in order to identify priority adaptation needs. In Mozambique, we 125

conducted fieldwork between March May 2011, including three weeks at the community-level spent surveying in the seven community cases along the Mozambican 2,500km long coastline, between April 9th May 1st 2011. In Senegal, fieldwork spanned the entire two years of this research, beginning in September 2010 and only ending in October 2011, with national interviews and community data collection spread throughout this period. The first set of VCAs took place in Senegal in October 2010 (concerning three villages in the Northern region of Saint-Louis), whereas the second round concerning three villages in the eastern central part of the country (Kaffrine region) only took place in June 2011. In Uganda, the last field site, fieldwork was clustered in the month of October 2011, including 2 weeks of community data collection in Bulambuli district, east central Uganda; there data collection proceeded much faster thanks to a larger team of VCA researchers. 4.4.1 Community cases selection protocol In order to fully understand the impacts of disasters on communities, I also carried out extended fieldwork in severely impacted communities within my principal country cases. Random selection was used to select amongst different climate disaster hotspots in each countrycommunities most severely impacted by HMDs in each country case. Within the selected communities, a significant sample of the population was surveyed (an average of 30% of all households in each target community), using a community-based Vulnerability and Capacity Assessment (VCA) methodology toolbox that utilizes both a standard questionnaire gathering data about my key variables under study, as well extended focus group interviews with male and female members of the community. 126

In each selected community, households were in turn selected for the survey using random sampling (every 5th household on my way, walking). Seven vulnerable communities at risk from rising HMDs were thus surveyed in Mozambique working with a research team of 5 people; six communities in Senegal working with a research team composed of 24 Red Cross volunteers; and 5 in Uganda working with a research team composed of 24 Red Cross volunteers. The VCA methodology The tools we employed to generate our community vulnerability data are derived from a participatory methods toolbox for assessing communities vulnerability and capacities to climate hazards, devised by the Red Cross/Red Crescent movement for use by its community-level volunteers for community work programming: the Vulnerability and Capacity Assessment (VCA) tools (IFRC 2009). The Vulnerability and Capacity Assessment (VCA) toolbox uses a range of different tools, which like the hammer, scaffold and measuring meter that one uses to construct a house, are the aides we have used to access data across different segments of the community studied and build a complete picture of the communitys Vulnerabilities to and Capacities to address climate changes. All the data generated across groups using the VCA tools were then triangulated to draw to a complete picture as close to the local reality as possible. The main attractions of this methodology are its ability to access the most vulnerable groups (women, youth, elders) and give them a voice within the process of collective solution design, as well as its emphasis on local capacities to address local vulnerabilities, before recourse to external interventions is considered. Additionally is the belief at the center of the VCA methodology that vulnerability is the product of multiple stressors, historically derived from the socio-economic 127

characteristics of the community; climate variability under this understanding is thus solely one factor of vulnerability, out of many others. The output is data on climate vulnerability in the context of the communitys wider vulnerability. The VCA serves first and foremost as a planning tool, at the very beginning of the project cycle, a useful method to identify/understand the problem before devising solutions/policy interventions, and to involve communities from the very outset: in the design of the projects activities, for their future ownership of these (see fig. 1).

One final note to make is that throughout our assessments, we have paid a lot of attention to gender-specificity. Indeed, generating gender-disaggregated data was an output of our assessments. Generating data on all vulnerable segments in the community was an important endeavor in order to place the gender-specific data in context, as well as set the entire community on a sustainable track of identifying their own vulnerabilities to climate risks, articulating by themselves who the most vulnerable groups are and developing a communal strategy to address these vulnerabilities. The rationale is that transforming womens vulnerabilities into opportunities will require action by both men and women, and if men are moved to become the champions of womens vulnerabilities, they will be their first allies in working to demolish them. 128

 More on the specific Vulnerability and Capacities Assessment (VCA) tools we have used for our community assessments are in table 1 below. All the activities detailed in table below were conducted within focus groups (men, women, elders and youth as relevant), save for the final community data validation meeting that brought the entire community together. Highlighted in Yellow in table 4.6 are exercises specifically designed to access gender-specific data. In Red are exercises specially aimed at collecting information on Climate Variability & Change.

129

Table 4.6: Vulnerability and Capacity Assessment (VCA) Tools employed to assess vulnerability in each target community (climate-vulnerable communities in Senegal, Uganda and Mozambique) Exercise Target sub- Purpose of Exercise (references in VCA manual) segment to Head This exercise will serve to generate data on how DAY TWO: Household Data Household participate ien) & households and their assets are vulnerable to climate risks. sampling (Women M n Household & Village exercise members This exercise also makes apparent inter vs. intra- Vulnerability Assessment Community introduction (surveyed Village authorities Introduction of Research aeam ato rof purpose of research household differences in tccess nd esources. (p. 107 in Manual) separately by (Mix, all) Refer to Household Data 3 gender) Transect walk guides: 1 Research team to walk through community scoping and Questionnaire & Collection Woman, 1 Man, 1 taking note on its most salient features of vulnerability and (pp. 86-92 in Manual) Sheets Youth capacities, accompanied by 3 guides (one woman, man and youth). DAY TWO (Household Data Household Men How do women specifically cope with to the surface when Community Mapping: Women Head Mapping exercises will serve to bring climate risks men, sampling contd) 1) Spatial Hazard/Risk (Women Men) & these occur? Which perspectives institutional and social Youth women, and youths particular on their community, as Livelihoods coping members mechanisms are triggered in response? What are be map and they see it. Results from these mapping exercises will the strategies analysis pathways of cause and effect? 2) Vulnerability maps (surveyed confronted across focus groups, towards a more complete (see pp. 109-119 nd people) separately by picture of vulnerabilities and co the heart ohe nderstanding This specific exercise gets us t apacities in t f u community. (places a in Manual) Refer to Household Data gender) the ways in which different members of the household are 3) Capacities map Questionnaire & Collection specifically affected by climate related disasters, which (pp. 75-86 in Manual) Sheets assets they sacrifice in the cope, and w of womens Seasonal calendar: past & Women Men Exercise to reveal order to seasonality hich safety nets they resort to i relative to present trends Youth vulnerabilities n order cope. those of men- during which Final Community Restitution Whole The final community meeting at the at their the study (p. 92-97 in Manual) times of the year are climate impacts felt end of peak? meeting: community reveals sense o vulnerabilities, capacities and gaps that Historical profile Elders & Youth To get a all the f the communitys history, cultural identity - V Environmental Hazards emerged by vulnerable sub-segment n t ut community and & alidation of Vulnerabilities, and environmental profile. Essential tio phe data in context. Capacities, Gaps (summary of (at least 50 In this exercise, we ask the group: front of the entire brings them to the surface in 1) General historical profile data collected), with breakout community community, the community; 2) apacity aps & relevant (pp. 98-105 in Manual) markers of for final validation of CHistory Gof hazards in session with Women members: vulnerability-reducing/capacity-building strategies in community; 3) How have you coped with hazards by the - Identification of relevant Men - Women past? community. resilience-building Community Organizations Youth E Men At the the focus group for this exercise, men and women Women lders- During end of this meeting, community members interventions Social network Youth identify/validate relevant rdifferent groups iand list the Mapping / are separated into two esilience-building nterventions - Prioritization of and prioritize these using the Ranking matrix (refer to data Analysis (Venn diagram) + institutions and social networks that frame their lives in interventions: capacity of key Ranking collection sheet 7). Following this initial scoping, the Assessing the their community. matrix (see rganizations Data collection Refreshments/light main to be organized for this meeting. community O capacity of the food local organizations and social sheet pp. 119-125 and pp. 7) (see networks identified is assessed (see data collection sheet In tin order to gauge nheir aomen ao anot speaking educe the 128-132 in Manual) 6), he event hat you t ote w bility t re ddress and r up. Break the community meeting into two focus groups to give space Refer to data collection Sheet vulnerabilities of women producers. for women to express themselves 6 A comparison of the different diagrams drawn by men and women will show differences in the way men and women perceive the patterns of relationships within the community. Direct observation Men Women This activity will be most critical to the understanding and (p. 71 in manual) Youth - Elders generation of reliable data on the particular climate- related vulnerabilities that people face at the community- level. The purpose of this exercise is to soak in the community and bear eyewitness to the vulnerabilities/capacities highlighted by the women and other sub-communities, following the anthropological participant-observant method. All observations from this exercise will be duly noted every day and will serve as qualitative input to contextualize the quantitative gender-disaggregated community-level data generated. DAY TWO: Household Data Household Head This exercise will serve to generate data on how sampling (Women Men) & households and their assets are vulnerable to climate risks. Household & Village members This exercise also makes apparent inter vs. intra- Vulnerability Assessment (surveyed household differences in access to resources. (p. 107 in Manual) separately by 130 Refer to Household Data gender) Questionnaire & Collection Sheets

 Community CCA-DRR needs appraisal: Then in each community we asked: what are local needs for adaptation to rising HMDs, in excess of local capacity to cope with their impacts? How relevant and adequate are local and national plans/policies in place to meet these local needs for adaptation? Based on answers to these questions, we reached a final assessment of each community CCA-DRR needs, and were able to proceed with our appraisal of the adequacy of current national plans and programs aiming to manage climate related risks, for the basic CCA-DRR needs identified by the communities weve surveyed. 4.5 Phase 3: Incentives analysis Following part I which enabled our selection of country cases by policy type, and part II that offered us the opportunity for an in-depth analysis of the adequacy of national policies to meet community needs, we arrived at a characterization of the national policy in each of our country case studies. In each of these countries, what factors prompt governments to adopt the climate disaster management policies and institutions they did? This question is the crux of the final part of our research plan. Part 3 of our analysis explores the factors prompting/deterring policy change in the context of Africas governance and delves into an incentives analysis of why a given policy in each country came to be, and what factors prompted / obstacles impeded the adoption of new socially beneficial institutions and policies to address the new challenge of CC, in the context of Africas governance incentives. We relied heavily on expert interviews using the snowballing interviewing method in each country case. We thus interviewed iteratively a total of nine informants, various experts on CCA-DRR, in Mozambique; seven in Senegal and five in Uganda. 131

Each interview lasted an average of an hour and half. The questionnaire protocol included in Annex I was dispensed in person to each in-country informant, yielding a precious goldmine of qualitative data on what prompts governments to adopt a given CCA-DRR policy, and deliver adequate disaster relief/prevention, posited here as a public good. Oft times, an expert was interviewed multiple times, and called on to provide additional insights as the research developed. In addition to expert interviews, an extensive events analysis was conducted in each country case, based on the review of newspaper clips from prominent national newspapers in order to enable an adequate process tracing of public response to HMDs between 1995-2010. In Mozambique, our events analysis is based on the review of 78 newspaper clips that matched our keyword search, from the archives of two of the countrys major newspapers (Notcias and J. do Domingo), catalogued in chronological order from 1995 to 2010. In Senegal, we consulted 95 newspaper articles drawn from three of the main national newspapers: Le Soleil, Populaire and Sud Quotidien. Finally one hundred and fifty (150) newspaper clips underpinned our events analysis in Uganda, drawn mainly from the Daily Motion and New Vision. This overall wealth of data formed the basis of our in-country analysis, and enabled us to conduct our process tracing of the circumstances, factors and variables that led to institutional change in CCA-DRR in each of the countries where such a change occurred. We provide more specific details on these in the individual country chapters.

132

PART III:

RESULTS & COUNTRY FINDINGS

133

 CHAPTER 5:

CLIMATE-RELATED DISASTER MANAGEMENT POLICIES ACROSS AFRICA 5.1 Results of Africa-wide classification Our country classification by climate-related disaster management policy type reveals that in Africa: thirty-three (33) countries are Unprepared Firefighters, nine (9) are Prepared firefighters and only six (6) countries qualify as Disaster Averters. A total of forty-eight (48) African countries were assessed and classified, and seven (7) countries remained unclassified because we could not obtain credible data, either because they were war-torn territories or countries just recovering from a crisis. Table 5.1, which presents each countrys point-based score on each of our twenty indicators for national disaster policy, provides further details on how each African country is preparing for climate-related disasters, and which countries are doing more than others, have more in place by way of disaster risk management frameworks and institutional arrangements, our initial research question.
COUNTRY RANK NAME 1 2 3 4 5 6 7 8 9 10 11 Mozambique Algeria Madagascar South Africa Mauritius Cape Verde Morocco Sudan Lesotho Botswana Tanzania OVERALL DM POLICY SCORE 17 16 15 14 13 12 10 9 9 8 8 FINAL SCORE (WEIGHED) 17 16 15 14 13 12 10 9 9 8 8 POLICY TYPE 3 3 3 3 3 3 2 2 2 2 2

POLICY TYPE NAME Disaster Averters Disaster Averters Disaster Averters Disaster Averters Disaster Averters Disaster Averters Prepared Firefighters Prepared Firefighters Prepared firefighters Prepared Firefighters Prepared firefighters

134

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Uganda Namibia Tunisia Ethiopia Egypt Djibouti Mauritania Comores Swaziland Sierra Leone Senegal Malawi Liberia Gambia Zambia Rwanda Nigeria Mali Ghana Cte d'Ivoire Cameroon Burundi Angola Togo Kenya Guinea DR Congo (Kinshasa) Burkina Faso Benin Zimbabwe

8 7 6 6 8 7 5 5 4 4 4 4 4 4 3 3 3 3 3 3 3 3 3 2 2 2 2 2 2 1

8 7 6 6 4

2 2 2 2 1 1

Prepared Firefighters Prepared firefighters Prepared firefighters Prepared firefighters Unprepared Firefighters (weighted)* Unprepared Firefighters (weighted)* Unprepared Firefighters (weighted)* Unprepared Firefighters (weighted)* Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters

4 1 4 1 4 4 4 4 4 4 4 3 3 3 3 3 3 3 3 3 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters

135

42 43 44 45 46 47 48 0 0 0 0 0 0 0

Somaliland Somalia Niger Gabon Chad Congo-Brazza Central African Republic Equatorial Guinea Eritrea Guinea-Bissau Libya

1 1 1 1 1 0 0

1 1 1 1 1 0 0

1 1 1 1 1 1 1

Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters Unprepared Firefighters

- - - - Insufficient data Insufficient data Insufficient data Insufficient data Insufficient data Insufficient data Insufficient data

Sahrawi Arab Democratic Republic - So Tom and Prncipe Seychelles - -

Table 5.1: Final Africa-wide country ranking by Disaster Management policy score (weighted; from most proactive Disaster averters to least prepared firefighters in Africa)
Note: Even though these countries have an overall policy score of 5 or above, they do not satisfy key indicators 5-6 (financial and human capacity to conduct DM duties); thus their weighted score classifies them as Unprepared firefighters.

Disaster Averters 11% (6 countries)

Unclassied 13% (7 countries)

Unprepared Fireghters 62% (33 countries) Prepared Fireghters: Eecgve Disaster Responders 14% (9 countries)

Fig. 5.1: Distribution of African countries by Disaster Management Policy Type

136

Fig. 5.2: The Patchwork of Disaster Management policies in Africa, basis of our country case selection

5.2 Discussion of Cross-continental Results The overwhelming red hue of fig. 5.2 confirms previous opinions about disaster management in Africa: the large majority of countries are at the very lowest rungs on the spectrum towards effective disaster management. However, what our classification lays bare, and table 3 in particular, is that great variation exists nonetheless within all these countries; for instance the Central African Republic and Congorespectively ranked 47 and 48 on the Africa-wide classification of countries, with an overall policy score of 0 countries that do not have as much as a standing Disaster Management body, can not be assimilated to Mauritania, Swaziland or the Comores, respectively ranked 17, 18 and 19th in our classification. Though all Unprepared Firefighters, the latter three countries are critically more advanced in the establishment of institutional arrangements to address climate-related 137

disasters, even if these are still poorly funded and staffed. Albeit being all at the very lowest levels of progress towards implementing recommended HFA recommendations for disaster-resilient societies, the latter countries lie at very different levels in terms of their national commitments and efforts already endeavored towards DRR. Thus it would be a mistake to group them indiscriminately under one category, and apply the same standardized policy prescriptions to them all. As fig. 5.1 further demonstrates, African countries lay at different levels on the spectrum of effective Disaster Risk management, and great variation and heterogeneity exist from country to country. Our results demonstrate this heterogeneity and reveal where each country stands on 20 standard criteria revealing progress towards effective Disaster risk Management (DRM) and implementation of the goals of the Hyogo Framework for Action (HFA). Differing baselines ought to now inform differentiated DRR-promotion measures in Africa. Differential progress towards achieving HFA goals across the continent begs for custom-tailored programs designed to support African countries on the critical areas/criterions where their progress is still slow (see table 5.1). 5.3 Limitations 5.3.1 Classification, work in constant progress: The first limitation of our Africa-wide classification is its coverage only up until September 2010, date on which research and classification were completed, often using data dating back as far as 2004 as information sources. Thus our classification runs the risk of being outdated, every time a country adopts a new DRM legislation or revamps its disaster response, passing from delayed to effective disaster response. In this sense, classification is a work in constant motion, and our classification is no exception: it only provides an adequate depiction of climate-

138

related disaster management policies across Africa only up until the point we stopped classifying. In the short time window between the end of our classification and submission of this manuscript alone, one new country was born in Africa, Southern Sudan, driving the number of countries to be assessed on the continent to 56, and one additional country adopted a new legislation on DRR, Tanzania, giving the country technically one more point in our point-based ranking, making it sit more comfortably among the Prepared Firefighters. Assessment being intrinsically a revolving exercise, our classification contains this major time coverage limitation. Nonetheless, our classification provides an adequate picture of the overall distribution and heterogeneity of policies to address climate-related disasters in Africa, one that is unlikely to change drastically over the next five years. Furthermore it offers an innovative qualitative method to assess progress towards DRR at the national and sub-national levels, one that can be built upon to integrate newer developments in each country, as well as extended to other countries / regions not assessed, beyond the scope of this paper, enabling cross-country and cross-regional comparisons. Indeed, the sequential spectrum of DRM put forth in this paper ensures that countries with the slowest progress towards achieving HFA goals and effectively managing disaster risk (many of Africas countries for instance) are not left out in classification, homogenized under the label of ineffective DRR which current binary DRR paradigms label them as (see fig. 3 for an example). Our three-partite continuum of Disaster Risk Management effectiveness ranging from the most unprepared of firefighters (with a score of 0) to the most effective Disaster Averter (score of 20) offers a means to capture the full variation and patchwork of policies that one sees at the national level and demonstrates comparatively which countries have very little in place by way of Disaster Risk Management and which others have much more in place. This constitutes a much more cogent way of analyzing developing countries national disaster management policies in our sense, 139

enabling cross-country and cross-regional comparisons, and allowing for policy recommendation and donor-supported strategies that are better tailored to the nuance and texture of national realities in disaster management policy 5.3.2 Note on confidence levels, data limitations and shortages: ID Out of the total thirty-five countries assessed, 39 Country name Togo thirty-one were classified with high confidence (4 or more data credible sources were used to generate final country classification); twelve were with Medium confidence (3 credible data sources used); five with Low confidence (1-2 credible data sources used) and seven countries were unclassified because no credible data sources could be accessed from country (country at war, in crisis or otherwise inaccessible). Table 5.2 summarizes confidence levels for each country classified.
Table 5.2: Classification confidence levels for each African country classified Key: High confidence=4 or more credible data sources used as basis for classification; Medium= 2-3 data credible sources used; Low= 1-2 credible data sources used.
1 40 2 41 3 42 4 43 5 6 44 7 45 8 46 9 47 10 48 11 12 49 50 13 51 14 52 15 53 16 54 17 55 18 19 20 21 22 23 24 25 26 Algeria Tunisia Botswana Uganda Burkina Faso Zambia Burundi Zimbabwe Cape Verde Comores frican Republic Central A Cte d'Ivoire Chad Djibouti Egypt Ethiopia Mauritania Gabon Rwanda Gambia Ghana Equatorial Guinea Eritrea Kenya Guinea-Bissau Madagascar Libya Malawi Western S Mauritius ahrawi A.D. Rep. So Tom Morocco and Prncipe Seychelles Namibia Nigeria Senegal Sierra Leone Somalia Somaliland South Africa Sudan Swaziland Mozambique Niger Benin Cameroon Angola Congo-Brazza DR Congo (Congo-Kinshasa) Guinea Lesotho Liberia Mali Tanzania

Classification Medium Confidence level High Medium High Medium High Medium High Medium High Low High High Low High Low

High Low High Low High No Data High No D High ata No D High ata No D High ata No D High ata No D High ata No D High ata High High High High High High High High High High High High Medium Medium Medium Medium Medium Medium Medium Medium

Though relatively numerous, the countries classified with high-to-medium confidence (a total of 43) are the only African countries we can venture to make inferences about; the remaining 12 need further research and data mining on to be able to yield conclusive insights about their national disaster management policies.

27 28 29 30 31 32 33 34 35 36 37 38

140

A further limitation in the interpretation of our results is the reliance on HFA mid- term national progress reports and VFL national reports to generate insights for indicators 11-20 (indicators of DRR measures); for lack of other data sources, we resorted to the HFA (self-assessments) and VFL (bottom-up assessments). This unfortunately replicated the methodological flaws decried about these two reports, namely that they reflect the reporting biases of the government officials who responded to the self-assessment reports on the one hand (HFA) and those of national organizations delegated to conduct survey work on the other (VFL). In addition, bottom-up perspectives were only gathered for nine countries evaluated in Africa (VFL). We attempted to circumvent duplicating these two reports flaws as much as possible however by confronting them against each other to generate a balanced account of countries progress towards implementing DRR measures and triangulating their respective data, as well as supplementing them with as many other qualitative assessments and informants views as possible. Data access remains a major challenge for studies about Africa, and the qualitative assessment we have here presented shows a way forward to generate new meaningful insights and methodological frameworks for the continent by accessing key informants (eg: Red Cross disaster managers, civil society representatives following closely the evolution of disaster management policies in their countries, etc.), and confronting their different accounts of the national public policy in place in their countries with public documents and declarations. Finally, a note on our chosen indicators and their respective weighting in the classification process of African countries is called for. It is important to recall that our twenty indicators were chosen in accordance with our storyline of what constitutes effective disaster management in Africa and worldwide, a storyline compatible with the major DRR frameworks and initiatives currently subscribed to by African countries, most notably the Hyogo Framework for Action, and the World Meteorological Organization recommendations. When analyzing African countries, converting this horizontal framework to the form of a continuum was important 141

however to be able to capture the variation existing among Africas countries, often all uniformly classified as low progress countries towards implementing Hyogo recommendations, offering no way to rate African countries against each other. By defining a spectrum with a baseline as not ineffective DRR, but rather the lowest progressing African countries so that it could be accounted for in our appraisal (countries whose score is 0), we enable a cross-country comparison of Africas disaster management policies. One needs to be cautious however when interpreting our results to not mistake Mozambique (with a score of 17) as a perfect instance of effective DRR. Indeed, Mozambique is a high-performer, compared to Equatorial Guinea for instance. However Mozambique still pales in comparison to other countries not on the continent that may have attained a perfect score of 20 on our spectrum of effective DRM, and have higher levels of DRM achievement. Based on this rigorous cross-continental identification of disaster management policies in Africa, and country classification by national disaster management type, we selected three country cases to underpin our institutional comparative analysis: Senegal, to represent the Unprepared firefighter policy group, Uganda to infer for the Prepared firefighter country group, and finally Mozambique to better understand the characteristics of the accomplished Disaster averters policy group.

Fig. 5.3: Selected country cases where fieldwork was conducted between September 2010 October 2011

142

We now turn to the findings from these three country cases to understand what factors spurred them to adopt the policies and institutions that they did to address rising climate risks in their countries.

143

 CHAPTER 6 SENEGAL COUNTRY CHAPTER: What thwarts Institutional Change for effective CCA-DRR in Senegal? Senegal was selected to represent the group of Unprepared firefighters, ineffective responders to hydro-meteorological disasters. Does the country in fact merit this classification? If so, what systemic factors render the country an Unprepared Firefighter, and thwart its engagement in the path of resolute institutional change to tackle the rising impacts of a changing climate? The present chapter elicits answers to these questions. 6.1 Country Profile As mentioned earlier, the Constitution of Senegal sanctifies the States primal responsibility to ensure security, well-being and integrity of all of its citizens in peace time and in time of crisis. The Law n64 -53 of 10 July 1964, promulgated a mere four years after the countrys Independence, further confirms our interpretation. Enacting the general organization of civil protection in Senegal, the 1964 Law states that: At all times the State must look after the protection of goods and physical persons and resources of the country174. Both these texts make it a constitutional responsibility of the State to provide relief to victims in times of national disasters, or otherwise pro-actively address disasters, and posit disaster management as a public good to be delivered to all citizens, regardless of their sex, origin or birth place.
174 Constitution of Senegal, 2001

144

Yet in practice, it is noticeable that the Government of Senegal (GoS) has shirked this responsibility. Over the past 30 years (1980 to 2009), flooding, the main climate hazard, has affected over 900,000 people in Senegal, caused the death of 45 people and caused damage estimated at over USD 142 million. In the suburbs of Dakar most particularly, flooding has become almost an omnipresent reality since 2005, directly affecting the poor175. During these times however, when thousands of people are impacted both in rural and urban areas, live in flooded homes knee-deep in green stagnant waters, and recourse from the State is needed most, it has been noticed between 1995-2010 that the GoS has not been, by and large, the guarantor of rights to security, protection and physical integrity. On the contrary, during these times of crises when the State is most needed are when the State has been least visible on the ground. Instead, donors and NGOs took the lead in relief operations, and the State found scapegoats to attempt to shirk its constitutional responsibility in times of crises, until this responsibility could no longer be evaded in 2009. This chapter offers a tale of how politics, popular pressure and public goods control are intertwined in time of climate crisis to force the State to deliver on its responsibilities. This tale offers mostly insights on what systemic factors explain the inability of the State of Senegal, typical in many ways of many in sub-Saharan Africa, to handle complex climate crisis situations, and what enabling factors can contribute to undo the situation. What conditions the governments inability to handle climate- related shocks? Is it merely lack of resources as is often purported, or does it go beyond? What are the systemic obstacles thwarting the efficient handling of climate shocks, and provision of needed relief to vulnerable groups during climate crises? These are the questions we address in this chapter. We aim to find what plans are in place to address HMDs, before turning to how the GoS has handled each climate shock between 1995-2010. We rest our analysis on two main case studies: floods in the cities of Dakar and Saint Louis where the GoS has clustered its few interventions 175 World Bank/Government of Senegal, Rapport dEvaluation des Besoins Post-Catastrophe Inondations
Urbaines Dakar 2009 (Dakar: World Bank, 2010)

145

in response to climate-related shocks. Following our description and analysis of the States flood response interventions, we will attempt to understand the systemic reasons for the Senegalese governments inability to handle climate-related shocks. What can be done to unlock government ability to properly take account of climate risks before they turn into full-blown crises? First, however, we delve into Senegals specific vulnerability to climate shocks; and why climate crises became a rising issue on the national agenda in the past decade. 6.1.1 Vulnerability Context The westernmost country in Africa, situated on the Western tip of the continent between 12.5 and 16.5 north latitude, Senegal is bordered by the Islamic Republic of Mauritania in the North, by Mali to the East, by Guinea Bissau and the Republic of Guinea to the South, and by the North Atlantic Ocean to the West. The country covers a total area of 196,712 km2, with the Gambia forming an enclave within Senegal on the lower river of the same name.

Figure 1: Senegal and its regions. Source: Official website of the Government of Senegal176.

176 Saint-Louis region was halved in 2002, to create Matam, Senegals youngest region as of 2002. Other

reconfigurations followed, bringing the total number of regions to 14 as of 2012.

146

 Development challenges: past & present A country of 12.8 million inhabitants classified as lower middle income, Senegal has a GDP/capita of a little over 1,000 USD and an annual GDP growth rate of 4,1%177. Formerly one of the prosperous countries in West Africa between 1990 and 2005, Senegal has been experiencing an economic downturn since 2006178. Senegal remains a poor country with more than half of its population (54%) living below the poverty line. It lies in the lower rungs of the human development ladder, classified 155th out of 187 countries ranked by the United Nations Development Program Human Development Index ranking179. High unemployment (48% of the labor force), most acute among the countrys predominantly young population (58% of population is below 20 years of age) continues to prompt illegal migrants to flee Senegal in search of better job opportunities in Europe. Poverty is most predominant in the rural areas, where 65% of the countrys poor live, while making up less than 55% of the total population. Country size (km2) : Population (million): Population growth (%): GDP (billions, current US$): GDP per capita (current US$): GDP growth (annual %): Income share held by lowest 20%: (2005 estimate) Births attended by skilled health staff (% of total) (2005 estimate) Life expectancy at birth, total (years) Mortality rate, under-5 (per 1,000) HIV prevalence rate (2009 estimate) Literacy rate, youth female (% of females ages 15-24) Primary completion rate, total (% of relevant age group) 196,700 12.8 2.6 12.9 1,040 4.2 6.2 52 59 75 1 56 57

177 National Statistics and Demography Agency Senegal, accessed March 25, 2011. http://www.ansd.sn/ 178 World Bank, Senegal Country Brief, accessed December 01, 2011. http://go.worldbank.org/PO6JPCB5P0 179 The United Nations Development Program Human Development Index (HDI) is a composite index measuring

life expectancy at birth, adult literacy rate, poverty, gender inequality, sustainability (impacts of natural disasters and carbon emissions per capita) and gross domestic product per capita, to provide a measure of countries human development levels.

147

Mobile cellular subscriptions (per 100 people) Agricultural land (% of land area) (2008 estimate) Agriculture, value added (% of GDP) Industry, value added (% of GDP) Services, etc., value added (% of GDP) Exports of goods and services (% of GDP) Imports of goods and services (% of GDP)

67 47.5 17 22 62 24 44

Table 6.1: Senegals Development at a Glance. Sources: World Bank World Development Indicators and Senegal National Statistics and Demography Agency (2010 estimates, unless otherwise noted)

A country with uneven development The Senegalese economy is based primarily on trade and services (for 62% of GDP), and to a limited extent on Industry (22% of GDP), while the primary sector (agriculture) dominated by groundnut production, cash crop introduced under French colonization contributes only 17% to GDP but employs the largest share of the population (77.5%). Following the devaluation of the CFA franc in January 1994, Senegal grew rapidly by 2.5% followed by a sustained annual GDP growth of over 5% between 1995 and 2001. In 2002, Senegal's economy was marked by a sharp contraction of business, with GDP growth dipping down to 1.1%. This was mainly due to difficulties faced in the agricultural sector. The country resumed with growth in 2003 with a rate of 6.3%, due to improvements in the agricultural sector. Dependence on petroleum imports for energy production and recurrent power cuts since 2005 constrain, however, economic performance and foreign investment. This also drives Senegal's trade balance in deficit, including foreign trade, in large part owing to the import of oil (petroleum), food and beverages (notably, rice, its staple food) and capital goods. The debt burden is also a major handicap, leading to reduced capacity for redistribution of state revenue and slowing the development of social infrastructure. The external public debt service as a percentage of state revenue, increased from 16.2% in 2000 to 17.6% in 2002. Since June 2000, Senegal has been 148

eligible for IMFs Highly Indebted Poor Countries initiative (HIPC) debt relief program, which wiped out two-thirds of its bilateral, multilateral and private sector debt180. However, with the energy sector heavily dominated by imports of petroleum products (oil bill doubled since 2000, nearing 861 million USD in 2005), the debt cancellation has had only a modest effect on social development. As for domestic energy consumption, Senegalese households mainly use wood and charcoal (56% of total energy use), to the detriment of timber resources that are becoming increasingly depleted by climatic and demographic pressures. Senegals deforestation rate is estimated to be of 45,000 hectares per year181. Senegals predominantly rural population more than three-quarters of the workforce are farmers is indeed heavily dependent on natural resources. The demographic pressure on these resources, correlated with a worsening climate, creates shortages and resource degradation, which exacerbate pauperization and render resource- dependent populations more vulnerable, especially in rural areas, where they face the negative impacts of a changing climate182. 6.1.2 Vulnerability To Climate Change and Variability 6.2.2.1 Physical vulnerability

The variety of bio-climatic conditions and diversity of geological substrata in Senegal, from the semi-desert of the Northern regions bordering the Sahara, to the lush green forests of Southern Casamance, including a 700-km long sea interface with the Atlantic Ocean, have endowed Senegal with several soil types with unequal ecologic usefulness from the dry, sandy soils of the Sahel to the Sudanian laterites

180 NAPA Senegal, 2006 182 NAPA Senegal, 2006 181 FAO, Statistic Yearbook 2009

149

of the Casamance. Overall, the soils are infertile and often fragile, susceptible to wind and water erosion183. Senegals climate is generally Sudano-Sahelian, characterized by the alternation of a dry season typically from November to May, and a rainy season from June to October. Average annual rainfall follows an increasing gradient from North to South, going from 300 mm in the semi-desert of the North to 1200 mm in the South, with strong inter-annual variations. Senegals climate is subject however to both geographic and atmospheric influences. Indeed, the presence of a coastline of 700 kilometers leads to climatic differences between coastal and interior regions. Also, atmospheric circulation, facilitated by the absence of mountainous obstacles mountainous, places the territory under the effects of maritime trade winds, the Harmattan blowing from the Sahara and the West African monsoon184. Three main rivers flow through the country: the Senegal, Gambia and Casamane, which all originate in the Fouta Jallon mounts of Guinea. Generally replete with water throughout the year, the rivers still suffer the effects of the dry season. The construction of major dams, Diama in particular, built to develop irrigation, animal husbandry and economic activities along River Senegal, and which Senegal shares with Mali and Mauritania, has impacted river flows and water availability. Finally, the country is highly vulnerable to Sea-Level Rise. Indeed with a 700- km long coastal area at risk, as well as 83% of Senegal's total population of 12.8 million and economic activities clustered within 100k of the coastline, even slight increases in sea level can have devastating effects on the country. Figure 6.2 depicts coastal areas at risk from a 1m rise in sea-level rise.
183 Ibidem 184 African Monsoon Multidisciplinary Analyses (AMMA), 2007

150


Figure 6.2: Vulnerability of Senegals coastline. Areas of heightened vulnerability to sea-level rise are hatched in green, areas of acute vulnerability in Red. Source: Niang-Diop, 1994.

As the following section demonstrates, Climate Change is projected to impact water availability and exposure to sea level rise even further. 6.2.2.2 Projected CC Impacts, Across Timescales and Sectors

Senegals National Adaptation Plan of Action offers the most up-to-date characterization of projected impacts of Climate Change and variability on the country (NAPA Senegal, 2006). Projected climate change impacts for Senegal consist mainly of: Table 6.2 further details each area of vulnerability. A rise in sea level, Decreasing rainfall, Increased variability in rainfall, Exacerbated evapo-transpiration, And a rise in temperatures (warming trend).

151

Fig. 6.3: Observed changes in Senegals average Temperature, 1950-2005. Source: NAPA, 2006

Fig 6.4: Water resources availability: Observed changes in Senegal River discharge levels as measured upstream (in Bakel) and downstream (in Kolda), 1960-2005. Source: NAPA, 2006. Sector at Risk Variable

Water resources & security

Access to clean drinking water remains major hardship, mostly for rural households. Only 56% of rural and 78% of urban households have access to clean drinking water Depleted groundwater table in many areas across country, mostly in arid

Agriculture & Food Security

Coastal Zones (Sea-level rise)

Sector Vulnerability

High dependance of 83% of population on rain-fed clustered within 100km of agriculture countrys 700-km long Limited irrigation coastline infrastructure Concentration of economic activities in coastal areas Fisheries, coastal infrastructure and tourist industry particularly sensitive to coastal risk 45th amongst countries most vulnerable to sea- level rise (SLR), out of 181 contries examined (Misdorp et al., 1990)

152

North and Center.

Observed changes

Consistent rise in temperature between 1950 2005, from 26.5C to 28.5C (fig. 6.3) Consistent decrease in river Senegal discharge levels between 1960 -2005 (fig. 6.4) Rise in average temperature by 2 to 4C between 2050 et 2100 Decrease in mean annual rainfall by 5 to 25%, with most dramatic drying projected in the South ; up to 58% decrease in mean annual rainfall by 2100 (Parton et al. 2004) -17% decrease in rainfall associated with a +2.5 increase in temperature Significant reduction in river flows and aquifer levels, by 5- 10m in the North and 15-20m in the South Significant reduction in water availability & security. Rational management of water resources Infrastructure for water retention (water harvesting, recharge areas of

Projection

35% decrease in rainfall quantity, reduction of rainy season length between 1950-65 and 1970-95 30% reduction in plant species Rapid loss in soil organic content Low regeneration capacity of forests Decrease in agricultural yields, leading to hightened food insecurity Loss of agricultural land Loss of cloud cover by 5 to 10%, with concommitent rise in evapo- transpiration

Fishing, one of countrys main exports, at risk from ocean temperature changes Currently observed rate of change : 1-2m annual withdrawal of shoreline along sandy beaches In the case of the Lagoba breach in the Sangomar Arrow (Saloum region), rate of degradation has been even more dramatic : 100-150m per year.

Proposed adaptation strategies

A minimum 21 to 48 cm rise in sea level by 2100 (note: global modeling of coastal zones project a 1m rise), Continued withdrawal of shoreline Destruction of seaside infrastructure / homes Exacerbation of Sea water inundation, coastal erosion, soil salinisation, mangrove loss, reduction in fish resources Main coastal zones to be affected: Saint-Louis, Bargny-Mbao sector, Small Coast between Ndangane and Djifer. Increase in swells (high waves) heights, with potential to lead to acceleration of rate of breaching of coastline arrows and further shoreline withdrawal Warming of ocean waters, associated changes in marine specie composition Changed structure of upwellings, impacts on fishing Urgent need to de- For 1m SLR: 1,350 km of link agriculture and coast will need to be rainfall, through: protected for a total cost of Irrigation; Storage 1,596 millions USD infrastructure for For 6m SLR: 7,450 km2 of rainwater/collection land area and approx. 3.7 Diffusion of agro- million people at risk, for a

153

aquifers) Protection of available reserves, through pollution controls, sewer system for houses, monitored extraction of groundwater

Medium. Low, no consistent Agreement over assessment of most likely increase in scenario of climate risk in temperature, Senegal decrease in rainfall and increased rainfall variability. However, envelope of uncertainty remains to be determined. Table 6.2: Climate Change impacts in Senegal: Observed and Projected. Sources: NAPA, 2006; UNDP, 2007 (unless otherwise noted).

Confidence level

forestry techniques Crop diversification Also : Short cycle/ Saline tolerant crops Forest fire controls Reasonable fertilizer use Functional early warning systems Training on CC Low, no consistent assessment of most likely scenario of climate risk in Senegal

total cost of protection of 3,623 millions USD (1.72% of current GDP). Halt to beach sand extraction and enforcement of law against this practice.

Photo 6.1: Fallen protective sea wall along the shoreline of Rufisque, Senegals Small coast. Rufisques coastal dwellers have already begun to suffer from advancing sea levels and sea water inundation into their homes. Source: Rufisque News.

6.2.2.3 Socio-Economic Vulnerability To Climate Variability and Change

Senegal is a country where climate variability on the seasonal-to-interannual timescale constitutes a practical problem with monumental humanitarian ramifications. With 77.5%, more than three-quarters, of the countrys active 154

population working in the agricultural sector185, a sector that is still highly climate- sensitive and dependent on rainfall, vulnerability of agricultural production and livelihoods to rainfall changes is endemic. Indeed only 2 percent of the total cultivated land in Senegal is irrigated or under some form of water management, the remaining 98% being exclusively rain-fed186. A growing majority of the Senegalese population also lives in unplanned shantytowns built on flood plains outside of the countrys major cities and coastal areas. These shanty towns were settled during the prolonged drought period of the early 1970s to the late 1980s (see Dai et al. 2004187 and Hulme 2001188), that drove peasants out of the countryside and into peri-urban settlements that were not designed to welcome them (no existing drainage systems, no planned infrastructure), in fact flood plains and water passages that had dried up during the prolonged drought. During the two decades that the drought lasted, waves of rural migrants settled these flood plains at the periphery of Dakar and Saint-Louis, obstructing natural water passageways that were soon to claim their right of way with the return of the rains in the mid-1990s. In these shantytowns, they are today exposed to the vagaries of changing rainfall regimes (see Pelling and Wisner 2009189; Diagne 2007190). With the return of heavier rains in the late 1990s and 2000s191, the vulnerability of these rural migrants to urban flooding has become a serious problem. Against such high vulnerability (high dependence on and exposure to climate factors) and limited capacity to cope with climate shocks (poverty and low adaptive
185 FAO, Statistical Yearbook 2009 186 Ibidem 187 Dai, A., P. J. Lamb, K. E. Trenberth, M. Hulme, P. D. Jones and P. Xie, The Recent Sahel drought is Real, Int. J.

Climatol. 24 (2004): 13231331

188 Hulme, M., Climatic perspectives on Sahelian desiccation: 19731998, Global Environmental Change 11 1 189 Pelling, Mark and Ben Wisner: Disaster Risk Reduction: cases from Urban Africa. London: Earthscan, 2009

(2001): 19-29.

190

Diagne, K., Governance and Natural Disasters: addressing floods in Saint Louis, Senegal, Environment and

Urbanization, 19 2 (2009): 552-562 191 Giannini et al., A Climate Model-Based Review of Drought In The Sahel: Desertication, The Re-Greening And Climate Change

155

capacity), when a climate hazard strikes in Senegal, a disaster of significant proportions often results, ones that further exacerbates the poverty of vulnerable populations in both rural areas and urban centers. The effects of climate-related disasters are most manifest when it comes to floods. Highly convective clouds more frequent since the mid-1990s passing over the country have led to flood disasters that have, in rural areas, destroyed harvests and deprived agriculture-dependent families of needed income, sending many into immigration; and in urban centers, compelled the urban poor living in the ill- planned shanty towns at the periphery Dakar and Saint-Louis to live in flooded homes, exposed to stagnant waters and the diseases they bear.

Fig. 6.5: The return of heavier rains in Senegal since the mid-1990s. Source: NAPA, 2006

The opposite also holds, with equally disastrous impacts: dry spells (consecutive days of rainfall deficit) have generated impacts that are as detrimental as those of wet spells the 1970s-80s drought in the Sahel serving as an eloquent illustration with its host of famines, rural exodus, cattle deaths, food price speculations and shortages it carted along192.

192 Though we make the distinction here that it is not the dry spells themselves that led to famines; rather it is

the other conditions of vulnerability inherent to the society in which they occur (e.g.: social inequalities in food distribution, poor protective mechanisms, propensity for food speculation, etc.) that, when compounded with rainfall deficits, generate famines. Sen, Amartya. Poverty as Deprivation, 1999.

156

In Senegal, and across the region, rainfall variability and changing climate patterns affect at once large sections of society dependent on climate factors for their livelihoods, food, health and well-being. 6.2 HMDs in Senegal and Government Response: 1995-2010 6.2.1 Main climate hazards in Senegal and impacts Major climate-related disasters and calamitous events have struck Senegal between 1995 and 2010. 1995 marks the return of heavier rainfall in the country (see fig. 6.5), and has been chosen as the inception point of our analysis in order to assess the adequacy of Governments responses to rising climate-related shocks. For the purposes of this analysis, the impacts of the preceding period of drought (1970- 1990) are not taken into account. Noteworthy climate shocks have included: River Senegal flooding in Saint-Louis in 1994, caused by above-normal rains and the rise of River Senegal out of its, leading to the inundation of its flood plain where the city of Saint-Louis is settled, and generating severe damages in the city, which was unprepared for them; Storm Cindy in 1999 in Casamance (South region) and Joal (Thies region), which caused the death of 99 fishermen, 10 disappearances, and considerable material losses (Civil Protection Department, 2007); Out of season rains in January 2002 in the regions of Saint-Louis and Louga that claimed 30 lives, caused significant material losses, including hundreds of collapsed homes, destruction of harvests and important cattle loss; Acridian (locust) invasions between 1998 and 2004 that destroyed hectares of cultivated land and caused significant damages to production in seven regions of the country; Recurrent flash floods that affect annually multiple locations across the country, with important impacts on populations and goods, and in some 157

instances human life loss. These floods are mostly an urban problem, caused by human settlement in depression areas that used to be riverbeds before the thirty-year drought, and which nature claimed back since the return of the rains in the mid-1990s. Flood impacts are unfortunately very poorly assessed. The best assessment of flash flood impacts to date is the World Bank-led Government report that quantifies in CFA francs and US dollars the impacts of the 2009 floods on the countrys economic sectors (see table 6.3).

Table 6.3: Estimate of losses during 2009 Floods in Senegal: Impacts of disaster nation-wide by sector (in CFA francs and USD). Source: World Bank/Government of Senegal, 2010

However even this report under-estimates the real consequences of the 2009 flood on the finances of local community groups. Flooding had a limited negative macroeconomic impact the report notes, the reduction in economic growth in 2009 resulting from the flooding is estimated at only 0.07 %, reducing the economic growth forecast to 1.4 % compared with 1.47 % before the floods. Damages represent only 0.4 % of 2008 GDP, and losses represent 0.3 %. In addition, the flooding did not change the fiscal position of central government because it had no 158

impact on government finance in 2009. However, the impact was felt more strongly by local community groups, whose losses represented over 10 percent of the 2009 provisional budget193 . The report goes on to explain: flooding had a considerable impact on households. Surveys conducted in these areas show that the affected households have lived in a more precarious situation after the flooding. During the 2008 floods, which were less extensive than those of 2009, 27% of flooded households in Pikine and Gudiawaye suffered food insecurity. In 2009, total revenue lost by the affected populations is estimated at 14% of average annual household revenue. The post- flood coping strategies adopted by households show that they had to temporarily or permanently forgo meeting some of their vital needs, for example by decreasing the number of daily meals (World Bank-GoS, 2010). The most critical impacts of the 2009 flash flood on vulnerable communities can thus only be gleaned, and are not accounted for in Table 6.3, leaving total damages much higher than 104 million USD. The hazards and calamities that occurred in Senegal between 1995-2010 have impacted all sectors of socio-economic human activity, mostly in urban centers, and have constituted threats that one can hypothesize jeopardized development gains. Table 2 summarizes our assessment of known HMD impacts in Senegal between 1995-2010, based on our events analysis.

193 World Bank/GoS, 2010

159


Photo 6.2: Distribution of flood zones in Dakar following the 2009 floods. Source: GeoEye satellite image, 14 October, 2009. Climate Shock Impacts 1994: River flooding in Saint-Louis No data Coastal erosion No data Impacts in Dakar, Rufisque, Mbour, Joal, Saint-Louis, Saloum Isld and Casamance 1999: Storm Cindy 99 fishermen dead at sea, 10 disappearances, and Impacts in Casamance and Joal considerable material losses January 2002: Out of season rains 30 dead, significant material losses, including hundreds of Regions of Saint-Louis and Louga collapsed houses, destruction of harvests and important impacted loss of cattle Acridian invasions between 1998 and Hectares of cultivated land destroyed by the 2004: 7 regions impacted grasshoppers, significant damages to production 2003 flooding: Saint-Louis impacted 50,300 people affected (DPC) 2005 flooding 200,000 people affected, more than 20,000 houses Saint-Louis, Joal, Kaolack, Fatick and collapsed or flooded (DPC) Dakar suburbs (districts of Pikine et 30 dead, 1,700 households relocated to Keur Massar and Gudiawaye) significant material losses 52 billion CFA Francs disbursed by central government to relocate flood victims to Keur Massar 2008 flash flooding Around 2,882 households impacted, or about 23,593 By mid-August most Dakar suburbs people stayed in their flooded homes and necessitated flooded emergency assistance (Senegal Red Cross/IFRC) by September 25 Some houses completely inaccessible, inhabitants contaminated by use of tap water for hygiene and from being in constant contact with stagnant waters, sewerage, solid waste and worms. Water-borne diseases explosion (cholera, malaria and

160

rash) (Views from the Frontline Report, 2009) 2009 flash flooding In Dakar, 7 dead by drowning, 814 families displaced, 500 Dakar suburbs, and other locations swallowed up by rain flood waters (Le Quotidien, 15 May across the country impacted 2009; ISE, 2009). Cost of 2009 flooding in Senegal is estimated at 104 million USD, including almost 56 million in damages and 47 million in losses. Sectors most affected by damages were housing (49 %), health (14 %), agriculture (11 %), education (10 %) and transport (8 %). Houses, schools, health centers and roads were severely damaged. Losses mainly concerned trade (20 %), public urban infrastructure (15 %), housing (16 %), energy (14 %) and transport (14 %) and these sectors account for almost 80 % of losses. The private sector was worst hit, bearing 65 % of all damages and 64 % of the losses. Location-wise, Peri-urban areas of Dakar were most affected. Cost of flooding estimated at $82 million therein. According to Government figures, about 360,000 people were directly affected in the peri-urban areas of Dakar, particularly in districts of Pikine and Gudiawaye, 125,000 were directly affected in the rest of the country. Those affected include most vulnerable groups of people living in Senegal (World Bank-GoS, 2010) 2010 flash flooding Dakar suburbs hit 3.000 families displaced (Le Soleil, July 26 2010) Table 6.4: Climate shocks in Senegal and their impacts, 1995-2010

As table 6.4 demonstrates, floods generate the most severe climate impacts in Senegal. Over the past 30 years (1980 to 2009 inclusive), flooding has affected over 900,000 people, caused the death of 45 people and caused damage estimated at over 142 million USD, according to World Bank estimates. In the suburbs of Dakar, flooding has become almost an omnipresent reality directly affecting the poor194. For the purposes of our study, we will circumscribe our analysis to the analysis of Governments response to recurrent flooding disasters, on which we rest our present analysis. We will thus look at the two types of floods that afflict Senegal: the first type occurring when the water level of a river rises beyond its bed and fills its floodplain (river flooding) and the second type concerning low elevation areas with a replete groundwater table that receive more rain than can be absorbed by the ground and the water table level consequently rising to above ground level (flash floods in
194 World Bank/GoS, 2010

161

Senegals peri-urban settlements). Both of these types of floods, although there is rural flooding in Senegal, are primarily urban problems arising from unplanned urbanization that has become problematic when conjugated with climate drivers. Humans settled in flood zones, and the spontaneous nature of their settlement has resulted in development without installation of public works such as closed sewer systems, rain drainage networks, paved roads, etc.195 . When the rains returned in the mid-1990s and found these high population densities of poor, uneducated Senegalese in flood plains on the peripheries of urban centers, the disasters that resulted were monumental. In the next section, we describe how the Government of Senegal (GoS) has organized itself to handle each type of flood, year after year. Beforehand, we diagnose the root causes of flooding in Senegal however, an explosive consequence of existing human- induced vulnerability combined with new rainfall increases. 6.2.2 Floods in Focus: Causes of Flooding In Senegal River flooding in Saint Louis: the Senegal River basin Written records of flooding along Senegalese rivers go back to French exploration logs of the territory. One such log kept during a 1658-1660 voyage recounts that during the rainy season the Senegal River rose so high that one-storey dwellings on the floodplains were completely washed away196. This type of record is not uncommon; the French catalogued weather conditions in the territory during their years of colonization. Most of their records concern conditions along the Senegal River near the city of Saint-Louis, an island at the mouth of River Senegal that served as the capital of Senegal and of the entire French West African colonial
195 Boudreau, Laura, Flood Management in Senegal: Past, Present, and Future, 2009 196 Dia, Aliou M., "Crues Et Inondations Dans Le Bassin Du Fleuve Sngal: Apport De La Tldtection Et Des SIG

La Cartographie Des Zones Affectes Dans l'Estuaire (Application Sur La Ville De Saint-Louis)", Diss., Universit de Marne la Valle: Institut Francilien des Gosciences, 2004

162

empire for many years. During the 19th and early 20th centuries, when Saint-Louis was very active as the French port of entry to the African continent, the French took extensive flood risk mitigation actions near the city197. All flood mitigation measures employed by the French, however, were forgotten when French control over Senegal ceased in 1960. Soon after Senegalese independence on April 4th 1960, the rains stopped and the era known as the 30-year drought began. The result of this succession of events was that the Senegalese government was not challenged by floods early on to develop policies or protocol to mitigate and cope with flooding198. Concentrated in the Saint-Louis region today through which River Senegal flows on its march to the sea in Northern Senegal, the problem of river flooding extends to the entire Senegal River Valley, occasionally affecting Richard Toll and other towns along River Senegal. It must also be noted that flooding in Senegal is not confined to imprudently urbanized zones, but is also common along rivers in rural areas throughout the country199. Because rural villages have a long experience in coping with yearly floods and often incorporate them into their crop rotations however, this type of flooding seldom reaches problematic proportions and is often handled using the affected communities own resources and capacities200. Flash floods in Dakars peri-urban settlements During the 30-year drought many of the conditions were created to turn Dakars peri-urban settlements into a ticking bomb, and render flooding a major issue today. Over the years the landscape changed; riverbeds shrunk, and areas that were once uninhabitable because of high water levels appeared to be open for settlement. These changes were accompanied by rapid urbanization: many of Senegals poorest
198 Ibidem 199 Ibidem 200 Ibidem 197 Boudreau, Laura, Flood Management in Senegal: Past, Present, and Future

163

rural residents fled to cities, a phenomenon known as rural exodus, looking for work because they could no longer grow their crops, keep livestock, or feed their families. Initially, it was primarily young Senegalese men who came, but full families eventually followed. Settlement of these poor migrants was spontaneous and done without thought of land ownership, a nonexistent concept in rural villages201. There was little to no government oversight or regulation of these settlements; they grew without urban planning or public works installations. This trend lasted two decades, resulting in high population densities of poor, uneducated Senegalese in floodable zones on the peripheries of urban centers202. When normal rainy seasons resumed in the mid-1990s, these areas began to suffer yearly flooding: their underlying vulnerability interacted with the new climatic driver, and the bomb exploded, rendering flooding an omnipresent problem defining the lives of the inhabitants of Dakars suburbs. Flooding in these imprudently urbanized zones203 occurs today in varying degrees in cities and large towns around Senegal. Flash flood problems are common in and/or around the following urban centers, ranked from most to least severely affected: Dakar, Thies, Kaolack, Joal, Tambacounda, and Ziguinchor (refer to map of Senegal in fig. 6.1). Around Dakar, flooding problems extend through the suburbs, along the corridor leading to the nearby city of Thies. How the Government handles these flood crises is the concern of the next section, and opens a door to inside Senegalese realpolitik.
201 Coly, Adrien, Inondations et Processus dUrbanisation, 202 Boudreau, Laura, Flood Management in Senegal: Past, Present, and Future 203 Diouf, Diouf, Catastrophes Naturelles Et Leurs Effets: Le Contexte Sngalais, Presentation by Ministry of the

Interior, Civil Protection Department, 2005

164

6.2.3 National institutions to address HMDs Institutional set-up of disaster response in Senegal Since its accession to Independence, the State of Senegal has implemented a number of laws and established institutions to take charge of climate shocks afflicting the country, and to a larger extent to manage all risks and national emergencies (natural or otherwise) when these occur. Firstly, the Law n64 -53 of 10 July 1964, promulgated four years after the countrys Independence, defines the general organization of civil protection in Senegal, and was shortly followed by a presidential Decree n64-563 twenty days later that instated the Department of Civil Protection (DCP), in charge of organizing and carrying out rescue operations, with the National Command of Firefighters. The 1964 Law asserted that: at all times the State is mandated to look after the protection of goods and physical persons and resources of the country. Furthermore, the Law of 28 February 1993 created the High Commission for Civil Protection, a consultative body in charge of prevention and organization of disasters in peacetime as well as during crises. Finally in 1999, a presidential decree instituted the Plan dORganisation des SECours, the Plan ORSEC, or National Emergency Management Plan. Each organization in this Emergency management institutional architecture has a role to play, as clearly outlined in the ORSEC plan and detailed below. 165

The National Emergency Plan, Plan ORSEC: coordination of relief operations in times of crisis

The Plan ORSEC serves as the main master plan for the organization of disaster response operations in Senegal. It was borne of the desire of different government bodies, in particular the Firefighters Command Unit, to have a commonly agreed- upon National Plan to coordinate disaster operations in time of crisis. It offers a good disaster preparedness plan, indispensable to prepare operations in advance of serious events susceptible of jeopardizing human lives, goods or the environment, and it is surprising that it took so many years for the Plan to be devised. One may read into this the lack of a need for a national Emergency management plan between 1960-1996, perhaps because of the paucity of multiple complex emergencies during this period correspondent with the thirty-year drought. Instated by presidential Decree n 99 172 of March 4 1999, updating and replacing decree n 93- 1288 of November 17 1996, Plan ORSEC defines the organization of rescue operations at the district, regional and national levels. Particularly, it establishes a plan in advance in each of Senegals 43 districts to allow for the rapid and effective summoning of all private and public means for rescue in the district204.
204 Administratively, Senegal is cut out into 14 regions, consisting each of three districts or

dpartements (except for the region of Dakar, which comprise four), for a total of 43 districts, which are subdivided in a number of arrondissements that are in turn each made up of villages (see fig. 8.5). Dpartements or districts in Senegal bear no political power, and are mere extensions of the central state, in charge of carrying out administrative functions (such as tax collection, paperwork processing, etc). Each dpartement or district is in turn composed of a given number of arrondissements, set by presidential decree according to its demographic size. Continuing down the flowchart, each arrondissement is made up of multiple villages, which constitute the bottom of the administrative ladder. The system is devised in such a way that State agents (extensions of the central State authority at the local level) reside at each level within the region to oversee and regulate the functioning of their political counterparts (the communes and rural communities). Thus each geographic unit in Senegal has both an elected political leader, and an appointed decentralized administrator: the region is administratively headed by a governor, each dpartement by the Prfet and each arrondissement by the sous-prfet. The ORSEC Plan established an Emergency Plan at the level of each dpartement or district, coordinated by the Prfet.

166


DEPARTMENT 1 DAKAR Prfet DEPARTMENT 2 PIKINE Prfet DEPARTMENT 3 RUFISQUE Prfet DEPARTMENT 4 GUEDIAWAYE Prfet

Region of Dakar Governor

4 ARRONDISSMENTS

2 ARRONDISSMENTS

2 ARRONDISSMENTS

2 ARRONDISSMENTS

Sous-prfet VILLAGES
Village chiefs

Sous-prfet

Sous-prfet

Sous-prfet

VILLAGES
Village chiefs

VILLAGES
Village chiefs

VILLAGES
Village chiefs

Fig. 8.5: Understanding Senegals Administrative Partitioning: Example of the Region of Dakar. Taking the example of Dakar as an illustration, Dakar region has four dpartements or districts (Dakar, Pikine, Gudiawaye and Rufisque), which are each sub-divided into arrondissements (leading up to a total of ten arrondissements for the region). The arrondissements, just as the dpartments or districts, hold no political power; they are mere local administrative extensions of the central State.

These dpartemental or district-level emergency Plans were designed to work as follows: only the Prfet (representative of the Central State in the district and coordinator of the district ORSEC Command Unit) can make the decision to trigger the district-level ORSEC Plan, since this plan only concerns events of exceptional proportions that overwhelm local capacity to cope (refer back to chapter 3 disaster definition). However, even in the presence of a disaster of exceptional proportions, it is the responsibility of the Mayor (elected political leader of the commune, 167

counterpart of the districts Prfet) to deploy all available communal means first, before calling on higher administration later. Once the ORSEC Plan is triggered, the Command Unit is divided into two components: the Operational Command unit that goes in proximity to the disaster zone to coordinate relief operations there, and the Fixed Command Unit that remains at the Prfecture, home of the Prfet. The same structure is replicated along all the administrative levels of the country: the Prfet triggers the district-level Plan, the Governor triggers the regional-level Plan, and the President (or his Minister of the Interior, head of the National ORSEC Plan Command unit) triggers the national emergency ORSEC Plan. Thus is designed Senegals disaster response architecture, defining an intricate interplay of administrative (central State) and political (locally elected) officials, and clearly outlining the responsibilities of each stakeholder at each geographical scale, towards the collective goal of ensuring effective and rapid rescue and relief delivery in times of disaster crises. As we shall see however, this elaborately designed response plan has not always functioned in practice. This was the case notably when opposition political parties began winning local elections, and local power was transferred to the hands of stakeholders other than the Central ruling partys. The 1996 Law did not simply institute disaster preparedness in time of crisis, but also in peacetime. Thus the High Commission for Civil Protection was established, in charge of coordinating all aspects of disaster prevention and advising the Minister of the Interior on relevant disaster prevention strategies to implement. A review of the minutes of meetings held by the Commission demonstrates however that its areas of intervention were circumscribed to: visits of public buildings (hotel establishments, fair premises, government buildings, etc.) for conformity with fire and safety 168 Disaster prevention in peace time

regulations, issuance of construction authorizations and, in recent years, television ads raising awareness on road safety measures and dangerous beaches forbidden for swimming205 Non-inclusion of climate-related disaster prevention is a noteworthy omission in the modus operandi of the organization, one that explains perhaps the lack of preparedness of Government bodies for HMDs, which became evident when recurrent floods began to inflict Senegal in the 2000s. Finally, since July 2007, Senegal is equipped with a National Platform for Disaster Risk Reduction, which brings together in concert all organizations in charge of disaster response and management (NGOs, State bodies, civil society, private sector), towards effective nation-wide prevention and reduction of disaster risks ex- ante of the disaster events, following prescriptions of the 2005 Hyogo Framework for Action agreed on in Kobe, Japan, of which Senegal was a signatory. A United Nations Development Program (UNDP)-funded project entitled, "Support to the National Program for Prevention, Risk Reduction and Management of Major Natural Disasters in the context of poverty reduction in Senegal", has played an important role in launching this platform, and helped to push forward the agenda of the Hyogo Framework for Disaster Risk Reduction in Senegal, which aims to accompany governments worldwide towards a new paradigm of pro-active ex ante disaster risk management and preparation, and not merely disaster response. One of its chief realizations has been the initiation of the national platform for DRR, and the elaboration of a national contingency plan, a useful preparedness tool which delineates roles and responsibilities for who does what when a disaster strikes. 205 Civil Protection Department (CPD), Activities of the CPD 2001-2011 169 The National Hyogo Platform for Disaster Risk Reduction (DRR)

However, limited resources constrain this national contingency plan and prevent it from being operational206 . Furthermore, the national DRR Platform fast became an additional institution that added itself to all the existing ones, not bringing them to life or delivering on its role to serve as a catalyst/synergy platform for all existing institutions to come together to define strategies and durable solutions to the issues of flooding and other climate-related crises in Senegal. Finally, the UNDP Project supported the development of a national strategy for DRR, all legislative texts for this strategy are today in place, and two national climate- informed national contingency plans have been developed since 2007. In practice however, the national strategy and Platform only exist on paper. When floods strike, all planned procedures are set aside. The disaster relief process is highly politicized; no official dares to trigger the ORSEC plan207. As proof, the platform member institutions met once only, and that was during the launch of the Platform on July 2007. When it comes to disaster prevention, in practice, the Department of Civil Protection, in charge of coordinating the platform, due notably to its limited financial capacities and lack of personnel trained to understand the complex linkages between Climate Change Adaptation Disaster Risk Reduction, suffices itself with disaster response and executing the ORSEC Plan when it is triggered. Non-governmental stakeholders such as Oxfam, Plan International, the Red Cross bypass the State and go straight to communities. To conduct isolated but high- impact disaster prevention initiatives is the only way to have impact, in the words of an NGO leader in Dakar, the State has committed suicide in Senegal208 .
www.preventionweb.net
207 Expert Interview, 2011 208 Expert Interview, 2011 206 UNISDR, Senegal Hyogo Framework for Action mid-term review report 2009, accessed March 15, 2011.

170

6.2.4 Integration of local needs in Disaster Planning and Preparedness A further zoom-in into the urgent CCA-DRR needs of local communities surveyed in a sample of Senegals vulnerable communities uncovers that accountability to local needs is also in practice not forthcoming from the national institutions in charge of climate disaster response and preparedness in Senegal. The national institution established hitherto to cater in fact is blind to priority community needs. It appears that no accountability mechanisms are built in or function in practice to channel priority local needs to the attention of national decision-makers, and link countryside with center. Table 6.5 reviews the priority CCA-DRR needs that emanated from our VCA assessment in communities vulnerable to both river and flash flooding between 2009-2011. Hazards Impacts Endogenous strategies (local capacities to cope) Family solidarity, digging of traditional canals, loan, migration, sale of assets (machine) Displacement, relocation Gaps: needed adaptation measures Harvest storage facility, Cash transfer, modern canal, availability of machines

Strong rains

Destroyed crops

House destroyed

Construction of a house with strong walls Loss of cattle Sale of cattle, Modern canals reduction in nutritional intake Difficult access to Traditional canals Water evacuation fields Loss of revenue Family solidarity Modern canals, creation of alternative income generating activities (IGA) 171

Research raids Early warnings Strengthening of Construction of homes brick houses Crops destroyed, Resignation Tree replanting harvested seeds lost Cattle lost Solidarity Construction of solid enclosures Loss of revenue Sale of assets Creation of IGAs (machines) Injured people Solidarity Construction of stronger houses Water run-off into Land destroyed Family solidarity Sturdier the village Construction Cattle loss Loss of human lives- Safekeeping of the Modern canals, people injured children, digging of Sturdier traditional construction sewers/dykes Loss of revenue Expenditure of all Construction of a cash and assets, strong Dyke, sale of necklaces creation of IGAs
Table 6.5: Vulnerability to HMDs, Local capacities to cope and Priority DRR-CCA needs in Kaffrine, Senegal.

Storms

People disappear Houses destroyed

We see that a wide gap exists between these needs and the national CCA plans as carried out, which reflects the wider disconnect between vulnerable rural areas and the hubs of decision-making in the capital city of Dakar, in Senegal. 6.2.5 Overall Institutional picture: Senegal, a Firefighter? From our overview of Senegals institutional infrastructure for climate disaster management clearly emerges that Senegal has on paper many of the attributes of an effective Firefighter, however in practice, it is an Unprepared Firefighter. Our preliminary classification of Senegal is thus confirmed. Table 6.5 summarizes our findings on national institutions for climate disaster management in Senegal, and their ability to respond effectively to urgent local needs in the face of rising HMDs.

172

Indicator Present? of Ex-post Disaster Responders (Present: (The Unprepared Firefighters policy) Absent: X) A national Disaster Management Strategy or law adopted, with clearly defined and appropriated procedures and responsibilities for disaster response In the national budget a dedicated budget line exists for disaster preparedness and emergency response A mandated agency for disaster management exists, able to coordinate across ministries and administrations and effectively command disaster response activities during an Emergency An annual Contingency plan is elaborated, and delineates the procedures, actions and responsibilities to be carried out in the advent of all possible disasters Regular drills of these disaster response procedures take place No more than three (3) to deliver needed relief to disaster- affected citizens Overall: Choosing not to, or unable, to address disasters until they occur. Countries in this category solely respond to disasters after they have occurred, mobilizing any personnel on duty and dispatching them to disaster sites. For this reason we label them: the Unprepared Firefighters! Official emergency response in the countries of this category is characterized by amateurism and improvisation, and there are no pre-established well- rehearsed procedures to follow. Also included in this category are countries with a dedicated agency for Disaster Management but one without capacity, ability for efficient inter-ministerial planning and coordination, or no budget line.
Table 6.6: Senegal, an Unprepared Firefighter in practice

X X

X X

173

6.3 Incentives Analysis: Which Factors Thwart Senegals Institutional Change For Effective Flood Risk Management And Preparedness? From our previous overview of national institutions to address flood shocks, emanates clearly that Senegal sets up, but does not implement. Which specific factors thwart the institutionalization of change for effective flood risk management and preparedness? What determines GoS implementation, or lack thereof, of the various elaborate plans devised for effective disaster preparedness and response at all levels geographic levels (from the national, down to regional and district levels)? In order to glean answers to these key questions, which get to the heart of our investigation, we review then analyze the policy responses provided by the GoS to flood shocks. What, in practice, has actually taken place during flood events between 1995-2010? How has the GoS handled HMDs? How have the devised existing plans (ORSEC, flood prevention and coordination within the Platform) been executed? 6.3.1 Process tracing: GoS Management of flood crises in practice, 1995-2010 The attention of the Government of Senegal has been hailed by a number of major HMDs and calamitous events between 1995 -2010 (see table 6.4). During this period, a total of 4 national flood crises have occurred (in 2005, 2008, 2009 and 2010), added on to 2 localized flood crises (in Saint-Louis in 1994 and 2003). The national ORSEC Plan has been triggered during 3 of those times (in 2005, 2008 and 2009), while the district ORSEC Plan had only been activated once (1994). 1994: Saint-Louis City floods Indeed in 1994, major flooding in Saint-Louis saw the rise of River Senegal to heights alarming for the safety of the city of Saint-Louis, which lies at its mouth. This had led to the triggering of the Saint-Louis district ORSEC Plan. 174

 2003: Saint-Louis City floods again Close to ten years later, in 2003, Saint-Louis city floods again. However, Senegals political configuration is completely changed then. We are three years into the new regime of Abdoulaye Wade, the popular president who rose victorious from the historic 2000 election that saw a change from the 40-year old Socialist Party regime to Wades new Democratic Party that had campaigned on the platform of Sopi or Change in Wolof. Following popular grievances and political tension in Saint-Louis due to the floods, the central government of Senegal decides to intervene. A engineering solution is decided upon to permanently resolve the flooding issue of Saint-Louis. The Langue de Barbarie Breach is thus opened in October 2003, a few kilometers away from the island of Saint-Louis, relieving the city dwellers from flooding, but sacrificing the inhabitants of neighboring villages, instantly exposed to the destructive force of Atlantic ocean tidal incursions (see Niang-Diop 2003209 and Tall 2009210). The following year, in 2004, disaster preparedness efforts begin anew, this time for Dakars suburbs. On 8 Jan 2004, within the auspices of the Commission for the Prevention and Fight Against Floods, instructions are given to the Minister of Urbanism to relocate inhabitants of the flooded zones of Pikine211. This indicates that the GoS knew what exactly to do as early as 2004. 2005: Nation-wide flood crisis 2004: Renewing with flooding in Dakar

209 Niang-Diop, Isabelle, "Les Apprentis Sorciers l'uvre Au Canal De Drivation", Walf Fadjri, October 16,
2003
210 Tall, Arame. Climate Change Vulnerability from Global to Local levels: the case of Doune Baba

Dieye http://start.org/blog/disproportional-impacts-of-climate-change-story-of-doune-baba-dieye.html
211

CPD, Activities of the CPD 2001-2011 175

In 2005, Senegal experienced its most severe floods, with unprecedented impacts (see table 6.4). The National ORSEC Plan is triggered. A nation-wide flood crisis is declared, and legislative elections due to take place that year are pushed back. The GoS responds by declaring that the 52 billion CFA francs initially budgeted for the 2005 legislative elections will be reused to relocate Dakar suburban flood victims and build 3,000 new houses for them in a new relocation site 25-km outside of Dakar, named Jaxaay or Eagle in Wolof. Thus is born the Jaxaay Plan. In a landmark speech on August 28 2005, Abdoulaye Wade speaks of needed radical measures to ensure populations no longer live in savage slums anywhere in the country, in their stead modern cities will rise: populations will be relocated to safer, dry grounds in a new modern city called Jaxaay, which will overlook the old slums, like an Eagle. The Jaxaay relocation plan will cost 52 billion CFA francs, 45 billions of which will be drawn from three-quarters of the total budget that had been set aside for the decentralized Independence festivities of 2006. Legislative elections will be postponed to 2007212. In Abdoulaye Wades epic declaration at the height of the 2005 floods, he also recalled his then Prime Minister, Idrissa Seck, who was in South Africa and declared: that the Government will no longer be taking vacations in August because that is the critical time of the rainy season when populations need us most213 . Following the rainy season, Government also began the construction of four water retention basins in the flooded suburbs214 of Dakar, to stock water during rainy season. Construction began in the districts of Pikine and Guediawaye neighborhoods of Baghdad, Gneeti-Mbaar, Wakhinan Nimzat and Medina Gounass215.
212 Le Soleil, 26 July 2010 213 Ibidem 214 Ibidem 215 Expert Interview, 2011

2007: Flood prevention at a standstill

176

Senegal then undergoes two years without any major flood crisis. Floodwaters from the 2005 disaster are still stagnant in many neighborhoods of Dakars suburbs. Disaster prevention activities come to a standstill in the country. In 2007, the political context is shaken again by the postponed legislative elections combined with a presidential election, which Abdoulaye Wade wins again, but this time with a close margin and in a victory contested by the opposition camp. In 2008, Senegal experiences major floods again, with Dakar being hardest hit. By mid-August, one month into the rainy season, most of Dakars suburbs are under water. The National ORSEC Plan is triggered. Faced with the severity of the floods, a Law rectifying the yearly budget is introduced in front of the National Assembly, and 1 billion 100 million CFA francs are gathered to fight against floods that year. 1,700 firefighters are deployed across the country216, 700 alone in Dakar. Firefighters come fast to the realization however that the problem is more structural. There are enough diesel pumps to pump waters out of flooded homes, but not enough canals to evacuate the waters declares Firefighter Captain Seck217. The West Africa seasonal rainfall outlook Forecast had been diffused, though, prior to the beginning of the season in May-June, warning of a probable above-normal season and recommending prevention measures218. The year later, on 22 March in 2009, local elections take place, and opposition coalition Benno Siggil Senegal which means Together to Uplift Senegal' wins in landslide victory. In Dakars suburbs, where many continue to dwell in homes flooded since 2005 and soaked anew in 2008, only one pro-Wade mayor is elected.
216 Sud Quotidien, November 7 2008 217 APS, Oct. 2008 218 Le Soleil, 26 July 2010

2008: Massive flooding

2009: Destructive flash floods amidst local election opposition victory

177

Dakar city is said to have lost its mayor due to his inability and lack of imagination to find solutions to flooding. A new opposition mayor is elected to replace him219. Some go as far as to purport that the Wade regime lost municipal elections over its poor management of the 2008 floods. On May 7, Dakars newly elected Regional Council President (elected counterpart of the Governor for the region) launches a National Solidarity Fund, constituted by private contributions and partner support as an attempt to permanently solution the flood problem in Dakar. Positively seen by populations as a creative solution to end the flooding problem in Dakar, the fund seeks to establish synergies between local communities and implement prevention measures to replace ORSEC responses220. As the newly elected and locally empowered opposition leaders continue their local- level flood disaster prevention activities, the battle over responsibility for flood management begins. It is the first time since his election in 2000 that Dakar locally- elected officials are not of the same camp as the Wade regime in place. Souleymane N. Ndiaye, Prime Minister and head of government in Senegals presidential system, announces on National television that management of floods will now be devolved to Local collectivities from then onwards, in the name of decentralization. Immediately following the Prime ministers address, Sitor Ndour, newly appointed Government speaker, in his first speech to the nation, announces that local collectivities are responsible for preliminary flood management efforts (e.g.: rehabilitation of underground canals to allow easy flow of rain waters which he highlights as solution to complex issue), according to articles 92 and 125 of the Code of Local Collectivities, which transfer responsibility for Sanitation to local collectivities. He affirms however that flood management is not a transferred competence, in the literal reading of the texts, and that the central State will be
219 Sud Quotidien, 3-4 April 2009 220 Sud Quotidien, 3 May 2009

178

available to provide help to the local governments. In his own words: Back in the days when local officials were from the same camp as Government, then it was an easy negotiation. Today well have to see. But the government will avail itself to support the local communities that request assistance. Ndour adds that this is not political retaliation against the newly elected officials of the Bennoo opposition. If we wanted to fight with elected opposition leaders, the retribution would be much more impactful221. But if we interpret articles 92 and 125 of Code of Local Collectivites, flood management is under the general prerogatives of Local collectivities222. Confusion seizes the country, a few months only ahead of another rainy season which local populations of Dakar, still knee-deep in last years waters, are dreading. The reaction from constitutionalists in Senegal and opposition is furious. On June 4, PIT opposition party leader calls to Resistance against the Executive powers attempts to devolve flood management to local collectivities and cut off fund supplies to them. Flood management has always been devolved to the State (mentioning the Plan ORSEC and Jaxaay plans as evidence), Why suddenly change? And mostly, right after defeats during local elections? (Amath Dansokho 223 ). Constitutional lawyers countrywide dissect the issue. Abdoulaye Mattar Diop, former minister in the Socialist regime and prominent political figure states that: The debate over flood responsibility is childish. We dont need a dictionary to fathom that flood management is a responsibility of the State. The 1996 decentralization Laws (Code of local collectives organizing responsibilities of rural communities, municipalities and districts), are very clear: Article 125 only mentions floods, and instates local collectivities to endeavor preventive measures. There is no talk of evacuation of rainwaters there. Furthermore by tradition, the government
221 Le Quotidien, June 9 2009 222 Walfadjri, 9 June 2009 223 Ibidem

179

has addressed floods. We remember Abdoulaye Wades epic declaration at the height of the 2005 floods, where he recalled his then PM, Idrissa Seck, and declared that the Government will no longer be taking vacations in August because that is the critical time of the rainy season when populations need us most. Mr. Diop continues to note when local collectivities were in the hands of the liberal regime, this issue of decentralized flood management responsibilities was never raised224 . The reaction from flood victims in Dakar was furious too. On June 4: Right before the beginning of rainy season, Djeddah Thiaroye Kao residents, organized in a Community-based Organization named ADDAD, march to demand that central State government take action. Not reassured by Interior Ministers announcement that Local collectivities are the sole responsible parties for flood management, they demand State intervention and provision of water evacuation infrastructure and sanitation equipment (water pumps): we are not frogs! We demand water evacuation equipment and water pumps. Aliou Sow, then Minister of Decentralization and Local Collectivities, reassures them the State will be by their side225; Aliou Sow days later however retracts himself to corroborate the Prime Ministers announcement that flood management is the responsibility of local governments. On June 29 2009, after letting the polemic go on for a month, Abdoulaye Wade during his weekly Counsel of Ministers, announces that he instructed the Prime Minister to support the action of local collectivities in the capital in favor of populations regarding flood relief and preparations. Thus, the President reasserted Governments role and responsibility to manage floods The State must continue to play its role, assuming fully its responsibilities in the cohabitation of competences226
224 Walfadrji, June 22 2010 225 Walfadjri, June 4 2009 226 Le Quotidien, 30 June 2009

180

While the constitutional polemic over flood management responsibilities was raging on, the West Africa rainfall seasonal forecast for the June-September season rainy season was released, warning of possible below-normal rainfall relative to 2008: rainfall will be less abundant than last year227. In August 2009, flash floods strike again. Rains may have been less than those of 2008, but they were still enough to trigger major flood disasters; this time, above- normal rainfall impacted other locations across the country, but Dakars suburbs, still reeling from past year flood impacts, remained hardest hit. On August 26, severe floods hit Dakar. Neighborhoods of Almadies and Grand Medine (in central Dakar, not the suburbs) are under water. The Patte dOie retention basin is submerged, and floodwaters take over the main roads and oil stations. The Diameguene retention basin floods and spills out onto the principal national highway. Its the deluge in Dakar. Flood disasters ensue in other regions across the country as well: in Mbour where a record 137mm of rainfall fell in a day, Bambey, Touba, Kaolack... It is a full-fledged flood disaster228 Impacts of the 2009 floods also reverberate on the academic year. Indeed many schools are flooded still in end September, while served as temporary relocation grounds for flood victims who would not leave once the schools re-opened. The 2009-2010 academic year was delayed as a resulted. As a result, many people begin to discredit climate forecasts. Were the seasonal forecasts wrong for 2009, or did Dakar indeed receive less rain relative to 2008, but the accumulation of rain waters and the gradual saturation of Dakars groundwater table (return of old lakes from the 1970s) caused flooding nonetheless?

227 Walfadjri, 12 June 2009 228 Le Quotidien, 26 August 2009

181

On August 29, faced with the severity of the flooding, the National ORSEC Plan is triggered. 2 billion CFA francs are disbursed by the Central Government to support in ORSEC Plan in 2009, with provisions for disbursement every year thereafter. Yet the sentiment of neglect from Dakar suburbs flooded victims could not be shaken off. In Gounass, Boune, Keur Massar, the administration was invisible on the field; despite the 2 billions they hear were disbursed to combat floods and organize firefighting activities. The motive? Populations are convinced this is in retribution for their non-vote for Wade during municipal elections, where populations voted massively for the opposition alliance of Benno229. Furthermore, of the 2 billion CFA francs set aside for the ORSEC Plan, the Firefighters Administration only received 125 millions230. Rumors of corruption and fraud on ORSEC funds stained the reputation and credibility of the ORSEC process231 . In light of the local protests, the Prime Minister is advised to cancel his visit of sympathy to suburbs to thwart risk of stoning232 . 2010: Floods hit again; ex-ante flood preparedness begins in earnest In 2010, floods again hit, not giving respite to the suburbs of Dakar, Joal, Mbour or Thies reeling from the impacts of yearly floods since 2005. On 14 January during the Counsel of Ministers, Abdoulaye Wade declares that flood zones are a real danger for populations that have chosen to erect houses therein. The president announces a draft law that would determine the areas that will not be authorized for the construction of houses233 . Three months later, on March 18, the President instructs to make the fight against floods a national priority. This marks
229 Le Quotidien, 26 August 2009 230 Le Quotidien, 26 August 2009 231 Expert Interview, 2011 232 Le Quotidien, 26 August 2009 233 Le Soleil, 15 January 2010

182

the beginning of the era when Government decides to take flood disaster prevention and risk management seriously. In April, the National Committee for the Fight against floods meets at a Saly seaside resort to elaborate the Program for the definite eradication of Floods, with a host of activities on schedule: rehabilitation and revitalization of old natural lakes, inter- linking of basins and lakes using canals, construction of underground water passage ways, new resettlement sites/tents and pre-made housing in Tivavouane Peulh and Thiaroye/sangalkam, additional water pumps. Total cost of this new initiative is estimated 167 Billions234. In May 20, two important meetings take place simultaneously at the local and highest policy levels. First is an Inter-ministerial committee held on flood prevention; at the end of it 8.7 billons are set aside for pumping and connection of basins and lakes. This was the first time that an Inter-ministerial committee was ever convened on the topic of ex-ante flood disaster prevention, and that money was set aside before the flood disaster. Simultaneously, in Pikine, recurrent flooding hotspot, the local development committee was also meeting on the topic of how to prevent flood disasters during the 2011 rainy season235. As a result of the Inter-Ministerial committee, in end, pro-active flood preparedness activities began in earnest, two months ahead of the rainy season. A second high level Inter-ministerial meeting is held on Flood prevention, during which the Prime Minister decrees that the army will be mobilized to support the resettlement of inhabitants of flood zones, pumping of water basins and evacuation of waters to the sea, and to help in the construction of major works, namely new canals linking different water retention basins in the Suburbs to adjoining lakes, that will then be

234 Le Soleil, July 26 2010 235 Le Soleil, May 20 2010

183

srained into the nearby Ocean. The total price tag for this operation is 9.2 billion CFA francs236. Flood prevention and finding durable definitive solutions to recurrent flooding now makes headline news. Le Soleil (government paper) dedicates its entire special supplement to the issue in its June 3 2010 edition. Ill urbanization, changing climate patterns (old lakes and water passages claiming their old territory again) and history of ill occupation of flood plains are heralded as culprits. The focus of the day is on the definitive eradication of floods in Dakar, and a complete media and popular campaign is launched to win the hearts and minds to this new endeavor. The governments campaign against floods is widely highlighted and publicized. The State was finally realizing that floods could make its regime fall, and that they needed to address the issue237.

Picture 6.5: Abdoulaye Wades caravan, half in flood waters, during a sympathy visit to Dakar suburbs flood victims (Le Quotidien, 15 May 2009).

236 Le Soleil, May 28, 2011 237 Expert interview, 2011

184

In July 26, as the first effects of the 2010 flooding set in, an Emergency Inter- ministerial counsel on flooding was held, to find solutions and position temporary shelter for the victims. Prime Minister, Ndene Ndiaye proposes pre-made houses, to be delivered in 6 weeks to relieve flood victims238. On August 17, 35 to 58mm of rainwater register in Pikine district; Aghrymet, the regional forecasting authority, alerts on possible overflowing of the Senegal River239. On August 19, Dakar is under siege! headlines Sud Quotidien, a Dakar independent daily. Diluvian rains hit Dakar and its suburbs. The road tunnel built by President Wades son, Karim, is submerged by the floodwaters. This is a point of irony which all local papers pick up on built in preparation for the 2008 Islamic Organization Meeting held in Dakar, for which Wade put his son forward entrusting him with increasing national responsibilities, the tunnel cost a total cost 22 billion of national francs, double the amount needed to solution Dakars flooding problem. By August 20, Dakar and its suburbs are under water240. As squatters flood evicted victims linger on school grounds, schools opening are delayed once again. The opposition coalition, Benoo, stages a protest march to denounce the governments poor management of floods and recurrent power cuts beginning in Senegal241.

238 Sud Quotidien, July 7 2010 239 Sud Quotidien, August 17 2010 240 Sud Quotidien, 20 August 2010 241 Walfadjri, July 30 2010

185

Photo 6.3: Drowning car in Rufisque, during the 2010 floods, and stranded people in the background. Source: Sud Quotidien, August 19, 2010

2011: Population still in flood waters

Though 2011 is not included in our analysis, we review the outcome of the Governments budding flood prevention measures endeavored in 2010. On January 6 2011, close to 4months after the end of the rainy season, populations of Dakars suburbs are still at risk242 . Pumping in retention basins has stopped, basins are replete, and the water table in Dakars suburbs is increasingly filling up causing septic tanks to retch out onto houses. Populations are afraid of a cholera outbreak. They resort to defecation pots for their natural needs. Furthermore, the roads destroyed by floods make life a nightmare for drivers and populations alike commuting between Dakar and suburbs, or anyone leaving Dakar. This is a particular hardship for the proletariat classes that have to go into Dakar every day for work.

242 Le PoPulaire, Jan 6 2011

186


Picture 6.4: Months after the rainy season ends, flood victims in Dakars suburbs still under water. Source: Le Populaire, Jan 6 2011

6.3.2 Analysis: Senegals Flood crisis management in review As a cursory reading of section 6.3.1 demonstrates, Senegals policy towards HMDs is one of response to flood crises, one that we can characterize as inefficient, disorganized, as well as ad-hoc and delayed. There is, by and large, no preparedness in practice, and disaster preparedness seems inexistent in light of the nature of responses provided to each flood shock that has occurred between 1995-2010. It clearly emanates from our review of public responses to flooding that only two stand out as drafts of long-term solutions to permanently resolve the issue of flooding in Senegal: the Jaxaay Plan in 2005 and geotechnical works endeavored in 2010. We analyze their adequacy in light of the severity of the floods that took place. 187

2005: The Jaxaay resettlement Plan

The Jaxaay Plan started off as a very innovative solution to end flooding in Dakars suburbs. Residents initially hailed Plan Jaxaay as their savior; however, they soon became disillusioned with it as little progress was made. Four years after the beginning of the plan, the Senegalese consider Plan Jaxaay a failure, and many of those whom it was supposed to help blame it for worsening their position243. A little over 1000 of the 3000 planned homes have been completed, and the retention basins planned for the flood-prone communities have been abandoned in various stages of completion. The public blames corruption for the poor progress made on Plan Jaxaay, which did probably contribute to the depletion of project funds. For instance, when the 2005 flood victims who had preferred to return to their homes after 2005 (for various reasons), were faced with flooding again in 2008 floods, many wanted to claim their old Jaxaay house allocation back; however they found their old attributed homes had been given away. 30 houses had been given away by Abdoulaye Wades son agency for the Organization of the Islamic Summit in Dakar (ANOCI) to resettled residents who lived behind presidency244. Among the reasons why flood victims did not rally the Jaxaay homes, and preferred to stay in their old flooded homes are the fact that Jaxaay homes were 2 bedroom homes, whereas the average family size in Senegal of 10people; this was highly inadequate and led to dislocated families. Isolation and distance from Dakar, work hub where suburbian proletariat has its professional activities clustered, were also factors. Furthermore, the paucity of social infrastructures in Jaxaay city, added to the fact that flood victims were asked to abandon 25-30 years of their life, and the lack of systematic destruction of old flooded houses and their

243 Boudreau, Laura, Flood Management in Senegal: Past, Present, and Future
244 Le Quotidien, 15 May 2009

188

transformation/valorization into lakes and parks, all made it so that did not go to the Jaxaay homes, preferring to stay in their old houses, feet under water. As of June 3 2010, the government claims that 1763 houses (out of the 3000 planned) were finalized and delivered245, with 2012 houses close to finished. In sum between 2005-2010 over 45 billion CFA francs were spent on Jaxaay, parcels ceded for 5million to each owner payable over 20years, Jaxaay city equipped with high school, waterways, health center and electricity246. The failure of the government to follow through on Plan Jaxaay has had the disheartening effect of causing Dakar suburbs residents to loose trust in the Senegalese government. The ORSEC Plan is a great Plan on paper, but in practice it faces a plethora of difficulties in its execution: financial, political and human blockages247 . DCP staff themselves, in charge of executing the Plan, admit these difficulties. Between 2008-2010, a total of 6 billions CFA have been injected into relief operations. Every time the ORSEC Plan is triggered, the Finance Ministry puts at our disposal 2 billions to conduct operations (Becaye Diop, Minister of Interior in Le Soleil, 26 July 2010)248. Not all this money reaches the implementers of the Plan however, most likely because of corruption and diversion of funds249. In 2008 for instance, 300 million were disbursed for ORSEC, but only 125million reached the administrator of the
245 Le Soleil, June 3 2010 246 Le Soleil, July 26 2010 247 DCP, Activities 2001-2010 248 Le Soleil, 26 July 2010 249 Expert interview, 2011

The ORSEC Plan

189

National firefighters grouping. Furthermore, most of the disbursed funds went to buying diesel for the operation of water pumps. In 2009 as well, money went to repair broken water pumps. Paraphrasing the words of a popular newspaper, The ORSEC plan is broken250. It has also come to be called the Plan Weurseuk, meaning the Plan of Good Luck, because every time it is triggered, a number of government officials seem to be able to build two-story buildings overnight and buy luxury cars. The rodents around flood management are NOT looking forward to a durable solution to the issue of floods: they thrive on the chaos. It is difficult to have a Long-term vision, define coherent strategies and get country on dry land when you thrive on the money destined to the disaster victims and cash in every time there is a disaster251. These corruption allegations cannot be confirmed, however they pose a credible hypothesis to explain the lack of a coherent exit strategy out of recurrent floods in Senegal, even when a major electoral bastion (such as Dakars suburbs) is at risk. Furthermore, an anecdote during the 2009 floods is revealing of the lack of institutional appropriation of the ORSEC Plan. In 2009, Abdoulaye Wade was absent on annual holiday in August, the height of the flood season, and the Minister of Interior, Becaye Diop, did not even dare trigger the National ORSEC plan in Wades absence. You want me to get fired? he retorted to one request to trigger the national ORSEC252. The then Minister of Interior during 2009 floods, had to wait for the return of the President, who was on vacation in Switzerland, since only his chief of staff had received instructions allowing him to call him in Switzerland. Such an episode, though anecdotal, demonstrates the extent of personalization of the Senegalese political system, which is in effect centered around one person, the
250 Le Quotidien, 26 August 2009 251 Expert interview, 2011 252 Expert interview, 2011

190

Supreme leader, and is in practice run as a one-man show. This corroborates the literature on Neo-patrimonialism ran on African States to which Senegal is no exception. In time of crisis, when leadership and strong institutions are most needed, is when the vacuum of functioning and appropriated institutions becomes most salient. The unfolding of flood events in Senegal, and governments poor inadequate response demonstrates it. Conclusion: which determinant factors thwart institutional change? A review of Senegals response to flood crises between 1995-2010 since national policy is reduced to just that: ex-post hurried responses to flood events wherever they occurred, according to where political tension was most rife and urgent, with no thought for pre-established protocol or action plans, and poor ability for effective disaster planning, response and rehabilitation reveals an interesting tale of political motivations and a non-responsive central State non-committed to facing an issue of rising importance, until it could no longer be avoided. It demonstrates how intertwined climate crisis management is with politics, and the political treatment of crises only when they compromise political interests. In 2003, the Saint Louis breach was opened only when floods became an issue threatening the city. However breach was not a long-term solution, but rather a publicity spot that got political results, appeasing rising grievances by Saint Louis populations, but did not really solve the issue of flooding. As proof: severe floods recurred in Saint-Louis in 2010. Political disaffection there could have motivated the States response. Furthermore, the fact that the State devised a solution that resolved only the problems of the city dwellers of Saint-Louis, at the cost of aggravating the environmental challenges of neighboring rural communities around Saint-Louis (for 191

instance, Doune Baba Dieye residents) exposed by the engineered Langue de Babarie Breach opening to destructive Atlantic Ocean tidal incursions, also appears consistent with the claims that State only reacts when a key electoral bastion is at risk. Those were the early days of the Wade honeymoon with his people, and he could not fall short of the expectations from his new regime in a city as important as Saint-Louis. In 2005 following the Dakar floods, the Jaxaay plan was launched. Politicians jumped on this opportunity, in a context of high loss of political ground, to postpone legislative elections. Retention basins were built, however these basins are reported to have filled up as early as the next year253 (Expert interview, 2011). This was not a durable solution either; rather it appears as though the Government simply was trying to appease contention. In 2009, after three years of relative calm when flood preparedness got off the national agenda, further floods occurred, in a context where municipalities fell to the opposition. Governments response here is interesting. In order to enervate a newly empowered political opposition that had won all of the municipal election seats, the central ruling Government tried to dodge its responsibilities for flood management, using the decentralization scapegoat, to claim that Sanitation (implied flood management) is one among the competences decentralized to municipalities. These newly elected opposition local leaders thus ought to be in charge of the well- being and security of their local municipalities. However this strategy backfired, and served only to pit local flood victims against a government which they found increasingly detached from their immediate needs and concerned (living in green stagnant waters). Faced with the fury of the victims themselves who took to the streets, mostly in Dakar, the State backed down on this decision, and reclaimed its responsibility of Flood management.
253 Walfadjri, 9 June 2009

192

Just as the State did, it is important to note that the opposition also instrumentalized floods and attempted to use them as political capital. Also as floods intensified, and people were in deep distress, we saw instances of flooding used as political capital by both ruling and opposition camps. A case in point was the film entitled Plan Jaxaay! screened at opposition party house in 2007, in order to discredit Wades regime and rally the people to demand financial accountability; however no further efforts from the part of the opposition were endeavored to find real solutions for local communities. Thus the State, attempted to shirk its responsibilities on climate crisis management until the issue could no longer be evaded, and it became a hot potato on the agenda. As commented by one expert interviewed, the State came to the realization that its regime could fall over the flood issue254. 2010 marks as such the break year with regards to flood management in Dakar. Indeed, that is the year when not just show construction works, but meaningful long-term actions began to be posed to resolve the Dakar flooding problem. For instance, upon the request of the Department of Civil Protection, a study was instigated to map out Dakar suburbs hydrographic network, towards a feasibility study for the digging of canals linking the 5 former natural lakes together, to reinstate natural (obstructed) hydrographic system. This project was in addition to the one pursuing continued pumping of water in the largest lake (Wouye) for drainage of excess waters into the nearby ocean, freeing 52 neighborhoods of Dakar from the hell of stagnant rainwater. Thus, it is noteworthy that in 2010, Knowledge began to be utilized to inform disaster management strategies and prevention, a major innovation. However outside of Dakar: no change in policies occurred. In Saint Louis, which at least had received some attention in 2003, the ensuing flood events garnered no public attention (notably the severe 2010 floods for which government refused to trigger national ORSEC Plan). Elsewhere in the country, in rural communities
254 Expert interview, 2011

193

impacted recurrently (such as Malem Hoddar) no actions have been posed to address those citizens needs, and deliver flood risk reduction to them as well. From this analysis we see a trend arising. As floods intensified, most particularly during the 2005-2010 period when floods recurring almost yearly became an omnipresent strain on the local populations, the urgency of meaningful flood preparedness actions could no longer shirked. The most vocal demands for such action (public good delivery) from the Central State came from the aggrieved flood victims of Dakars suburbs, who faced higher densities of vulnerability to yearly flooding disasters, and were better mobilized by the opposition. The central Government of Senegal could no longer ignore these local demands; they had to deliver on flood preparedness measures, and respond to popular pressures, which they finally did in 2010. However as our preliminary review of the outcome of the 2010 preparedness measures shows, the minute the State can get away with no longer delivering on the public good, it will (e.g.: pumps no longer function 4 months after flood disaster, while populations of Pikine still at risk from flood-related injuries and epidemics). Thus here, THE LOCAL PRESSURE AND DEMAND HYPOTHESIS obtains. Faced with repeated strain from intensifying HMDs that became a yearly reality in Dakar since the 2005 record flood, which in turn led to stronger local pressures from a mobilized and aggrieved local population, the State was forced to act on Disaster preparedness. It could not shirk its responsibilities any longer. Also pressure from donors, including the World Bank with whom the Government of Senegal jointly assessed flood damage in 2009, and called upon yearly to contribute to flood disaster response operations, possibly weighed on the balance, adding in an additional voice to demand for more effective and rational flood management measures to avoid the yearly disasters. 194

Finally, our analysis also suggests an ELECTORAL BASTION HYPOTHESIS. In 15 years of flood management (1995-2010), flood disaster preparedness was only delivered to two cities: Senegal and Saint-Louis, two major electoral hubs for the Wadist regime. This theory is a second-order one however, which has to be tested more systematically and rigorously across different cases within Senegal, and outside, using data on electoral results in search for systemic correlations with areas where public good delivery on CCA-DRR was forthcoming. Furthermore, a counter-argument that raises concerns on the validity of the electoral bastion at risk theory is that other parts of Senegal, somewhat as equally important electoral bastions (Thies city for instance), only received flood response; there no DRR-CCA measures aligned with local needs were forthcoming. In light of these arguments, the first theory local pressure and demands as means of enacting effective CCA-DRR from the State, appears as a more cogent explanation of States delivery on CCA-DRR measures. In 2003, a Saint-Louis engineering solution to flooding was delivered as a result of local demands and protests. Long-term solutions to Pikine flooding were also endeavored in response to the severity of the floods that gripped this area of Dakars suburbs from 2005, and to the ensuing popular grievances and calls for State action therein. The State as such only appears to react where local demand is rife. However, it is important to note that whereas localized solutions to Pikine flooding began in earnest in 2010, they were never generalized to the entire country. They were never institutionalized or translated into nation wide policy for improved effective disaster preparedness and relief. This confirms our theory on localized popular pressure as a trigger; and begs the question of scale: how much and how widespread local demand is needed to trigger national institutional change in the face of more intensive HMDs? In the final section of our study, we bring our analysis to a close with a listing of all the factors that seem to have come into play in Senegals management of floods, to have contributed to the 2010 inception of long-term flood prevention measures, and 195

the notable absence of such plans in other parts of the country, as vulnerable if not more. 6.4 Summary of Variables that Potentially Explain The GoS Inability to Handle Complex Climate Shocks in Senegal Our case suggests that the government of Senegal does not manage climate crises well; rather it manages these in an ad-hoc fashion, handles crises as they arrive, running from one to another, according to where political pressure is strongest; without any foresight or effort at anticipating, and that without this political pressure, they will not act. It is governing through response; when in fact, governing is the art of planning. Today science offers ways to anticipate climate risks, monitor vulnerability factors, so that hazards do not turn into full-blown disasters that disrupt local lives livelihoods, and jeopardize feeble development gains. Yet, it is unfortunate to note that the government, despite availability of international support, is unable to manage climate risks well. From our study, the main systemic factors that seem to explain this inability and obstruct the adoption of meaningful long-term solutions to the issue of recurrent flooding, is the lack of Political will to tackle the issue resolutely, and deliver on CCA- DRR. Our case suggests that this lack of political is caused by: 1) Differing timescales between needed flood interventions and political rationales: politicians abide by a political rationale (short term, electorally motivated), whereas the imperatives and policy planning needs for flood prevention are in the long term. Indeed, Politics in Senegal as in many other neighboring countries is the medium to rise from nothingness to the heights. Politicians have a short-term view as a result of this, and will pillage public coffers, and do everything necessary 196

to ensure their reelection/stay in power. In such a context, long term planning is difficult. 2) Overpoliticization and instrumentalization of floods: by the government as well as by the opposition leaders 3) Corruption- Government officials thriving on corruption around inflowing disaster relief money, notably during the Plan ORSECs implementation; 4) Institutional vacuum: institutions exist for the effective preparedness and response delivery on disasters, but they do not come to life, no institutional synergies are weaved around the issue. Existing institutions would rather send a fax warning, than implement innovative strategies to get the information to vulnerable communities at risk, such as SMS In 2010 however a shift in policy occurred in Senegals climate crisis management, from inefficient delayed ex-post response to the seeds of a disaster prevention strategy, at least in a localized turf of the territory. The factors that seem to have enabled such a shift are as follows: 1) A key electoral bastion was at risk: Dakars suburb of Pikine It could be that Abdoulaye Wades regime lost 2009 local elections as a result of cumulated frustrations from populations over an array of issues (increasing inflation, the Islamic Summit wastes, permanent electricity shortages, etc.); however it is clear that flood grievances played an important role as well in Wades defeat at elections in 2009. Thus making politicians incentives align with CC policy planning incentives (build political interest for handling issue) seems to be a way forward to force government to implement adequate climate crisis management policies. Its been demonstrated to work in Senegal, when Dakars suburb of Pikine, important electoral bastion 197

became at risk, the government took on the issue and placed it on the national agenda. 2) International pressure and donor funding also seemed to have played a key role. Indeed, the World Bank 2009 report was instrumental in bringing the State in front of its responsibilities after 2009; furthermore funding from the UNDP and UN-ISDR enabled capacity building to place the issue of floods on the agenda; 3) Finally, the most important lessons from this case is that once the flood afflicted populations of Pikine took their grievances in their own hands, in the streets and at the polls, the Government had no other choice but to move to action. The governments defeat at the 2009 elections served as eloquent illustration of the power of the people to reclaim their destiny, when a government is unresponsive to their needs. Will this policy of climate crisis prevention be sustained? That is a question that future generations of scholars researching this question, as more and more disasters hit Senegal and the rest of the continent and climate crises prevention moves onto national agendas, will have to answer. The Senegal case suggests a pathway for institutional change to effectively address the challenge of intensifying HMDs however, in this case flooding (see fig. 6.6). Our next investigation next takes us to Uganda, where we will assess whether the same triggers of institutional change on CCA-DRR hold. It is important to note the limitations of the present study are that the floods between 2008-2010 are the most documented (a total of 30 newspaper articles between 2008-2010), and as such received more attention in their treatment. 198

6.8 Pathway of Change According to the Senegal Experience: Enabling Conditions for the Adoption of Socially Beneficial Institutions to Address the Challenge of CCA-DRR

LOCAL demand for effective vlood management

Donor support

Political will to address CCA-DRR

Rise in HMD
2005 massive vlooding in Dakar 2008 food Floods yearly after
Donor Support

Political will to address CCA-DRR

Rising Popular disaffection

LOCAL Demand for improved vlood management

Fig. 6.6: Pathway of Institution Change for effective Climate Change Governance, based on the Senegal experience of managing intensifying floods, 1995-2010. Note: Rise in Popular disaffection as an Intervening Variable, the Independent Variable (nd true driver) being intensifying HMD.

199

CHAPTER 7 UGANDA COUNTRY CHAPTER: What Triggered Institutional Change on CCA-DRR in Uganda? Our investigation of national triggers of institutional change in the face of rising climate hazards now takes us to Uganda, in the heart of East Africa. Uganda was selected as the country case to infer for the Yellow policy group of Prepared Firefighters, the efficient responders to hydro-meteorological disasters (HMDs). Is this classification verified by facts? If so, what factors led to Ugandas transition to the category of a Prepared Firefighter? 7.1 Country Profile

Fig. 7.1: Uganda in East Africa

Embedded in the Great Lakes Region of Center-East Africa, Uganda is a lush green country of rolling hills that straddles the equator, and occupies 241,038 km2, of which 43,941 km2 of open water and swamps (18.2% of the countrys total area). 200

Landlocked and bordered in the North by Africas youngest Republic of South Sudan, in the east by Kenya, in the West by Congo (Zaire) and in the South by Tanzania and Rwanda, Uganda comprises 45% of the worlds second largest freshwater lake, and Africas largest lake: Lake Victoria, source of the White Nile (see fig. 7.1) and largest inland fishery on the continent. A British colony until its independence in 1962, Uganda is often referred to as the Pearl of Africa for its natural beauty. Africas fruit basket would have been more adequate. With humid and hot equatorial climate throughout the year, Uganda is endowed with two rainy seasons in the year (one from March to June, while the second season is from August to November), which merge into one long rainy season as you move northwards from the equator. Mean annual rainfall ranges from 700mm in the drier areas to about 1,500mm in the humid areas. Most parts of Uganda are also in altitude, with an average height of 1,200m above sea level. The lowest altitude is 620m (in the Albert Nile) and the highest altitude (Mt. Rwenzori Peak) is 5,110m above sea level. 7.1.1 Vulnerability Context Uganda is a low-income country of close to 33 million and a half inhabitants, with a low GDP/capita of 503 USD, albeit a strong annual GDP growth rate of 5.2%255. Uganda was one of the first Sub-Saharan African countries to embark on liberalization, structural reforms and pro-market policies in the late 1980s, at the end of its Civil war in 1986, which firmly instated the power of President Yoweri Museveni. Since then, the government has maintained a stable macroeconomic environment, accelerated growth from an average of 6.5% in the 1990s to over 7% during the decade of the 2000s, and sustained private-sector oriented reforms that graduated Uganda into a mature reformer in 2006. Ugandas growth over the years

255 World Development Indicators, 2010/2011

201

has remained well above Sub-Saharan Africa average256. However, rapid population growth over the years, one of the highest in the world, has limited the distribution of the fruits of growth, driving real GDP per capita down to its current levels. Furthermore, progress in primary education completion, child and maternal mortality, access to reproductive health and incidence of malaria and other diseases have been slow (see table 7.1). Ugandas urbanization rate is low compared to other African countries, with over 80% of Ugandas population still rural and dependent on rain-fed agriculture257. The Ugandan government is becoming increasingly concerned about this uneven progress and the distinct geographical patterns of outcomes in health and education resulting from unequal access to services, which pose challenges to Ugandas aspiration to become a middle-income country within one generation. Nonetheless, strong national economic growth enabled substantial poverty reduction and progress towards MDGs. With the proportion of people living in poverty at 25 percent in 2009/10, Uganda has surpassed the 2015 Millennium Development Goal of halving the 56 percent poverty rate recorded in 1992/93258. There is also significant progress towards reducing the share of population suffering from hunger. Uganda may achieve the goals of universal primary education, gender parity, and combating HIV/AIDS which was also slashed from a record 29% prevalence rate in urban areas in the late 1980s to the current rate of 6.5%, through strong government leadership, broad-based partnerships and effective public education campaigns259. Thus, while Uganda has been one of the few durable success stories of Africa, its
256 World Bank, Uganda Country Profile, accessed Dec 1, 2011. http://go.worldbank.org/8XKQR04V10 257 NAPA Mozambique, 2007 258 World Bank, Uganda Country Profile 259 Avert Uganda, accessed Dec 1, 2011. http://www.avert.org/aids-uganda.htm

Development challenges: past & present

202

current development challenges consist of sustaining macroeconomic stability while moving the economy to a higher productivity level, inclusive of all regions into the development process. Ensuring equal access to quality social services, in particular education and health, while promoting growth through infrastructure investments in places where it will earn the highest return are key priorities260 . One of the countrys newest challenges as well is to manage its resources, in particular its fast growing youthful population and newly discovered oil. According to the World Bank, in order to reap the demographic dividend, Uganda must invest in fertility reduction, quality education and skills development, and job creation. To reap the oil dividend, Uganda must maximize the social benefits through adequate investment and prudent macroeconomic management of the oil sector, as well as transparency and management of expectations261. Unfortunately, perceived mal-governance and increase in corruption threatens to deteriorate Ugandas image as a development model and trustworthiness vis--vis its development partners and donors. Indeed, the country needs to decisively address increasing petty corruption, the perceived growing culture of impunity for grand corruption and pervasive quiet corruptionthe failure of public servants to deliver goods or services paid for by governmentssuch as unchecked teacher and health worker absenteeism262. Uganda remains a poor country with a low level human development, and lies in the lower rungs of the human development ladder, classified 161st out of 187 countries ranked by the Human Development Index ranking263.
261 Ibidem 262 Ibidem 263 The United Nations Development Program Human Development Index (HDI) is a composite index measuring

260 World Bank, Uganda Country Profile

life expectancy at birth, adult literacy rate, poverty, gender inequality, sustainability (impacts of natural disasters and carbon emissions per capita) and gross domestic product per capita, to provide a measure of countries human development levels.

203


Table: Ugandas Development Indicators Country size (surface area, thousand km2) 241.038 Human Development Index Ranking Low (Rank 161) Population (million) 33.4 Population growth (%) 3.2 GDP (current US$) (billions) 17.01 GDP per capita (current US$) 509 GDP growth (annual %) 5.6 Income share held by lowest 20% 5.8 (2009 estimate) Births attended by skilled health staff (% of total births) 42 (2006 estimate) Life expectancy at birth, total (years) 53 (2009 estimate) Mortality rate, under-5 (per 1,000) 99 HIV prevalence (% of population ages 15-49) 6.5 (2009 estimate) Literacy rate, youth female (% of females age group 15-24) 85 Primary completion rate, total (% of relevant age group) 57 Mobile cellular subscriptions (per 100 people) 38 Agricultural land (% of land area) 70 Agriculture, value added (% of GDP) 24 Industry, value added (% of GDP) 25 Services, etc., value added (% of GDP) 50 Exports of goods and services (% of GDP) 24 Imports of goods and services (% of GDP) 34 Coastline length area at risk from SLR 0 km (landlocked) Water bodies (% of total land area) 18.2 Total Number of HMDs (1960-2011); source: CRED 29 % change in average number of HMDs +7.7 (1960-1994 vs. 1995-2011) (Exponential rise) Net official development assistance and official aid 1,786 received (current US$) (millions) (2009 estimate)

Table 7.1: World Development Indicators, Uganda (2010 est., unless otherwise noted)

7.1.2 History & Political Context Uganda features a diverse ethnic and cultural mosaic, with more than 40 different ethnic groups and over 30 distinct indigenous languages264. It is not surprising in
264 The largest ethnic group in Uganda is the Baganda, comprising about 17% of the population, and located

largely in the central region, followed by the Banyankole and Bahima (10%), the Bakiga (7%), the Banyarwanda (6%), the Bunyoro (3%) and the Batoro (3%) in the southwest. In the North live the Langi (6%) and the Acholi (5%). In the northwest are the Lugbara (4%). The Karamojong (2%) occupy the considerably drier, largely pastoral territory in the northeast. In the east dwell the Basoga (8%) and the Bagisu (5%) (US State Department Background Notes: Uganda, 2011).

204

this context that when Ugandans speak about pluralism, they think of ethnic diversity. Colonial boundaries grouped diverse ethnic groups that had distinctive political systems and territories. When Arab traders moved inland along the Indian Ocean coast of East Africa and reached the interior of Uganda in the 1830s, they found several African kingdoms with well-developed political institutions dating back several centuries. These traders were followed in the 1860s by British explorers searching for the source of the Nile River. Protestant missionaries entered the country in 1877, followed by Catholic missionaries in 1879. In 1888, control of the emerging British "sphere of interest" in East Africa was assigned by royal charter to the Imperial British East Africa Company, strengthened in 1890 by an Anglo-German agreement confirming British dominance over Kenya and Uganda. In 1894, the Kingdom of Buganda was placed under British protectorate265. Institutional soul-searching in the wake of the new nation

In 1962, the country became independent from British colonial rule, without having to fight for it. Indeed, the British determined a timetable for withdrawal before local groups had organized an effective nationalist movement. Uganda's political parties thus emerged in response to impending independence, rather than as a means of winning it266. In part due to the result of this fairly smooth transition to independence, the near absence of nationalism among Uganda's diverse ethnic groups led to a series of political compromises. The first of these compromises was a federal government composed of coalitions of local and regional interest groups, loosely organized into political parties. Following the Uganda Constitutional Conference held in London in 1961, Uganda achieved internal self-government and organized legislative elections in March 1961. A first
265 US State Department Background Notes, 2011 266 Encyclopdia Britannica Online, 2011

205

institutional solution of a National assembly was attempted then, headed by a Chief Minister, Benedicto Kiwanuka, leader of the Catholic Democratic Party. However in April 1962, new elections were held (where turnout was higher among the Buganda) and Kiwanuka lost to a coalition of the Baganda royalist party, Kabaka Yekka, and Milton Obotes Uganda Peoples Congress (with support mainly from the West and North). The Chief Minister position was dissolved. A new national government was formed, led by a ceremonial king, and presided over by a prime minister whose principal role appeared to be that of a broker, trading patronage and development projects--such as roads, schools, and dispensaries--to local or regional interest groups in return for political support. It was not the strong, directive, ideologically clothed central government desired by most African political leaders, but it functioned. This system might reasonably have been expected to continue to function, because there were exchanges and payoffs at all levels and to all regions. In the wake of full independence on October 9 1962, Edward Mutesa, King of the Buganda, became the first ceremonial president and Milton Obote, the first executive Prime Minister. Uganda experienced a short-lived period of political and economic stability thereafter. In February 1966, Prime Minister Milton Obote suspended the constitution and assumed all government powers, removing the positions of president and vice president. In September 1967, a new constitution proclaimed Uganda a Republic, gave the president even greater powers, and abolished traditional kingdoms. This marked the beginning of tyranny and oppression. In 1971, while Obote was away at the Commonwealth Heads of State meeting in Singapore, then army officer Idi Amin conducted a military coup, which settled Uganda on a trajectory of violence, terror and mismanagement, and reduced the The reign of terror

206

country to a failed state and a collapsed economy267. In 1972 notably, Idy Amins regime expelled 50,000 Asians, the largest non-indigenous ethnic group in Uganda, who had been engaged in trade, industry, and various professions, and seized their property for personal gain. The legend of Idy Amins cruelty and dictatorship extended beyond the borders of Uganda, when he invaded Tanzania to distract away for his countrys collapsing economy. Political and economic turmoil continued between 1979 and 1985, with successive coups, one of which ousted Idy Amin, and a disputed election in 1980 that brought Milton Obote back to power for a second time, and led to civil conflict across the country. In 1986, the National Resistance Army (NRA), led by Yoweri Museveni, took power after a coup dtat. Then began a period of sustained economic and political renewal. National Resistance Army (NRA) was transformed into the National Resistance Movement (NRM), and a one-party system was established. During the first decade of NRM rule, the government focused on reconstructing the economy through pro-market reforms and increasing the legitimacy of government institutions through political liberalization, to the contentment of Ugandas new strategic partners, the US, in a context dominated by Cold War politics. However, in 1987 a brutal civil war erupted in Northern Uganda, waged by the Lords Resistance Army (LRA) of Joseph Kony who mobilized on the basis of Holy Spirit fanaticism, preaching among frustrated Northerners, mainly Acholi. The Acholi since Independence had felt as second-class citizens sidelined from the benefits of agricultural development and growth, centered in the countrys south. The civil war left thousands dead and millions displaced, constraining economic activity and entrenching poverty in the North. After twenty years of protracted peace efforts, the LRA was pushed out of Uganda in

267 World Bank, World Bank, Uganda Country Profile

207

2005. Since then, there have been no major security incidents. Economic activity has resumed in northern Uganda, and most internally displaced persons have left IDP camps and returned to their land. This greatly contributed to the reduction in poverty from 60 percent in 2005/06 to 45 percent in 2009/10268. Uganda has also slowly progressed towards multi-party democracy. In 1995 a new constitution was promulgated, opening the political sphere to elections. President Museveni was elected to a first term through a non-party election in 1996. He was re-elected in a contested election in 2001. The constitutional amendments approved by a referendum in July 2005 introduced multi-partyism and Parliament lifted the two (5-year) presidential term limits, which allowed President Museveni to seek a third term in office and his NRM Party to win the first multi-party elections in 2006. On February 28, 2011 President Museveni was re-elected for another 5-year term after garnering 68 percent of the vote, with his party having an overwhelming majority in Parliament. Serving his 25th year in power, President Museveni is one of the longest serving leaders in Sub-Saharan Africa269. Decentralization Context and State of Accountability Uganda is administratively and politically subdivided into 112 districts (see fig. 7.2),

then counties, sub counties and villages as the lowest political administrative unit. Each district has a District council, a Municipal council with
Division councils, Town councils and multiple Sub County councils. The Chief Administrative Officer (CAO) heads all the technical teams in the councils. Below the CAO at the district level, are 11 directorates with several departments. The technical teams at the local governments are headed by Administrative Officers, below who are extension officers. The political arm is responsible for policy implementation

268 World Bank, Uganda Country Profile 269 Ibidem

208

while the technical teams undertake the implementation of lawful decisions of the councils.

Fig. 7.2: Ugandas 112 districts in the four regions (Central, East, West and North). Source: Wikipedia Commons, 2011.

7.1.3 Vulnerability To Climate Change and variability

Uganda is a country highly sensitive to both longer-term Climate change and annual and inter-annual climate variability.
Ugandas main strength is that it is richly endowed with the bounties of nature: good soils and biodiversity, ample vegetation cover, attractive climate and abundant water resources. In some areas, these gifts of nature still remain intact, thereby availing the opportunity to utilize them sustainably270

Physical vulnerability

270 Government of Uganda, Vision 2025

209

Climate, perhaps Ugandas most valuable natural resource, is the most neglected. The climate of Uganda is not only a natural resource, but a key determinant of the status of other natural resources, such as water resources, forest, agriculture, ecotourism and wildlife. Ugandas climate can be broadly subdivided into: Highland climate: cool temperatures (as low as 4 degrees in some parts of the highlands) and moderate rainfall (mean annual rainfall of over 900mm). In the Rwenzori Mountains, which have a permanent ice cap, temperatures of below 0 degrees Centigrade are experienced. The beauty of this climate zone is reflected in its plentiful waterfalls as illustrated by figure 7.3; Savannah tropical climate, including the lake basin climate: with moderate average temperatures of 28 C and high mean annual rainfall of over 1200mm. The tropical rainforest is found in this climate. Swamps, found mostly in this climatic zone, provide an excellent habitat for birds; the most notable being the Crested Crane, national bird of Uganda; Semi-arid climate, with relatively high average temperatures, ranging from 26.3 to 29.0 C and a relatively low mean annual rainfall, between 887 905 mm. Animal rearing is the dominant activity in this climate. The high animal population has led to serious land degradation. Although the mean annual rainfall is relatively low, some drought-tolerant land crops can still grow.

210


Fig. 7.3: Luzzi waterfall, Bulambuli district, one of the multiple waterfalls in the highlands of Mt. Elgon (Photo: Arame Tall)

Each bioclimatic region thus features distinctive climate-determined biotopes and environmental conditions. The subdivision of the Ugandan climate also conditions the distribution of natural resources such as water, forest and vegetation. Ugandas great climatic diversity explains then how we can have within the same country temperatures below 0C on the mountain ranges of Rwenzori and mount Elgon (Rwenzori has a permanent ice cap), while Gulu, Kitgum and Moroto in the north and North Eastern part of the country can experience temperatures over 30C. CLIMATIC CHANGE The same diversity in rainfall regimes is also true (see Figure 7.4)

211

Fig. 7.4: Mean annual rainfall in Uganda. Source: Uganda Department of Meteorology.

One particularly fragile ecosystem, dependent on rainwater for human consumption and production, is the Cattle Corridor (see fig. 7.5). Stretching from the northeast to the southwest, The Cattle Corridor lies in the semi-arid climate zone, and is predominantly a pastoralist area. The prolonged drought of 1999/2000 caused severe water shortage in this area, animal loss, decreased milk production, food insecurity and increased food prices. Frequent droughts result in lowering the water table and drying boreholes.

Fig. 7.5: The Cattle Corridor (shaded brown). Source: NAPA, 2007.

212

Socio-economic vulnerability

Natural resources constitute the primary source of livelihood for the majority of Ugandans. Indeed, the economy of Uganda depends on exploiting its natural resources and will remain so for the foreseeable future (See Uganda Vision 2025271 ; also Mozambique NAPA 2007272). Management of these natural resources is therefore important and critical to Ugandas long-term development. Forest products for instance (timber, poles, rattan, bamboo, food, fodder, medicine, and firewood) and services (biodiversity habitat, moderation of micro-climate, shade/wind breaks for enhancing agricultural productivity) play an important role in the soco-economic development of Uganda, contributing 6.1% of national GDP. Forests are especially critical to rural community livelihoods. Over 99% of Ugandas rural people use wood or charcoal as fuel. In economic terms, the work by subsistence users of firewood and charcoal is equivalent to about 725,000 jobs273. Similarly, commercial users of forest products provide work equivalent to 120,000 jobs274. High Sensitivity to Rainfall fluctuations Social and economic activities in Uganda are most affected by rainfall variations, the most sensitive climate variable. The rainfall regime allows two planting and harvesting seasons a year in most parts of the country, without the use of irrigation. Fig. 7.4 shows mean annual rainfall distribution in Uganda. The typically wettest districts are located within the Lake Victoria Basin, the eastern and the northwestern parts of Uganda. These areas include Kalangala, Kampala, Mpigi, Mukono, Jinja, part of Masaka and Bugiri (Lake Basin), Mbale and Kapchorwa (eastern) and Arua (northwestern). In recent years however, it has been observed that rainfall is heavier and more violent on the one hand; on the other, the western,
271 Government of Uganda, Vision 2025 272 Uganda NAPA, 2007 273 Ibidem 274

Moyini, 2005

213

northern and northeastern districts are experiencing long droughts, which are becoming more frequent. Increasingly erratic rainfall coupled with increasing frequency of droughts has made Uganda more vulnerable to climate changes275. Water resources About 34% of the country is covered in wetlands with a dense network of rivers, lakes and swamps. Indeed, Uganda has abundant water resources, with up to 15% of Ugandas total area covered with water (80% of which is accounted for by Lake Victoria) and a mean annual rainfall ranging from 700mm in drier areas to about 1500mm in the humid areas. However its distribution is not even, particularly in semi-arid areas. The semi-arid areas of the country experience water stress. Prolonged and severe droughts led to lower water levels in rivers, underground aquifers and reservoirs, affecting the hydrology, biodiversity and water supply. The severe drought of 2004/05 notably contributed to the reduction of the Lake and Nile River levels with serious impacts on power generation, leading to power rationing in the domestic and commercial sectors, and resulting in the interruption of economic activities and to a decline in manufacturing outputs276. Furthermore, the rural poor depend on streams and swamps. A large proportion of the population depends on streams, which tend to dry up during droughts causing serious water stress for a significant proportion of rural communities. Water scarcity in such areas has resulted in population movements into neighboring districts in search for pasture and water. These movements have frequently led to ethnic conflicts and disruption of production, affecting community development. Water scarcity in dry lands is likely to worsen with CC277. 275 Uganda NAPA, 2007 276 Ibidem
277

Ibidem

214

Agriculture Agriculture is the backbone of Ugandas economy. It constitutes about 42% of GDP, over 90% of export earnings and employs over 80% of the labor force. Ugandas agriculture however is subsistent, rain-fed and, therefore, vulnerable to climate variability and climate change. The contribution of agriculture to total GDP has decreased from 45.7 percent in 1995/96 to 41.5 percent in 1999/00. The decline in agricultural contribution to GDP is not a sign of a diversified economy, but due to the impact of the factors that influence agricultural production. These include soils, climate, agricultural implements, management practices and access to markets (both domestic and international). The decline in agricultural production is also partly explained by the 1999/2000 drought that caused severe water shortage leading to animal loss, decreased milk production, food insecurity, increased food prices and generally negative effects on the economy. Prolonged droughts can have serious impacts on agricultural production. Even long dry spells during the rainy season are sufficient to reduce agricultural production, thus seriously impacting livelihoods of rural communities. Poor agricultural production has direct negative effects on the national economy; increasing food prices and resulting into inflation and an unstable macro-economy, which not only discourages foreign investment but also leads to frequent health breakdowns, thus affecting production and decreasing standards of living278. Although it is predicted that climate change will likely lead to increased rainfall in Uganda279, its distribution during a season is critical to agricultural production. Erratic rain seasons have been observed in recent few years, and rainfall seasons have been more variable. Floods have led to waterlogged fields or washing away of crops. Poor people frequently settle in or close to wetlands; during floods, such families are vulnerable since their livelihood source is no longer accessible for
278

Ibidem

279

Ibidem 215

agricultural production. In agriculture, there is also high crop sensitivity to temperature increases, the most likely impact of global Climate Change. Other factors that make the sector vulnerable, beyond climate variability, include land degradation, following subsistence fallow agriculture, and demographic pressures on the resource base. Indeed, the increase in human population has increased demand for food, putting additional pressure on natural ecosystems. Non- traditional crops such as maize, sesame and soya beans have gained value in the last ten years, which has enabled farmers to make a choice of what type of crop to grow depending on demand, thus improving their incomes. However overall, agricultural performance fluctuates with climate variability and climate change, and is adversely affected by rudimentary means of production, poor markets and storage facilities. Any serious efforts for poverty eradication in Uganda will have to target the agricultural sector, and build its resilience to withstand climate-related shocks. Fishing Fishing is a key sector in the Ugandan economy, representing over 80million USD in 2002, as well as source of food for the population. It contributes to food security, increased household income and economic growth. It is estimated that over 200,000 people, most of who poor men and women, are directly involved in the fishing industry. As such the industry is particularly sensitive to changes in sea level and sea temperatures, driving associated modifications in upwelling mechanisms and in the composition of marine ecosystems (and consequently fish availability). Health 216

Uganda is characterized by high vulnerability to water-borne epidemics linked to flood disasters (malaria and cholera chiefly). Temperature rise would have significant impacts on health as well as agriculture. People in the highlands have not developed immunity for malaria and have therefore been susceptible to it. 7.1.4 Projected Climate Change impacts, across timescales, across sectors Climate Change, which has started manifesting itself through increased frequency of extreme weather events, such as droughts, floods and landslides, poses a serious threat to Ugandas natural resources, as well as its socio-economic development. The events of the past few years clearly illustrated the magnitude of the problem. Although rigorous and detailed vulnerability and adaptation options were not done for Uganda, the literature review of empirical information and observations by the communities surveyed during Ugandas NAPA participatory rural appraisal process has given interesting results280.

Fig. 7.6: Potential impacts of temperature rise on coffee growing

280 Uganda NAPA, 2007 217

Current temperatures and rainfall permit the cultivation of coffee in most parts of Uganda. However, an increase of 2 degrees Centigrade can have significant impact on coffee growing (fig. 7.6). Other crops like cassava and soya may be sensitive to temperature increases. Increase in temperatures may lead to emergence of new pests. There is, therefore, need to orient and widen the research focus to meet future challenges. A changing climate is projected to impact the following variables and sectors most decisively: Table 7.2 summarizes the observed and projected changes on each variable.
Observed Changes HMDs Increase in drought frequency. 7 droughts experienced between 1991-2000 (fig. 7.7-8) W estern, northern and northeastern districts experiencing long and increasingly frequent droughts Increase in climate variability More intense and

Hydro-Meteorological disasters (HMDs) Temperature Rainfall & water resources Highlands

Temperature Previously malaria-free highlands are now invaded by malaria Higher yields of potatoes, sorghum, maize and wheat reported in the highland ecosystems of Mbale, Kapchorwa, Kisoro and Kabale

Rainfall Rainfall seasons have become more variable Observed heavier and more violent rainfall Erratic onset and cessation of rainfall seasons in recent years

Projected Climate Change impacts

Increase of 2C likely to have dramatic impact on coffee growing areas

Highlands Increase in intensities and frequency of heavy rains, floods, landslides in highland areas and outbreaks of waterborne diseases. Decreased yield of certain crops (Arabica coffee), rise in the yields of others (potatoes, sorghum, maize and wheat) Increased Rwenzori Mt.s rainfall in Uganda permanent ice Consistent with cap vulnerable to IPCCC prediction global warming that wetter areas

218

(see fig. 7.6); will become also applies to wetter. other crops. Likely Temperature exacerbation of rise likely to lead water scarcity, to new particularly in pests/diseases & semi-arid areas change in crop Projected flood growing areas increase may Possible higher pollute sources of yields for crops drinking water such as potatoes, and lead to with outbreaks of indeterminate waterborne growth habit, diseases (cholera, provided rising typhoid, Ts do not exceed dysentery), with optimum T for the poor being crop most affected development Projection Low Robust for Low Low Confidence coffee; Low level otherwise with large remaining uncertainties Table 7.2: Observed and Projected climate changes in Uganda. Source: NAPA, 2007

frequent extreme weather events (drought, flood, landslides and heat waves) Dry conditions and more frequent, prolonged droughts frequently likely to lead to outbreaks of forest fires

219

7.2 Hydro-Meteorological Disasters and Government Response: 1995 2010 7.2.1 Main climate hazards in Uganda: 1995-2010 Uganda is most affected by droughts followed by storms, floods and landslides which are confined to highland ecosystems. Droughts (frequent and prolonged) represent the highest percentage of total disaster impacts in Uganda- about 36%281 . The most drought-prone areas are the districts in the Cattle Corridor. In extreme cases, particularly in the Karamoja region (North-east), frequent failure of the rains leads to starvation. Severe drought results in human and livestock deaths, as well as a reduced water table and dwindling water levels in the major lakes and crop failure282. The next most frequent disaster is flooding. Floods in Uganda are seasonal and usually occur during periods of intense rainfall as well as on el-Nio years. They are associated with death and injury to human population. Besides causing death due to drowning, floods destroy public health facilities such as water sources and sanitation facilities. Floods also trigger outbreaks of water-borne diseases and malaria, hence compounding community vulnerability to health hazards. They cause physical damage by washing away structures, crops, animals and submerging human settlements283. Direct loss of lives is particularly high in the highlands where floods and landslides merge to create human tragedies. At national level, during the 1997/8 El Nio event, about 1,000 people died in flood-related accidents, and 150,000 were displaced from their homes. In Mbale and Kapchorwa, a further 60 people were injured and 33 281 Uganda NAPA, 2007
282 MRDPR, 2010 283 Ibidem

220

died. 1998 floods were followed by severe drought in the western region, with Mbarara being the most affected. The rest of the country had good seasons. In 2000, Uganda again experienced widespread drought. Drought was more severe in eastern and northern regions with Arua district being the most affected, while western region experienced good seasons284. Analysis of rainfall variability over the years clearly indicates that not all parts of the country are affected by drought at any given time. Therefore overall impact of climate variability and climate change can significantly be mitigated through provision of climate information and promoting its utilization so as to take advantage of good seasons in some parts of the country285. Figure 7.7 shows the alternation of dry and wet years in Uganda, variability in Ugandas climate system that has become more frequent in recent years. Floods are also becoming more frequent, a prime example being the Teso region, which received its heaviest rainfall in 35 years in 2007. At that time, floods and food insecurity due to harvest loss affected an estimated 50,000 households. In March 2010, some parts of Eastern Uganda also experienced unusually heavy and prolonged rains that resulted into floods and landslides. For example, in Butaleja District floods submerged crop fields and vital infrastructure, including some schools and houses286 . Finally, landslides, storms (rain, wind and hail) and thunder are major hazards that cause significant damage when they strike. The areas mostly affected by landslides are the Mt. Elgon, Rwenzori region and Kigezi regions. In March 2010 a tragic landslide occurred in Bududa on the slopes of Mt. Elgon, burying three villages and

284

285

Uganda NAPA, 2007 Ibidem

286 UNDP, 2011

221

killing 81, including 35 pupils who could not walk to school, and destroying property. In 2005, hailstorms hit in Rakai, resulting in three deaths and displacement of 2,800 people. Violent windstorms also claimed many lives when boats capsized on lakes. In one incident, 21 people lost their lives when a boat traveling to Kalangala capsized on Lake Victoria.

Fig. 7.7: Regional dry and wet years in Uganda, 1943-1999. Source: State of the Environment Report 2000-01 in NAPA, 2007.

7.2.2 A Rising HMD trend In recent years however, the number of HMDs recorded in Uganda has risen exponentially, by a staggering 7.7% positive change in average number of disasters from the 1960 1994 period (when the average number of HMDs to hit Uganda was 222

0.8) to a 6.5 average number of HMDs during the 1995 2011 period. Of all the countries in Africa surveyed, Uganda records the most exponential rise in HMDs from 1960-1994 to the recent period of 1995 - 2011287. Figure 7.8 and table 7.3 confirm our initial intuition to focus on the period of 1995-2010 to study the impacts of more frequent disasters in the country. 1960 -1994 1995 -2011 HMDs in Uganda period period Drought 3 6 Flood 0 16 Storm 0 4 Total 3 26 Average for period 0.75 6.5 Source: EM-DAT (1960 2011) Total 9 16 4 29 3.6

Table 7.3: Exponential rise in selected HMDs in Uganda between 1960 1994 and 1995 2011

Fig. 7.8: Drought occurrence in Uganda throughout the 20th century. Source: Department of Meteorology, in NAPA, 2007.

287 CRED, 2011

223

7.2.3 Understanding HMD impacts in Uganda 7.2.3.1 In 1995-2010, impacts of rising disasters at both the national and community-levels have been felt heavily across Uganda. Annex 7.1 details the impacts of HMDs that hit Uganda during our period of study, in terms of lives lost, property destroyed, population displaced or otherwise affected by flood, drought, famine and landslides.

Impacts at national level

7.2.3.2

Zooming in: Impacts at the community level and priority community CCA-DRR needs

In Uganda, the 2007 NAPA, which rested heavily on Participatory Rural Appraisal (PRA) methods and extensively collected community perspectives on CCA across the country, offers a very good initial snapshot of community DRR-CCA needs. In the communities sampled during the NAPA, table 7.4 displays the disasters reported as most recurrent (rank-ordered), and their associated main impacts.

224


Table 7.4: Major impacts of HMDs in communities surveyed by the NAPA. Source: NAPA, 2007.

The NAPA study concludes that in the communities sampled, the priority community adaptation needs and intervention areas are as follows (rank-ordered): 1) Indigenous Knowledge documentation and awareness creation288; 2) Farm forestry; 3) Water resources management; 4) Weather and climate information provision, access and use; 5) Policy, legislation and planning; 6) Land and soil management; 7) Disaster preparedness; 8) Alternative livelihoods; 9) Health; and 10) Infrastructure. The NAPA study remains shallow nonetheless in its appraisal of community vulnerabilities in relation to their already existent capacities to cope with HMDs, and identification of what outstanding CCA-DRR needs are, above existing local capacity to cope. Through our community fieldwork conducted in Uganda from October 24 30th 2011, we investigated these precise needs in relation to local capacities to cope. In the following section we present the critical community resilience-building needs that have emanated from our community climate Vulnerability and Capacity Assessments (VCAs) to address rising HMDs.
288 Communities used to know their local climate relatively well and indeed relied on this knowledge for

planning of their farming activities. Knowledge of local climate was augmented by indigenous knowledge such as appearances of specific bird species, sprouting of particular plants and flowers. NAPA, 2007. For a more in-depth study of the value of local / indigenous knowledge in CCA, see Kniveton et al., forthcoming.

225

Case selection criteria:

Bulambuli district, in Center-East Uganda, one of Ugandas newest districts, was selected as the site for our community vulnerability assessments to impacts of HMDs. Bulambuli is hit by all three major hazards that afflict Uganda: drought and floods in the lowlands, as well as deadly landslides in the highlands. Bulambuli is found at the foothills of Mount Elgon, 30km away from the commercial hub of Mbale town, capital of the Bugisu sub-region.

Fig. 7.9: Bulambuli in Uganda. Source: Wikipedia Commons

Until July 1st 2010 when a parliamentary instruction mandated its split289, Bulambuli used to be part of the district of Sironko, which in turn composed the larger district of Mbale, with which it still shares the same common attributes. With a total population of 955,247 and a population growth rate of 3.4% per annum, it is one of the most densely populated zones in the country. Over 90% of the areas population live in the rural areas and rely heavily on the natural resources base. The climate in the region is humid tropical with a fairly uniform temperature throughout the year. Average rainfall around Mbale town varies from 880 to 1,775 mm per year with a mean of 1,186 mm. The average rainfall increases to 2,000 mm around Mt. Elgon, where all of the rivers of the region originate.

289 In 2009, Ugandas Parliament voted to create six more new districts, which became effective on July 1, 2010

(New Vision, January 1, 2010). Many purport that the creation of the new bill was to secure additional wins in the areas (turned into new districts) where Museveni had a sure majority (Expert interview, October 2011).

226

Agriculture is the principal economic activity, engaging over 77% of the adult population compared with a national average of 72%. Most agriculture in the region is centred on maize and beans as principal crops: but, with the intervention of the National Agricultural Advisory Services (NAADs), there is a shift towards commercialised agricultural production of high-value crops such as carrots, potatoes, arabica coffee, and horticultural crops in the upland areas; and rice, bananas and coffee with horticultural crops in the lowlands. Under its former denomination, Mbale districts economic growth stalled during the period of civil strife in the 1970s-1980s. The region boasted high-quality coffee production in the 1960s and early 1970s. The coffee industry has suffered from climate variations, in particular higher temperatures and unpredictable rainfall, and the decline of global coffee prices. Nonetheless, high-quality Fair Trade coffee is produced by some farmers on the slopes of Mt Elgon. Most of the funding for local government plans (nearly 98%) comes from central government in form of conditional grants. These include the Plan for Modernisation of Agriculture (PMA), the National Agricultural Advisory Services (NAADS) and the Local Government Development Programme, and Poverty Allevation Funds (PAF). PAF funds are for central government priorities focused on poverty alleviation. Local Government Development Program funds mainly support government priorities, which are education (construction of schools, teachers houses, furnishing schools), health/sanitation and road construction. With 79% of the regions population living in the rural areas and 21% in the urban, the regions leadership is convinced that agricultural growth is essential for economic growth, particularly as it will stimulate growth in other sectors and improve food security, employment and poverty alleviation290.
290 Mbale Region draft Climate Change framework Plan, 2009

227

We selected five drought, flood and landslide prone communities in the low and highlands of Mount Elgon in Bulambuli as target sites for our VCAs, to represent the diversity of vulnerability contexts in Bulambli. Table 7.5 describes the communities where VCAs took place in Bulambuli district. Community Hazard Type Name 1 Bufumbula Flood 2 Wanga 3 Busangai 5 Namungu Flood Bufumbula Nabongo Bufumbula Bufumbula Bulambuli Bulambuli Muyembe Bulambuli Parish Sub-county County District

Drought/Flood Buwebele Landslide Luzzi

Bunambutye Bulambuli Bulambuli Bunambutye Bulambuli Bulambuli Bulambuli Bulambuli Bulambuli

4 Bunambale Drought/Flood Buwebele

Table 7.5: Communities at risk in the district of Bulambuli selected for climate VCAs, Center east Uganda, October 2011

Critical community adaptation needs emanating from our VCAs:

Community where intervention is needed Bufumbula Wanga Busangai Bunambale Namungu (Luzzi) X X XX X X X Priority Intervention
Relocation to lower-risk zones Support for tree replanting, with provision of seeds and training on tree replanting techniques Dig trenches to divert floodwaters Support to build strong permanent houses resistant to heavy rainfall and landslides Build water sources (eg: dig borehole) / protect existing water sources Scale-up local irrigation from river

228

Food security program with food storage support and income diversification IGA/ Credit access to secure income during climate disasters when agricultural income is affected (gender-specificity) Agricultural support to: - access and plant drought/flood resilient seeds (fast maturing crops/drought-resilient crops) - access advisories for early planting Support local irrigation using river water Lightening protection/shelters Early Warning Medicine to health center Sensitization and awareness on hygiene and sanitation practices

X X

X (drought resilient)

X X

X X X

X X

Support with building of X latrines Table 7.6: Aggregation of common priority CCA-DRR needs across 5 target communities

Farmers surveyed confirmed that rainfall is already becoming more intense. The particularly heavy rains of late 2007 across northern and central Uganda may be an omen of a future climate regime. As such, future generations may see a switch away from the current pattern of two rainy seasons across much of the country towards a new pattern, of a unimodal and much more pronounced rainy season (with heavier rains) later in the year, with the rest of the year hotter and drier than at present. Climatic changes are thus happening in Uganda. On the one hand there is more erratic rainfall in the March to June rainy season, bringing dry spells and reductions in crop yields and plant varieties, on the other hand rainfall, especially in the later

229

rains towards the end of the year (June-November) is reported as coming in as more intense and destructive downpours, carting along floods, landslides and soil erosion. 7.2.4 National CCA-DRR Policy Analysis In the face of the dramatic impacts of HMDs in recent years, both at the national and community levels, what is the Government of Uganda (GoU) doing? What measures has it adopted to address rising climate risks? How has it organized itself to meet the CCA-DRR challenge? How adequate are these measures to meet the priority CCA-DRR needs of base communities, affected by HMDs year after year? This section attempts to understand and characterize the national climate disaster management architecture of Uganda. We had concluded from our cross-continental mapping that Uganda was in the yellow policy group: is this verified in practice? 7.2.4.1 In Uganda, the primary responsibility for disaster risk management in Uganda rests with citizens. Government plays a supportive role291. Paragraph XXIII of the National Objectives and Directive Principles of State Policy of the 1995 Constitution obliges the state to institute effective machinery for dealing with any hazard or disaster arising out of natural calamities or any situation resulting in the general displacement of people or general disruption of normal life. Aware that disasters disrupt the productive capacity of her people, destroy the infrastructure and resources, divert the planned use of meager resources and so interrupt programs and retard the pace of development, government is therefore, committed to creating and promoting a disaster preparedness and management system that safeguards the country against disasters and ensures continued productive capacity of her people.
291 National Policy for Disaster Preparedness and Management, 2010

Introspection into Ugandas Climate Disaster Management Policy:

230

The States responsibility to institute an effective machinery for dealing with any hazard or disaster arising out of natural calamities has not be delivered homogeneously across time, however. Before 1999, Ugandas policy towards HMDs has mainly been that of response to disasters; response that was inefficient and delayed, in practice a Firefighting country (refer to Annex 7.1, HMDs before 1999). In 1999 however, the Ministry of Disaster Preparedness, Management and Refugees was established, with a Department for Preparedness, under the Office of the Prime Minister (OPM). DP/OPM was mainly established with a mandate to address the massive internal refugees problem following the end of the civil war. This new institution was to take charge of all affairs related to disaster management. In 2010, after much drafting and re-drafting, Uganda adopted its National Policy for Disaster Preparedness and Management, an ambitious national plan to holistically prepare for and manage the multiple hazards hitting Uganda. This plan offered a comprehensive blueprint for planning and coordinating disaster operations across multiple sectors and agencies, in collaboration with all government ministries, humanitarian and development partners, the private sector, local governments and vulnerable communities. The objective of this plan was to ensure that in times of disaster, appropriate administrative, legislative and technical measures including procedures and resources are in place to assist those afflicted by the disaster and enable them to cope. In effect, this plan proposed a new institutional framework, or under Norths definition, new rules of the game, to manage climate-related risks in the country, defining the roles and boundaries of different organizations within Ugandas DM institutional landscape towards improved public delivery on national and local CCA- DRR needs. 231

What led to the initiation of this new institutional framework, the new rules of the game so to say? We get to this question in our next sections. Before we delve into the heart of our question, let us first attempt to understand the provisions of Ugandas new policy for disaster preparedness, and its adequacy to meet fundamental community needs in DRR-CCA. Provisions of the national policy: Initiated by the Directorate of Relief, Disaster Preparedness and Refugees (DRDPR/OPM), Office of the Prime Minister, mandated unit to address disasters in Uganda, the National Policy for Disaster Preparedness and Management (NPDPM) was approved in September 2010. The goal of the NPDPM was to establish institutions and mechanisms to reduce the vulnerability of people, livestock, plants and wildlife to disasters in Uganda. The specific policy objectives were to: i. ii. iii. iv. v. vi. vii. 232 Establish Disaster Preparedness and Management institutions at national Equip Disaster Preparedness and Management institutions and ensure that the Integrate Disaster Preparedness and Management into development processes Promote research and technology in disaster risk reduction Generate and disseminate information on early warning for disasters and Promote public, private partnerships in Disaster Preparedness and Create timely, coordinated and effective emergency responses at national, and local government levels; country is prepared at all times to cope with and manage disasters; at all levels;

hazard trend analysis Management. district and lower level local governments.

What is noteworthy about this policy is its recognition of the growing role of Climate Change in shifting disaster hazard trends, explicitly stating the connection between CCA and DRR and the need for disaster management to be an integral part of the development process, from the local to the national levels. It also recognizes the countrys hitherto poor efforts to address disasters efficiently. As detailed in the Policy document: Owing to meager resources, the administrative and technical measures necessary for disaster preparedness and management in the country are scanty and hampered by poor equipment. The country lacks a depository and repository of vital data on the incidence of disasters and their effects on the communities. Enforcement of legislation on risk avoidance is weak such that disruptions arising from disasters have continued to grow without corresponding measures. This policy, therefore, is an effort to put in place a systematic disaster prevention, mitigation, preparedness and management framework for the country (National Policy for Disaster Preparedness and Management, 2010) The National Policy most of all established clear rules of the game for the management of HMDs in Uganda, a social contract from the national to the very local levels. At the national level, the Ministry responsible for Disaster Preparedness and Refugees in the Office of the Prime Minister is the lead agency responsible for disaster preparedness and management in the country. It coordinates risk reduction, prevention, preparedness, mitigation and response actions in the country in consultation with other line ministries, humanitarian and development partners, Local Governments and the Private sector. The Minister responsible for relief, disaster preparedness and refugees links the Department to Cabinet, which directs policy and advises the President. On day-to-day and routine matters, the Minister makes rules and regulations on the management of likely disasters and presents annual reports relating to Disaster Preparedness and Management to Cabinet. The 233 Institutional framework for CCA-DRR:

Minister also links the Office of the Prime Minister to inter-governmental organizations, the donor community, the private sector, regional and international frameworks (OPM, 2011). The following flowchart depicts how disaster preparedness responsibilities are devolved and coordinated, as nationally mandated, from the supreme authority in the country, President, to the basic community level where HMDs occur, reflecting Ugandas conception of DM as a shared responsibility between the State and citizens.

234

 Fig. 7.10: Organization of Disaster Preparedness (2010 National Policy)

- The president: Article 110 of the 1995 Constitution gives the President the mandate to declare a state of emergency in any part of the country in the event of a disaster. The Minister in charge of disaster preparedness and management shall provide the President with all the relevant details on the cause and effects of the disaster and mitigation and relief measures to be undertaken; in a situation where the disaster is 235 Roles in the National CCA-DRR Institutional Framework:

caused by a natural or human-induced hazard. The President in consultation with Cabinet shall declare an area or the nation to be in a state of disaster. A declaration of a state of disaster shall mean the provision of the required relief services and goods to the affected population takes precedence over all other programs of government, thus calling for immediate re-programming and re- allocation of resources from other sectors until such a time when the President declares the end of the state of disaster in the defined part of the country. - The cabinet: Advises the president on all disaster related issues. - The Ministerial Policy Committee (MPC): This standing committee of Cabinet handles cross-sectoral matters relating to disaster preparedness and management. The committee is responsible for ensuring that disaster preparedness and management is mainstreamed in governance. Chaired by the minister responsible for relief, disaster preparedness and refugees, the MPC composed of the following ministers, whom have each been designated as lead institution in the advent of a specific disaster pertaining to their area of expertise: Minister of Relief, Disaster Preparedness and Refugees (Chair) Minister of Internal Affairs; Minister of Finance, Planning and Economic Development; Minister of Agriculture, Animal Industry and Fisheries (lead institution in advent of famine/food insecurity); Minister of Health (Lead Institution on Epidemics); Minister of Water and Environment (Lead Institution on Drought, floods and storms, as well as landslides through its subsidiary body, NEMA); 236

Minister of Defence; Minister of Education; Minister of Local Government; Minister of Gender, Labour and Social Development; Minister of Justice and Constitutional Affairs; Minister of Works, and Communications; Minister of Information and National Guidance Minister of Lands, Housing and Urban Development

The National Environmental Management Authority (NEMA), that does not sit on the MPC is the lead institution on Landslides and mudslides. - The National Emergency Coordination and Operations Centre (NECOC): The NECOC is responsible for the effective coordination and networking of the various emergency response institutions of government such as the fire brigade, Police Rapid Response Units, UPDF Emergency Support Units, Uganda Red Cross Society, hospitals emergency units and the private emergency firms. NECOC is established under the Office of the Prime Minister- Directorate of Relief, Disaster Preparedness and Refugees. Finally, at the community level, every district was mandated by law to form its District Disaster Management Committee (DDMC), to serve as the community platform for disaster preparedness and response. Across Uganda, many DDMCs have been established, but have yet to be rendered operational through capacity- building of district administrators and community members on the stakes of ex-ante disaster management and contingency planning (expert interview, February 2012).

237

7.2.4.2

Institutional Coordination and Coherence in the Arrangement of Organizations to Deliver on Climate Change Adaptation

Prior to the adoption of the 2010 national policy, no institutional coordination for effective disaster response existed. Disasters were responded to by the NECOC in a delayed, and ineffective fashion. Following the 2007 floods, it is related that State rescue services did not get to afflicted sites. The Ugandan Red Cross (URCS) was consistently the first organization on the disaster site292. The relative novelty of Ugandas new institutional architecture on CCA-DRR (2010) however hinders our ability to rate its effectiveness. Its concrete implementation at the local level, through the local committees for Disaster Management in every village, will in all likelihood take time to realize, with the support of donors and partners in DRR-CCA. However what we can state is that it is a major effort in the direction of institutionalization of better coordination and preparedness of HMDs, which lifted the blockages thwarting cross-institutional integration for improved delivery on CCA-DRR and gave a national mandate for disaster preparedness. 7.2.5 Overall Policy picture: Uganda, an Effective responder Thus, Uganda qualifies for all of the attributes we had initially hypothesized to characterize a Prepared Firefighter (see table 7.7). For the purposes of our study, we can then proceed to use the country case as an adequate representative of the Prepared Firefighter policy group. The only concern remains the relative novelty of the countrys effective transition to the Yellow policy group (begun in 1999, but effectively completed only in 2010). However institutional change being inherently incremental and a process, we are confident using Uganda, to understand the factors that turn a country into a Yellow case.
292 Expert interview, Oct. 2011

238

Indicator for Ex-post response+ ex-ante disaster preparedness (The Prepared Firefighters policy) A national Disaster Management Strategy or law adopted, with clearly defined and appropriated procedures and responsibilities for disaster response In the national budget a dedicated budget line exists for disaster preparedness and emergency response A mandated agency for disaster management exists, able to coordinate across ministries and administrations and effectively command disaster response activities during an Emergency An annual Contingency plan is elaborated, and delineates the procedures, actions and responsibilities to be carried out in the advent of all possible disasters Regular drills of these disaster response procedures take place No more than 3 days to deliver disaster relief to affected citizens Overall: Country in this category have shifted from response (responding to disasters only after they occur), to preparedness for disasters before they occur, so as to ensure more efficiency and swiftness in disaster response and relief operations
Table 7.7: Uganda, a Prepared Firefighter as of 2010

Present? (Present: Absent: X)

What led to Ugandas directional shift in the management of disasters? We now turn to this fundamental question.

239

7.3 Incentives Analysis: which factors prompted institutional change for effective CCA-in Uganda? After reviewing government institutions in place to address rising HMDs, we realize that the GoU has adopted bold measures since 1999 to improve the effectiveness of its response to HMDs. Now we turn to the central question of our investigation. What triggered such a change in government institutions towards HMDs in Uganda? What are the systemic enabling conditions? To answer this question, we first review government responses to the impacts of rising HMDs that hit Uganda during our period of study, proxy for impacts of a changing climate in this highly disaster-prone country of the African great lakes. We then proceed to retrace the process of response of the GoU to each HMD during the period 1995-2010, focusing on the identification of the critical break years and disasters during which a transition occurred in the public policies towards hydro- meteorological disasters. Finally, we end with an analysis of government institutional change towards HMDs during the period of study, when these began to increase. We derive from our analysis the factors that may explain breaks in the continuity of public policies and institutions towards the management of HMDs in Uganda. Our analysis rests on a thorough event analysis based on 115 newspaper articles that matched our keyword search on two of the countrys major newspapers (New Vision and Daily Monitor), catalogued in chronological order from 1995 to 2010. We then confronted the information from our events analysis with insights from iterative interviews with five experts on DRR-CCA in Uganda, who were pivotal actors in the countrys transition from Firefighters to Prepared Firefighters in the end 1990s. We also consulted secondary literature at great length, in search of areas of agreement and triangulation. Only the findings confirmed by both sources of information (events analysis and expert interviews) are presented in the sections following. 240

7.3.1 Process Tracing: Review of Government responses to HMDs, 1995-2010 A quick glance at the GoUs response to HMDs that made landfall with increasing frequency and magnitude during the 1995 2010 period (see Annex II, 7.2), reveals that three main moments characterize Ugandas response to rising climate risks: 1) Pre- 1999: Period of the Firefighters Response Indeed before 1999, the GoUs response to disasters was in practice that of a Firefighter. Its response to HMDs was disorganized, uncoordinated and characterized by amateurism and poor efficiency. A look at government response to disasters pre-1999 (see Annex II) proves this point adequately. Management of the 1994 earthquake illustrates the GoUs inadequate response to disasters. Not included in our log of HMDs in annex II (7.2) due to its non-climate related nature, the 1994 earthquake is still important to recall because of its impact on the countrys institutional evolution on the governance of HMDs. Indeed the 1994 earthquake marks the very first trigger of Ugandas awakening to the importance of a new and improved institution to manage disasters. A severe earthquake (magnitude: 6.2 on Richter scale) rocked many parts of Uganda on 6 February 1994, causing extensive damage to three western districts of Kabarole, Bundibugyo and Kasese. While casualties were limited to seven deaths, heavy damage was inflicted to private, commercial and public buildings (such as schools, churches and hospitals). As a result, many people were left homeless and forced into displacement. The government of Uganda requested the United Nations to assist in damage assessment, provision of relief and co-ordination of donor activities. The government set up a national earthquake disaster committee chaired by the Ministry of Labour and Social Affairs. Furthermore each affected district had a disaster committee. UN agencies, bilateral donors and NGOs collaborated with the government through and with the UN disaster management team, chaired by the UN Resident Coordinator and DHA representative. In practice, the Government stepped 241

aside and allowed the UN system, more efficiently organized, to conduct relief operations. This is reminiscent of the Senegal case. Shocked, however, by the sheer damage caused by this earthquake, the Government of Uganda decided to form a Commission to institute an effective machinery for dealing with any hazard or disaster arising out of natural calamities or any situation resulting in general displacement of people or serious disruption of their normal life. The 1995 constitutional revamping, the first of its kind under Musevinis rule, offered the opportunity to instate such a Commission, and provisions were included in the 1995 Constitution for a Disaster preparedness Commission. Chapter 17 Article 249 in the 1995 Constitution thus stated that" There shall be a Disaster Preparedness and Management Commission for Uganda to deal with both natural and man made disasters"293. Its formation was due to the need for an effective institution to deal with hazards and disasters, as there was none in place in the country prior to its formation. However, the Commission was never operational although there was a provision for it in the Constitution. One of the reasons for this was that the composition of the Commission was top heavy that is, it had too many partners from different departments and ministries and the costs involved were exorbitant294. This first attempt at institutional reform in 1995 in favor of improved disaster preparedness thus died in the shell. It is noteworthy that this period corresponded to a relative lull in the occurrence of disasters, between 1994-5 to the 1998 flood event (see Annex 7.1). This correlation seems to indicate that the loss of momentum may have been driven by the lack of disaster occurrence, and hence lack of urgent need for the establishment of decisive new institutions to manage HMDs.
293 Uganda Constitution, 1995 294 Expert interview, February 2012

242

2) 1999: New institutional agreement to better manage Disasters In 1999 however, following the 1998 severe flooding and 1999 droughts, another momentum for institutional change came. In 1999, Cabinet approves a National Disaster Preparedness Policy and Institutional Framework, to be later revised in 2003. This led to the creation within the Office of the Prime Minister (OPM) of a Ministry of Disaster Preparedness, Management & Refugees, with a Deputy Minister for Disaster Preparedness, coordinating the Department of Disaster Preparedness and Management (DDPM). The DDPM was formed in 1999 after a merger of two departments: the Department of Relief and Resettlement (under Ministry of Labour and Social Affairs) whose main role was to provide relief to and resettle citizens from over-populated areas (Kabale, Rukungiri, Bushenyi, Busia in Western Uganda) to less populated areas within the country; and the Directorate of Refugees and Resettlement (under the Ministry of Local Government) whose main role was to provide relief to and resettle victims and displaced populations from the LRA war in Northern Uganda and ADF rebellion in Western Uganda. It also provided relief to refugees within the country. However, the two departments had the same mandate although they were in different ministries and this at times created duplication of activities and interventions. In order to ease planning and as a result of the roles compatibility, the Permanent Secretaries in the two ministries sought to combine both departments to form the Department of Disaster Preparedness and Management (DDPM) with a Minister for Relief, Disaster Preparedness and Refugees. Its elevation to a department under the Office of the Prime Minister created the conditions for cross- ministry coordination on the issue. Many purport that the first successful attempt at institutional change of 1999 mostly ensued as a response to the massive man-made disaster of the 1990s in Uganda: the civil war with LRA in the Northern region of Acholi, which displaced 1.2 million, 243

creating about half a million refugees to be dealt with. This was a monumental national problem that the State was ill equipped to deal with. An efficient cross- ministerial framework was needed to handle the surge of internal refugees. It was in response to this initial national demand that the DDPM was created295. The title of Refugees that still lingers on to the Departments name is indicative of the nature of the initial impetus that led to its creation. In addition Uganda has in place a National Policy on Internal Displacement of Persons and is writing a National Policy on Conflict Resolution and Peace Building296. Today, the war is over, and most if not all IDPs have returned to their initial districts of origin. The refugee problem is in effect solved in Uganda. However the institution, which confirms the hypothesis of path dependence, in this case the accident of nature that led to the creation of the institution being the very unfortunate Nothern Uganda war, and its host of refugees, but which, in turn, set Uganda on a track of holistic management of large-scale national disasters, bringing to bear all o the sector expertise and ministerial abilities in the nation. The United Nations also played a significant role in the institutional shift of 1999. The UN system in Uganda played a key support role, notably through UNDPs Bureau for Crisis Prevention and Recovery (BCPR)297. The 1990s also coincided with the UN Decade on Natural Disasters, a global initiative that actively advocated across all countries of the world in the 1990s for a paradigm shift from response to natural calamities to the prevention of disaster risks and efficient disaster preparedness for a changing climate. 3) 1999-2010: Consolidation era: changes and adjustments to the new DM institution
295 Expert interviews, October 2011 296

DRDPR, 2004

297 Expert interview, October 2011

244

The National Policy drafted with UNDP BCPRs support has gone a long way though since 1999. The ratification process of the policy was a long odyssey lined with multiple successes and roll-backs (expert interview, October 2011). Indeed, in 2002-03, the country, still coming to grips with the large-scale refugee problem bequeathed by civil war, was still focused on IDP policy. This sidelined the Disaster National Policy. The Policy draft still continued its course however, fortifying its provisions in the meantime. In 2007, the draft Policy was submitted to Cabinet for approval. However it was rejected, on the grounds that Uganda needed to first learn more about how Disaster Preparedness and prevention should be done well, in light of Ugandas weak institutional capacity on the issue. The policy draft was thus returned to its drafters with a solid list of recommendations. The DDPM was indeed being sent back to the drawing board298. Following the 2007 rejection of the Policy, both the Disaster Minister and his Deputy, along with a team of technical advisers from the DDPM, travelled to Bangladesh on exchange visits to better understand how that small Asian country, also highly exposed to multiple HMDs, succeeded in propping up its national preparedness to HMDs and establishing effective early warning systems able to prevent losses and destruction from HMDs. It can also be hypothesized that the severity of 2007 floods followed by the 2010 landslides which killed more than 300 people, and which triggered national states of emergency both times, played a key role in bringing back on the political agenda the urgency for a National Policy of Disaster Preparedness, focused on HMDs this time. Thus, in 2010, following many years of drafting and redrafting, the Government took it upon itself to seriously fulfill its constitutional responsibility for establishing an efficient machinery for dealing with natural hazards in the country and preventing loss of life. It posed two significant acts to this effect.
298 Expert interview, October 2011

245

 Creation of the National Emergency Coordination and Operations Centre (NECOC) First off, it created the National Emergency Coordination and Operations Centre (NECOC). This institution was created to coordinate disaster relief and disaster operations. Its creation was a significant addition in an institutional landscape where relief and response to HMDs were still disjointed and poorly coordinated. This was the case in 2007, for instance. During the severe flood disaster of 2007 that disrupted the lives of 400,000 people, the State, overwhelmed by the extent of the disaster, responded in a very disorganized fashion. As one of interviewed expert who was a key actor in the 2007 relief effort depicts, many districts in the Teso sub- region were hit (Amuria, Katakwi, Soroti, Kaberamaido). There was no loss of life but 20,000 people were affected there, with roads were cut off. When the state of emergency was declared, big international disaster people from Kenya Red cross came in, UNICEF brought in tents for children, UN planes came in to drop off food. Many agencies came in donations were received through Uganda Red Cross. However there was no real coordination of the relief effort, which led to much inefficiency. No real procedures existed to be followed. Coordination of relief efforts was ensured by OPM with UN OCHA; however it was blatant that no well-rehearsed pre-existing procedures were there for different intervening agencies to follow299. Indeed, in the 2007 floods President Yoweri Museveni declared a 90-day State of Emergency in the flood-hit parts of eastern and northern Uganda. The Government and partners launched a US$ 41 million Flash Appeal for the Floods. Some US$ 6 million in Immediate funding for the Appeal was made available from the United Nations Central Emergency Response Fund (CERF). Over the next three months, the UN clusters delivered life-saving emergency assistance to those affected by the flooding, achieving success in preventing loss of life, malnutrition and disease.
299 Expert interview, October 2011

246

Emergency responses by the Uganda Government and humanitarian agencies included provision of temporary shelter, food, drinkable water, sanitation facilities, medicines, and even helicopters and boats to assist those in hard to reach areas. Six camps were created for Lira flood victims. The United Nations chartered a helicopter to airlift emergency supplies to areas where floods rendered roads impassable. The World Food Programme (WFP) reached 129,152 flood-affected people, 39% of targeted beneficiaries, with 2,131 metric tonnes (MT) of food aid to date. Some 3,000 beneficiaries received food aid as air operations resumed out of Soroti on 10 October. As of December 2007, the emergency response has been replaced by longer-term recovery and disaster preparedness programming across the flood-affected area. From all this clearly appears that the relief effort in 2007 was driven mostly by the UN system, and the State, overwhelmed, abdicated and let the UN, better organized and more capacitated, take over. Thus despite the 1999 institutional reform, coordination and capacity issues remained during disaster response, as evidenced by the governments stunted response when faced with the massive disaster of the 2007 floods. Uganda was not quite prepared enough to handle large-scale HMDs. Following the 2007 floods fiasco in disaster response, under the impetus of UN OCHA, the NECOC was created to improve disaster coordination before and during disasters, and mandate a Government leader, to serve as the maestro during disaster operations. UN-OCHA also played a large role in bringing about the NECOC300. Still a relatively new institution, the NECOC has yet to show its true mantle. But its existence already bodes of brighter days to come for Ugandas efficient coordination of disaster response.
300 Expert interview, October 2011

247

 Adoption of the National policy for disaster preparedness and management Finally, in 2010, Ugandas cabinet adopted its National policy for disaster preparedness and management, an ambitious national plan to holistically prepare for and manage the multiple hazards hitting Uganda. This plan offered a comprehensive blueprint for planning and coordinating disaster operations across multiple sectors and agencies. The objective of this plan was to ensure that in times of disaster, appropriate administrative, legislative and technical measures including procedures and resources are in place to assist those afflicted by the disaster and enable them to cope301. In effect, this plan proposed a new institutional framework, or under Norths definition new rules of the game, to manage climate-related risks in the country, defining the roles and boundaries of different organizations within Ugandas DM institutional landscape towards improved public delivery on national and local CCA- DRR needs. We can thus claim that Uganda only achieved full transition to a Prepared Firefighter in 2010. The transition process lasted a total of 11 years, from 1999 to 2010. 7.3.2 Analysis: Ugandas pathway of Institutional Change From our analysis of the events that led to substantive institutional change towards effective Disaster Preparedness in Uganda, emanates that Ugandas institutional transition was a slow and staggered process, which occurred in response to a pressing national demand/need.
301 NPDPM, 2010

248

It can be further asserted that Ugandas was a two-stepped phased transition towards effective disaster preparedness. Step one occurred in 1999 in response to the national refugee crisis, whereas step 2 waited until 2010, when the large-scale impacts of rising recurrent HMDs (floods, droughts, landslides) on communities and national welfare could no longer be ignored. Step 1: 1999, first successful institutional change in Ugandas disaster management infrastructure with the Cabinets approval of a National Disaster Preparedness Policy and Institutional Framework and the creation of the Department of Disaster Preparedness and Management (DDPM) with a Ministry of Disaster Preparedness, Management & Refugees, under the Office of the Prime Minister (OPM) However this change in the rules of the game of disaster management was in practice restricted to the management of war refugees (the war disaster). Uganda thus did not fully turn into an accomplished prepared firefighter right after 1999. It still was not capacitated to address multiple risks, notably HMDs, as poor public response to the massive 2007 floods demonstrates. However, 1999 marks the moment of institutional revolution, and triggered the process of change to the institutional framework on CCA-DRR in Uganda, which was to be completed over the next decade. Step 2: 2007-2010: re-drafting of National Disaster Preparedness policy and approval by Cabinet We indeed had to wait until 2010 to see Uganda substantively take measures to place itself comfortably in the Prepared Firefighter category, notably with the creation of its National Emergency Coordination and Operations Centre (NECOC) and adoption by the Cabinet, in September 2010, of the National Disaster Preparedness and Management Policy. 249 Trigger: War refugees disaster (at the end of civil war in N. Uganda) Support for formulation provided by: UNDP BCPR

Trigger: Severity and magnitude of HMDs in the decade of the 2000s (2007 floods, 2010 landslides) Support: UN-OCHA

2007-2010: Incremental changes Revision of National Policy for Diaster Preparedness & Management

2010: Uganda becomes a Prepared Firevighter NECOC created, National Policy for Diaster Preparedness and Management adopted

1999: Ministry of Disaster Preparedness & Refugees created, under Ofvice of Prime Minister

1995: Constitutional provisions for a Disaster Preparedness &Management Commission, but never materializes

Fig. 7.11: Ugandas transition to become Prepared Firefighter: a long and phased process

7.3.3 Identification of relevant Independent Variables What we retain is that both spurts of institutional change in Ugandas management of disasters first building capacity to handle war refugee disaster, second building preparedness for HMDs were driven by a national need. Indeed, both phases in Ugandas transition towards becoming a Prepared firefighter responded to an urgent national problem, which the country had to deal with, or perish. The institutional solution devised endogenously by the GoU was driven by the existence of a need. 250

As such, our Uganda case confirms the finding from the Senegal case that the most critical triggers of institutional change on CCA-DRR are: 1) The occurrence of more frequent high-impact disasters; 2) The rise in the frequency of HMDs in turn led to demands for a new climate risks/disaster management institutional framework. This became a national need (country-driven-ness); 3) Donor and development partners technical and financial support to countrys desire to formulate a new national institutional framework. Both Senegal and Ugandan cases thus appear to suggest that the country-driven- ness of institutional change is fundamental, and that key is a countrys appropriation of the desire to adopt effective rules of the game to govern the new issue of Climate Change. This desire under both instances is, itself, driven by the sheer increase in HMDs, which places CCA-DRR, and the search for more effective institutional arrangements to manage and prevent the impacts of HDMs, as a priority on the national agenda. National leaders first had to experience the effects of not addressing HMDs properly and not climate-proofing development grains from the rising impacts of climate risks (climate-compatible development), before they began to offer seriously engage with the battle of HMDs. In the context of Ugandas hypertrophied Executive branch, where the same president, Yoweri Musevini, has been in power since 1986, this finding is not surprising. In a system revolved around one-man rule, the one leader and executive manager of the patrimony needs to perceive the dangers and risks to the patrimony caused by HMDs, before he can react decisively and place the issue high on the national agenda. In Uganda, in turn we purport that national will to tackle rising climate risks was built by the rise in frequency and severity of high-impacts HMDs floods, droughts and landslides in the 2000s decade, which increased by a staggering 7.7% relative to the 1960 1994 period. 251

Also in Uganda, when they decided to become serious on HMDs, those in charge of drafting the National policy even had to go and seek outside expertise to learn how to do this right and what institutions to devise to rightfully handle the new nation issue. As we will see shortly, this pattern was also replicated in Mozambique, where multiple HMDs also took the country by surprise, and the State was left with no choice but to confront the problem, and devise new institutions or rules of the game, to handle multiple HMDs. When given the choice between Adapt and Perish, these countries chose to Adapt. This determines a course of gradual institutional adaptation as countries are increasingly faced with rising climate-related risks and disasters. It bodes negatively though of countries abilities to engage in ex-ante action until the projected climate impacts have started to hit them. This theory claims that countries in Africa, and elsewhere, will not take climate impacts seriously until these begin to manifest themselves. They will not learn form the age-old wisdom that prevention is better than cure. However when the projected predicament will hit them, then countries will scramble to get on their feet and seriously begin to tackle the problem, finally implementing all of the known recommendations for addressing climate impacts. Faced with the ultimate choice of Adapting or Perishing: they will finally choose to Adapt. However, not before the ultimatum gets down to that. Will Uganda then sustain this course, and continue its transition up the spectrum of effective disaster risk management all the way to the green Disaster averter category? Only time will tell. If more and frequent intense disasters occur (which we do not wish for this beautiful country), and Uganda intensifies it efforts to transition towards ex-ante prevention on top of its existent prearedness, then our theory will be proven. 252

We now turn to our investigation of the factors that make countries transition to the final category in our spectrum of effective climate disaster management: the Disaster Averters. Are there any additional identified causal factors, beyond national drivenness to address HMDs and donor support with this endeavour?

253

 CHAPTER 8 MOZAMBIQUE COUNTRY CHAPTER:

What Triggered Institutional Change for Effective Climate Governance in Mozambique? 8.1 Country Profile The Republic of Mozambique is a thin, elongated stretch of land situated on the oriental coast of Africa in the southern hemisphere, between longitudes 3012 4051 and latitudes 1027 1652. Covering a total land area of 799.380 square kilometers (km2), of which 786.380 km2 of firm land and 13.000 km2 of interior waters, Mozambique is a coastal country, which boasts of a two thousand five hundred kilometer (2,500 km)-long ocean interface. This elongated interface with the Indian Ocean, right at the Mozambique Channel separating the country from Madagascar, renders Mozambique highly vulnerable to multiple HMDs, among which drought, flooding, sea level rise and recurrent cyclonic activity. It is the third country in Africa most vulnerable to weather-related hazard302. Mozambique is endowed with a tropical climate with two seasons, a wet season typically from October to March and a dry season from April to September. Climate conditions vary, however, depending on altitude. Rainfall is heavy in the highlands and along the coast and decreases in the north and south. Annual precipitation varies from 500 to 900 mm depending on the region, with a national average of 590 mm. Bordered in the north by Tanzania, in the West (from north to south) by Malawi, Zambia, Zimbabwe, South Africa and Swaziland, and in the South by South
302 Global Facility for Disaster Risk Reduction (GFDRR), Disaster Risk Management Programs For Prioritm

Comntries Africa- Mozambique, 2011, 97-105

254

Africas KwaZulu-Natal Province, the country is administratively sub-divided into 11 provinces, which are in turn each subdivided into 128 districts (see fig. 8.1). The country also boasts of many cities, among which: Maputo (capital city and province of its own), Beira, Nampula, Nacala, Chimoio, Quelimane, Inhabane, Tete, Xai-Xai, and Pemba303.

Fig. 8.1: Map of Mozambique in Africa, and its 11 provinces. Source: Mozambican Ministry of Foreign Affairs and Cooperation, 2011.

8.1.1 Vulnerability Context 8.1.1.1 Mozambique is a country of 23 million inhabitants classified as low-income, with a GDP/capita of $428 and a strong annual GDP growth rate of 6.3%, which has been sustained over the past two decades304. One of the poorest countries in the world after its independence in 1975, Mozambique has emerged from two decades of armed civil conflict as one of Africas best-performing economies, achieving an average annual rate of economic growth of 8% between 1996 and 2008, the highest Development challenges: past & present

303 Mozambique National Directorate of Statistics, 1994 304 World Development Indicators- Mozambique, 2010

255

growth rate among African oil-importers305. Mozambique remains a poor country with a low level human development, and lies at the very bottom the human development ladder, classified 184th out of 187 countries ranked by the United Nations Development Program Human Development Index ranking306. A country devastated by civil war until 1992: At the end of its civil war in 1992, Mozambique was left with decimated infrastructure, a weak economy, a citizenry with limited skills and fragile institutions. Since then, the country has made significant strides in its development endeavor307, restoring growth and improving welfare. Growth was primarily driven by investments in physical capital high levels of public spending in roads, bridges and the rehabilitation of infrastructure, needed to gain market access. In turn, public spending fueled growth in rural incomes. Growth was sustained by the positive response of family farms and family-owned businesses, which employ more than 90 % of the labor force in Mozambique, to the pro-growth economic policies of the government, which held down inflation while reducing the cost of doing business and lowering restrictions on competition308. According to Patt, donors were ready to fund good projects in underserved areas, and the government was determined to get these funds and use them to reduce poverty. The greatest successes were in roads and in education, where high levels of spending (20% of total government expenditures) produced sizeable improvements in both access and outcomes for poor and rural families309. Indeed, investments in social and economic infrastructure extended access to public services, reduced 305 World Bank, Mozambique Country Profile, accessed Dec 1, 2011. http://go.worldbank.org/70UK6S1X30 306 Ibidem 307 Ibidem
308 Instituto Nacional de Gestao de Calamidades (INGC), Study on the Impact of Climate Change on Disaster Risk

in Mozambique: Main Report, Asante, K., Brundrit, G., Epstein, P., Fernandes, A., Marques, M.R., Mavume, A , Metzger, M., Patt, A., Queface, A., Sanchez del Valle, R., Tadross, M., Brito, R. (eds.), 2009. 309 Ibidem

256

welfare inequalities, and supported the livelihoods of the average Mozambican310. They reduced monetary poverty from 69% of the population in 1997 to 54% in 2003311. As the World Bank further confirms, net primary school enrollment reached 95 percent in 2010, and infant and under-five child mortality decreased from 201 per 1,000 in 1997 to 138 per 1,000 new born in 2008312. Sustaining pro-poor growth into the 21st century, a major challenge: It is unclear today however whether Mozambique is continuing to beat the odds on shared growth. Results of the latest household survey indicate that poverty reduction has stagnated. The latest comprehensive data on living standards are from 2003, when about half the population was classified as poor313. Using national human development statistics as a proxy for the countrys overall vulnerability today, ones eye is caught by Mozambiques low average life expectancy at birth (49 years), its high infant mortality rate (135 out of 1,000 live births) and the still high percentage of the population living below the national poverty line (55% of the population). These figures denote of a still restricted success in the development endeavor, with many more gaps needed to be filled before full national / local capability and resilience is achieved (see table 8.1). Mozambiques development challenges today are to sustain the countrys impressive performance over the last two decades, notably by boosting production and productivity in sectors such as agriculture, making economic growth more inclusive and stepping-up exports and job creation314.
311 Ibidem 312 World Bank, Mozambique Country Profile

310 INGC, Study on the Impact of Climate Change on Disaster Risk in Mozambique: Main Report

313 314

INGC, Study on the Impact of Climate Change on Disaster Risk in Mozambique: Main Report World Bank, Mozambique Country Profile 257

Mozambique- Country Profile

2000 Overview Population, total (millions) 18,2 Population growth (annual %) 2,6 Poverty headcount ratio at national poverty .. line (% of population) GNI, Atlas method (current US$) (billions) 4,2 GNI per capita, Atlas method (current US$) 230 People Income share held by lowest 20% .. Life expectancy at birth, total (years) 47 Fertility rate, total (births per woman) 5,7 Births attended by skilled health staff (% of .. total) Mortality rate, under-5 (per 1,000) 177 Primary completion rate, total (% of relevant 16 age group) Ratio of girls to boys in primary and 75 secondary education (%) Prevalence of HIV, total (% of population 8,6 ages 15-49) Environment Forest area (sq. km) (thousands) 411,9 Agricultural land (% of land area) 61,2 Land area where elevation is below 5 meters 1.8% (% of total land area) Improved water source (% of population 42 with access) Improved sanitation facilities (% of 14 population with access) Energy use (kg of oil equivalent per capita) 394 CO2 emissions (metric tons per capita) 0,1 Electric power consumption (kWh per 122 capita) Economy GDP (current US$) (billions) 4,3 GDP growth (annual %) 1,1 Agriculture, value added (% of GDP) 24 Industry, value added (% of GDP) 25 Services, etc., value added (% of GDP) 51 Exports of goods and services (% of GDP) 16 Imports of goods and services (% of GDP) 37 Gross capital formation (% of GDP) 31 Revenue, excluding grants (% of GDP) .. Cash surplus/deficit (% of GDP) .. States and markets Time required to start a business (days) .. 2005 20,8 2,6 .. 6,1 290 .. 48 5,3 .. 158 42 83 11,2 400,8 62,0 45 15 409 0,1 445 6,6 8,4 27 25 48 32 44 19 .. .. 153 2008 22,3 2,4 54,7 8,6 380 5,2 49 5,1 55 144 59 87 11,4 .. 62,0 47 17 420 0,1 462 9,9 6,8 30 24 46 32 46 16 .. .. 26 2009 22,9 2,3 .. 9,9 440 .. 49 5,0 .. 140 57 88 11,5 .. .. .. .. 427 .. .. 9,8 6,4 31 24 45 25 43 20 .. .. 26 2010 23,4 2,3 .. 10,3 440 .. .. .. .. 135 61 89 .. 390,2 .. .. .. .. .. .. 9,6 7,2 32 23 45 25 43 24 .. .. 13

258

Military expenditure (% of GDP) 1,3 0,9 0,8 0,9 .. Mobile cellular subscriptions (per 100 0 7 20 26 .. people) Internet users (per 100 people) 0,1 0,9 1,6 2,7 .. Roads, paved (% of total roads) 19 .. 21 .. .. Global links Merchandise trade (% of GDP) 35,8 63,7 67,3 60,4 .. External debt stocks, total (DOD, current 7 255 4 211 3 450 4 168 .. US$) (millions) Total debt service (% of exports of goods, 12,5 3,6 1,1 1,6 .. services and income) Net migration (thousands) 75 -20 .. .. -20 Workers' remittances and compensation of 37 57 116 111 132 employees, received (current US$) (millions) Foreign direct investment, net inflows (BoP, 139 108 592 881 789 current US$) (millions) Net official development assistance and 906 1 297 1 996 2 013 .. official aid received (current US$) (millions) Table 8.1: Mozambiques socio-development indicators. Source: World Development Indicators 2010, World Bank

8.1.1.2 One of the younger nations of Africa, Mozambique won its independence in 1975 from the Portuguese, after an auspicious military coup in Lisbon and a 12-year guerilla struggle by FRELIMO- the Frente de Libertao de Moambique. Its charismatic leader and freedom fighter, Samora Machel, then installed a hardliner socialist, single-party FRELIMO government and harbored other freedom fighters from the neighboring Zimbabwe African National Union (ZANU) (then Rhodesia) and from South Africa's African National Congress (ANC). As a result, Mozambique was embroiled in a 16-year civil war with opposition party RENAMO- Resistncia Nacional Moambicana, originally sponsored by the white Rhodesian government and then the South African apartheid government. A peace treaty was signed between the two parties in 1992, under the leadership of Joaquim Chissano, Machels successor, and the countrys first multi-party elections were held in 1994. Mozambique has enjoyed relative peace, stability and prosperity since. 259 History & Political Context

Political vulnerability and State of Accountability:

Mozambique is a multiparty democratic Republic315, characterized by a strong presidential system and weak parliament. To date, Mozambique has held four general elections (1994, 1999, 2004, 2009), the first two won by Joaquim Chissano, and the last two by Armando Guebuza; three municipal elections (1998, 2003, 2008); and one provincial election in 2009. Suffrage is universal at 18 years of age. A strong Executive The current President of the Republic, H.E. Armando Guebuza, was elected for a second term during the presidential race of October 2009 and was inaugurated in January 2010. His ruling party, Frelimo, has held power since independence, and secured sizable majorities in the national parliament (about 75 percent), and provincial assemblies (80 percent). Frelimo also controls the totality of municipal assemblies, and holds 42 of 43 municipal presidencies. The RENAMO party, led by presidential hopeful Afonso Dhlakama, never succeeded in overtaking FRELIMO at general elections, collecting as low as 32% of the general vote in the 2009 elections. The next municipal elections are set for 2013, while the presidential and legislative elections will take place in 2014. In the meantime, the ruling party is set to hold its congress in 2012, during which it will designate a presidential candidate. Legislative power The country has three legislative bodies: the Assembly of the Republic with 250 deputies; provincial assemblies with over 800 delegates; and municipal assemblies in 43 municipalities. The ruling party has large majorities in all three legislative bodies, granting the president a de facto grasp over the entire political system. In addition, there are 11 provincial governors appointed by the President and who are a part of the executive. Thus in practice, the legislative cannot serve as an effective check to the power of the president, and ruling party.
315 For the purposes of this study, we define a Republic as a representative democracy in which the people's

elected representatives (deputies), not the people themselves, vote on legislation (CIA world fact book).

260

Three Political Parties Besides the ruling Frelimo party, originally a guerrilla movement credited for having liberated the country from colonial rule in 1975, the largest opposition party is Renamo, followed by the Democratic Movement of Mozambique (MDM), a Renamo breakaway which gained seats in the national parliament for the first time in 2009 elections. MDMs leader, Mr. Davide Simango, had been elected in 2008 as mayor of Beira, Mozambiques second largest city, after running as an independent, and thus becoming the first-ever independent candidate to win an election in the country. Up to the introduction of a democratic system in the 1990 constitution and the holding of the first multiparty elections in 1994, Frelimo was an unchallenged political force under a one-party system. The two first multiparty elections in 1994 and 1999 were relatively easy victories for Frelimo, although its popularity started to crack under perceived corruption and an increasingly vocal independent media and opposition. In the 2004 elections, the party replaced then President Chissano with Armando Guebuza who led a vigorous electoral campaign against corruption, a new theme in electoral campaigning, leading to an easy victory. The election also marked the start of Renamos drop in popularity as it struggled with internal disputes. Frelimo, under Guebuza, made it a priority to reinforce its political base, which led to another comfortable victory in the 2009 elections. Decentralization policy In recent years, Mozambique has engaged in significant efforts toward greater de- concentration and devolution of power to local levels. The decentralization process has been marked by the introduction in 1998 of municipalities, autonomous political circumscriptions in charge of administrating the countrys cities, led by a mayor. Indeed, after some delays, in 1998 the country held its first local elections to provide for local representation and some budgetary authority at the municipal level. RENAMO boycotted the local elections, citing flaws in the registration process. 261

Independent candidates also contested the elections but won seats in municipal assemblies. Turnout was very low. In the aftermath of the 1998 local elections, the FRELIMO government resolved to make more accommodations to the opposition's procedural concerns for the multiparty national elections in 1999316. In 2003, Parliament adopted the Decentralization law of 8/2003, setting the principles of organization, competencies and functioning of Local State authorities at the provincial, district, localities and administrative post levels. In 2009, the provincial assemblies were introduced, replicas of the national assembly in each of the countrys 11 provinces. The roles of these bodies are yet to be broadly understood however by citizens and even some public officials at the municipal level317. Figures 8.2-4 schematically represent the partitioning of the country, and decentralized sub-units of the State apparatus.

316 US State Department Background Notes Mozambique, 2011

317

World Bank, Mozambique Country Profile

262

Republic of Mozambique President of the Republic

Province Governor

District Administrador de District

Administrative Post

Locality Chefe de Localidad

Circle (rural) Secretario de Circulo

Community Village (rural) Chefe de Aldeias comunais

Neighborhood (urban) Chefe de bairro

Celula

Zona

Fig. 8.2: Administrative Partitioning of the country: Flowchart318

318 The title in italics designates the function of the officer at the head of the administrative entity. For

example, the district of Pebane in the province of Zambezia province is headed by a District Administrator or Administrador do Distrito.

263

Municipality (eg: Pemba) Presidente de Municipio

Bairro Chefe de bairro Unidades Chefe de unidade

Fig. 8.3: Political Partitioning of the country: Flowchart

Note: Municipalities serve as the headquarters of Provinces, but are considered as autonomous entities, whose leaders are elected by the municipal citizens. They have their own budget, own administration and heading the Municipality is the Mayor or Presidente de Municipio.

Community Lider Comunitario (Regulo) 1st level Sub-community Lider comunitario (Regulo) 2nd level

Cellular Secretario de Cellular

Zona Chefe do zona

TRADITIONAL ADMINISTRATIVE/POLITICAL
Fig. 8.4: State apparatus extensions in rural areas: co-habitation between elected local traditional leaders (Regulo) and appointed politico-administrative leaders (Secretario)

264

8.1.2 Vulnerability to Climate Changes 8.1.2.1 Due to its geographical localization right on the Mozambique Channel on the track of tropical cyclones borne from monsoonal activity in the Indian Ocean, and lying at the receiving end of nine major international rivers making their way to the Indian Ocean through the country, Mozambique is subject to multiple extreme weather events. The country ranks third among African countries most exposed to weather- related hazards. Floods, epidemics and cyclones are the most frequent disasters, although drought affects by far the largest number of people, with sea level rise emerging as a significant hazard319. During the past 50 years, Mozambique has suffered from a total of 68 major hydro- meteorological disasters, which have killed more than 100,000 people and affected up to 28 million. Up to 25 percent of the population is at risk from natural hazards320. High Exposure: Mozambique, 3rd African country most exposed to HMDs

Fig. 8.5: Disaster Impacts in Mozambique. Source: GFDRR, 2011

319 GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique, 2011 320 World Bank, Mozambique Country Profile 265

Drought is the most severe disaster in Mozambique. Drought occurs primarily in the southwestern region (west from the Gaza province) and in the central region (west from Tete province), with a frequency of 7 and 4 in 10 years respectively (see MICOA 2005321; also GFDRR 2011322). Provinces particularly vulnerable to drought are Gaza, Inhambane, and Maputo, where rainfalleven when above averageis often insufficient and results in critical water shortages and limited agricultural productivity. According to the World Bank, between 1980 and 2000, droughts contributed to an estimated 4,000 deaths. Drought disaster costs to the national economy have been estimated at $1.74 billion during 19802003, but this largely underestimates losses and impacts on the poor323. The second most impactful disaster in Mozambique is flooding. Floods in Mozambique occur every 2-3 years and can last up to several months. They are caused by a number of geographical factors. Nine major international hydrographic systems find their way to the Indian Ocean through Mozambique324. In fact, 50 percent of the water in Mozambiques rivers originates from outside the country. Flooding is linked not only to heavy rainfall events, but also to water drainage from rivers in upstream neighboring countries. Epidemics have been generally associated with flood disasters325. The southern and central regions are where floods have been most recurrent, in river basin areas, low-lying regions, and areas with poor drainage systems. At least 1.7 million hectares, or 6 % of the national territory, could be at risk from flooding, principally along the coast and the main riverbeds326. Areas most vulnerable to
321 Ministry for the Coordination of Environmental Affairs (MICOA), National Adaptation Programme of Action,

2007
322 323

GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique, 2011 World Bank, Mozambique Country Profile GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique, 2011 MICOA, National Adaptation Programme of Action, 2007 266

324 These are the Buzi, Incomati, Limpopo, Maputo, Pungo, Save, Rovuma, Umbeluzi and Zambezi rivers. For all

325 326

but Rovuma, the flood plains are inside Mozambique (GFDRR, 2009).

flooding are, but not limited to, areas close to the coast. In 2000, Mozambique experienced its worst floods in 50 years, killing about 800 people and displacing 540,000327. Finally, Mozambique is subject to three or four cyclones every year, which travel up the Mozambique Channel due to monsoonal activity in the Indian Ocean, from October to April. The whole coast is affected, but areas most affected by cyclones are located between Pemba and Angoche, on the northern coast of Mozambique328. In addition to the extreme wind and rainfall brought by these cyclones, cyclonal activity can exacerbate seawater inundation and storm surge threats. It also warrants noting that in the last decades, the southern parts have been hit recurrently by Cyclones such as in 2000, 2001 and 2007. Other significant climate-related hazards are coastal erosion, tidal incursions and storm surges. More than 60% of Mozambiques population of 23million live in coastal areas, and is therefore highly vulnerable to seawater inundation along its 2,700 km coastline. Seawater inundation includes saline intrusion deep into river mouths, beach erosion, and short extreme rises in sea level due to tropical storms and cyclones. Saline intrusion of the coastal aquifers and estuaries holds serious ecological and economic implications for agriculturalists settled along rivers. This is partly due to the fact that some areas in the country are below sea level such as Beira. The issue of beach erosion is also very serious, threatening coastal infrastructure such as roads and housing as well as tourism. In some portions of Beira (central region), 30 to 40 meters of beach have been eroded in the past 15 to 20 years, destroying natural mangroves and encroaching on homes, roads and touristic

327 World Bank, Mozambique Country Profile 328 MICOA, National Adaptation Programme of Action, 2007

267

establishments329. Finally, storm surges pose a huge threat to coastal infrastructure as they can temporarily raise sea level as much as 5 meters. While many of the major coastal cities of Mozambique have infrastructure in place to stem the effects of such an extreme event, many are in need of serious maintenance330. All in all, areas highly exposed to multiple HMDs in Mozambique are: coastal areas which lie at the receiving end of the nine hydrographic basin and are thus triply at risk: from flooding, sea level rise and cyclones. Areas particularly prone to rainfall deficits are the Southern and Central regions. 8.1.2.2 Following from our definition of a Disaster as the conjugation of a naturally driven Hazard and human Vulnerability, we understand that multiple hazards that make landfall in Mozambique turn into large-scale disasters not only because of high exposure, but also due to acute socio-economic vulnerability they meet on ground. This is particularly obvious in the case of drought, where areas experiencing rainfall deficit (hazard) do not necessarily correspond to areas suffering from drought disasters. Provinces particularly vulnerable to drought are Gaza, Inhambane, and Maputo, where rainfalleven when above averageis often insufficient and results in critical water shortages and limited agricultural productivity. The SETSAN nation- wide study of food security conducted in 2008 confirms these insights. The study found that Sofala, Tete and Nampula provinces suffer from acute food insecurity, reflecting their vulnerability to the adverse effects of climate-related disasters and limited adaptive capacity; while the provinces of Zambzia, Tete and Maputo, on the other hand, feature the highest level of chronic food insecurity, a result of persistent High socio-economic vulnerability, Low Adaptive Capacity

329 MICOA, National Adaptation Programme of Action, 2007 330 World Bank, Mozambique Country Profile

268

poverty and systemic food access issues (see SETSAN 2008)331. Therein, an estimated 35% of the population is thought to be chronically food insecure. Economic impacts of drought disasters have been most significant in Zambzia Province, where production losses could range between $12-$170 million for maize alone, depending on the severity of the drought332 . Despite strong economic growth (6,3% in 2010), over 80% of the population continues to depend on agriculture for subsistence. High rates of poverty (55% in 2010), malnutrition, HIV/AIDS, and endemic diseases contribute to what is already a high physical vulnerability333. Furthermore, above 60% of the population lives on the 2,700km coastline, rendering the coastal areas hotspots of multiple hazards 334. Characterizing Mozambiques adaptive capacity in the face of HMDs remains a tricky endeavor. Vulnerability to HMDs can only be assessed using proxies of overall inherent or social vulnerability assumed to compose vulnerability to HMDs, such as level of poverty, access to health, education and food security. However evidence on vulnerability at the sub-national level in Mozambique is rare; information related to the exposure of Mozambique is available at a geographically disaggregated level, but the same does not hold for sensitivity335. Most data available on sensitivity is focused at the national level, or zooms in on a particular local case study, providing no basis for comparison. A thorough vulnerability mapping of the country of Mozambique will require access to detailed and reliable datasets on both exposure and sensitivity 336.

331 SETSAN (Technical Secretariat for Food Security and Nutrition), Report on the Monitory of Food and

Nutrition in Mozambique, May 2008

332 333

World Bank, Mozambique Country Profile GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique, 2011 334 Ibidem 335 Ibidem 336 Ibidem 269

Table 8.6 from Furgusons in-depth study in the district of Buzi provides a rare snapshot of local vulnerability, and low adaptive capacity, to HMDs.

Fig. 8.6: Risk and Vulnerability to climate-related hazards in Buzi, Buzi district. Source: Ferguson, 2005

8.1.2.3

Conclusion: A country highly sensitive to climate change and variability

The literature reviewed provides a consistent picture that Mozambique is particularly vulnerable to climate change and variability. It is not clear whether Mozambique is more exposed to multiple HMDs than other countries in tropical regions (see UNDP 2007), however its particularly high vulnerability to the multiple HMDs it is exposed to renders national adaptation to climate change and variability a high priority for Mozambique. At the national-level, the direct impacts of extreme weather hazards combined with acute vulnerability generate severe effects on Mozambiques developing economy. Indeed, the majority of the populations reliance on rain-fed agriculture for 270

subsistence337, absence of traditional and modern irrigation systems to mitigate the effect of erratic rainfall of agricultural yields, inadequate infrastructure and weak institutional framework are important factors. A regression analysis over the period 19812004 suggests that Mozambiques GDP growth is cut by an average of 5.5 percent when a major water shock occurs338 . According to Bambaige (see UNDP 2007), the sectors that should be the focus of adaptation measures in order to increase the resilience of socio- economic systems are agriculture, water, energy, and health339. At the local level, multiple HMDs reduce opportunities. Natural disasters do not affect all communities in the country evenly, as their impacts are correlated to the ability of those communities to withstand them340. Disaster hotspots in Mozambique are areas where multiple hazards interplay with acute poverty for instance, subsistence farmer coastal communities in Pebane (Zambezia province) or inner city neighborhoods in Pemba (Cabo Delagdo province) triply at risk from cyclones, drought and flooding are a prime example. Such particularly high vulnerability, combined with significant exposure to extreme weather events, makes adaptation of utmost importance for Mozambique. A look at regional climate model projections reveals that climate change is expected to only increase the frequency and magnitude of shocks, while cyclones are likely to continue to pose large threats to Mozambiques growth. 337 MICOA, National Adaptation Programme of Action, 2007 338 World Bank, Mozambique Country Profile 339 GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique, 2011 340 MICOA, National Adaptation Programme of Action, 2007 271

Drought Extreme temperature Flood Storm TOTAL AVERAGE

Ethiopia 14 0 49 0 63 15,75

Madagascar 6 0 6 45 57 14,25

Morocco 5 1 28 2 36 9

Mozambique 12 0 29 17 58 14,5

Niger 8 0 15 1 24 6

Senegal 6 0 17 3 26 6,5

South Africa 8 2 30 21 61 15,25

Tanzania Uni Rep 9 0 31 4 44 11

Uganda 9 0 16 4 29 7,25

1960-2011 Source: EM-DAT

Table 8.3: Mozambiques exposure to HMDs, as compared to a sample of other African nations (data source: EM-DAT)

8.1.3 Projected Climate Change impacts in Mozambique The Mozambique country case of the World Banks Economics of Adapting to Climate Change (EACC) study comprises the most comprehensive assessment of projected Climate Change impacts on Mozambique341. The study compares growth paths with climate change to a baseline of growth without climate change, considering five priority sectors impacted by climate shocks: (1) agriculture; (2) hydropower generation and water availability; (3) transport infrastructure, notably roads; and (4) coastal zones, which do not constitute a sector per se, but a hotspot of climate risk. The study thus projects the impacts of climate change in these priority sectors and assesses costs of adaptation. Adaptation is defined as a set of actions designed to reduce or eliminate the deviation from the baseline development path caused by climate change342. It needs to be noted that this assessment of the impacts of climate change on Mozambique is a conservative one, as it does not project impacts on other sectors such as ecosystem services or health (malaria incidence).
342 Ibidem 341 World Bank, Economics of Adapting to Climate Change, Mozambique, 2010

272

Just as for the rest of Africa, scientists are more confident about projections of temperature changes over Mozambique than projections for precipitation. In order to represent the full range of uncertainty in the climate projections derived from Global Circulation Models (GCMs), the World Bank study describes for Mozambique a wet and dry scenario, and introduces Mozambique wet and Mozambique dry models to represent the range of possible outcomes for Mozambique343. The picture that emanates from the model runs for Mozambique are as follows. By 2050, Mozambique is projected to experience an increase in temperature of 12 degrees Celsius according to both Mozambique Wet and Dry scenarios; more precisely, temperatures will increase by 1.15 to 2.09 degrees Celsius, though with regional variations (see figure 8.7).

Fig. 8.7: Mozambique Temperature in 2050 Dry (left) and Wet (right) scenarios. Source: World Bank EACC, 2010

With regards to precipitation, depending on the scenario, precipitation in the southern region is projected to either decrease relatively little (in the dry scenario) or increase dramatically (in the wet scenario). Thus, depending on the model used, precipitation will either increase or decrease, again with regional differences (see
343 It is important to note that Mozambique wet (IPSL_cm4 sres_a2) and Mozambique dry

(UKMO_hadgem1 sres_a1b) models differ from the two extreme GCMs used in the World Banks EACC global track study, labeled global wet (NCAR-CCSM sres_a1b) and global dry (CSIRO- MK3.0 sres_a2). Indeed, a globally wet scenario is not necessarily wet in Mozambique. In fact, the global wet scenario projects a slight drying and the global dry is in fact somewhat wetter in Mozambique. The two models Mozambique wet and Mozambique dry were introduced to correct this discrepancy and display the range of possible outcomes specific to Mozambique. World Bank, Economics of Adapting to Climate Change, Mozambique, 2010

273

figure 8.8). The main message here is that climate will become increasingly variable and uncertain, and that people and decision-makers need to plan for this uncertainty344.

Fig. 8.8: Mozambique Precipitation in 2050 Dry (left) and Wet (right) scenarios Source: World Bank EACC, 2010

Comparing Figures 8.7 and 8.8, it becomes clear that regional variation in precipitation is more significant than regional variation in temperature. As shown by the maps above, regardless of the model used, regional variation in precipitation between northern and southern regions of Mozambique is projected to remain important345. Finally, changes in precipitation and temperature from the four GCMs (the two global scenarios plus two extreme scenarios for Mozambique- extreme wet and extreme dry) were used to estimate for the 2050 time horizon: (a) the changes in yield each year for both irrigated and rain-fed crops, as well as irrigation demand for six cash crops and eight food crops; 2) changes in river flow into hydropower generation facilities, and resulting changes in generation capacity; 3) impacts on transport infrastructure and the increased demand and costs of road maintenance. Simulations of sea level rise were constructed independently of the climate
344 World Bank, Economics of Adapting to Climate Change, Mozambique, 2010 345 Ibidem

274

scenarios, using both an integrated model of coastal systems to assess the risk and costs of sea level rise in Mozambique, as well as focused analyses of interactions between cyclone risk and sea level rise in two coastal city cases Beira and Maputo, the two largest cities in Mozambique346. Table 8.4 describes results of climate change projections for each key sector of vulnerability considered. Sector vulnerability Agriculture
Agriculture in Mozambique accounts for 24% of GDP and 80% of employment. Traditional slash & burn farming techniques plus frequent uncontrolled fires result in poor soils vulnerable to erosionand hence to further productivity losses during floods / drought. In all scenarios, net national average crop yield lower relative to baseline yield without CC 24% reduction in yields of major crops, especially in central region Agricultural GDP loss of 4.5 9.8%

Only 7% of Mozambicans have access to modern electricity. The primary source of energy (the grid) is hydropower from barrages on the Zambezi Basin. Fuelwood and coal were not considered in the study.

Energy (hydropower generation)

Transport (Roads)
Mozambique has one of the lowest road densities per person of any African country. Losses from climate change projected to be substantial, in large part because of the importance of current and required investments in the sector. Impacts of severe rainfall events on roads, culverts, and bridges would lead to damage / loss of access Reduction in stock of roads (measured in km) without adaptation estimated between -2% (Moz dry scenario) and - 12% (Moz wet scenario); -9% reduction with

Coastal zones
60% of Mozambiques population is clustered in the countrys 2,700km- long coastline

Projected Climate Change Impacts by 2050 (highest confidence)

Under all scenarios except most pessimistic, 1.4% less energy generated than without CC (only slight negative impact) Potential energy deficit due to CC is of approx. 110,000 GWh relative to baseline generation in 200550 Impacts limited by new energy generation plants integration of

By 2040s, with no adaptation, Mozambique could lose up to 4,850 km2 of land from today (0.6% of national territory) 916,000 people could be forced to migrate away from coast (2.3% of projected 2040s pop.). Worst-case scenario: total annual damage costs to reach $103 million/year in the 2040s forced migration being a large contributor to that cost. High geographic concentration of costs in

346 World Bank, Economics of Adapting to Climate Change, Mozambique, 2010

275

Projected costs of adaptation (in discounted USD billions, const. 2003)

Damage avoidance strategy: Beach shore nourishment & river and sea dike building. Cost: $890 million/year in the 2040s for the high sea level rise scenario. Strategy would reduce total damage expected by a factor of four (from $103million to $24million/year by 2040s), reducing total land area lost by a factor of 80 (to only 61km2) and number of forced migrants by factor of 140 (to 7,000 people). However costs of this strategy outweigh its economic benefits by a large margin. More viable adaptation options: strategic construction of hard protection infrastructure in high value / unmovable vulnerable locations (e.g: Port of Beira). Costs: $190 $470 million/year in the 2040s, depending on the sea level rise scenario. Table 8.4: Projected Impacts of Climate in Mozambique, with associated levels of confidence. Source: World Bank, 2010.

Expanding Irrigation. Cost: $ 0.6 b. Alternative soft & less costly measures advised given extent of land requiring irrigation: Water harvesting, soil/moisture conservation and agro-forestry. Agriculture R&D. Cost: $ 6.1 b. Softer resilience building strategies include: Support for improved access to markets/ inputs, value addition and reduction of post- harvest losses.

changing patterns of temperature and rainfall in planning Main source of negative change: increased evapo- transpiration in the reservoirs (less water availability for electricity) Across all sectors, in order to regain remaining loss in welfare: Human capital accumulation needed (Education). Cost: $ 6.1 b.

adaptation.

Zambezia, Nampula, Sofala and Maputo provinces (low-lying, densely populated areas) Even small levels of sea level rise dramatically increase the probability of severe storm surge events

Sealing unpaved roads and culverts in highly strategic areas (no-regret strategy recommended even under the baseline). Cost: $ 1.5 b.

Using a dynamic computable general equilibrium (CGE) model, an aggregate picture emerges of CC impacts on Mozambiques macro-economic performance as a whole, across all sectors. This picture points to significant declines in national welfare by 276

2050 and slowing rates of economic growth. In the worst-case scenario, the net present value of damages (discounted at 5%) reaches about $7.6 billion dollars, which is equivalent to an annual payment slightly above $400 million. GDP falls between 4 and 14% relative to baseline growth in the 204050 decade if adaptation strategies are not implemented347.

Fig. 8.9 (Left): Projected CC impacts for Mozambique using four models (2 GCMs and two Mozambique specific), by priority sector, relative to baseline of no climate change. Fig. 8.10 (Right): Costs of adapting to Climate Change in Mozambique Source: World Bank, 2010.

347 World Bank, Economics of Adapting to Climate Change, Mozambique, 2010

277

8.2 HMDs and Government Response in Mozambique: 1995-2010 8.2.1 Understanding Impacts of HMDs in Mozambique, 1995-2010 The break year in Mozambique that marks the accentuation of inter-annual variability and increase of HMDs is 1994, similar to the trend observed across Africa. Figure 8.11 illustrates eloquently the staggering increase in HMDs from 1994.

Number of HMDs that hit Mozambique, 1979-1997

350 300 250 200 150 100 50 0 n of HMDs

Fig. 8.11: Number of HMDs that hit Mozambique between 1979-1997, a notable hike starting in 1994348. Source: DesInventar Mozambique data, GRIP-Arendal.

We now turn to investigating the impacts of these HMDs, since they began to occur with increased frequency, at both the national and community levels. 8.2.1.1 Impacts at National Level Disaster costs to the national economy have been estimated at $1.74 billion during 1980-2003, but this largely underestimates economic losses and impacts on the 348 HMDs include: drought, floods, heat waves, alluvions, cyclones, strong winds, rains and
tropical depressions, as well as disaster-related epidemics.

278

poor349. Table 8.5 reviews HMDs that have hit Mozambique during our period of study, and their adverse impacts across sectors, from local to national. From the above table clearly appears that central and southern regions bear the brunt of multiple HMDs (droughts, floods, cyclones) hitting Mozambique simultaneously within the same year, with increasing frequency, and causing extensive damage in terms of lives lost, livelihoods interrupted and families affected.
Climate shock 1994-1995: Cyclone Ndia in 1994 affects coastal area of Nampula, exacerbates drought that occurs in 1995 1995, August and September: Drought Impacts in the province of Tete 1997, January to March: Floods Impacts in the Zambezi and Shire Rivers 1997 July: Drought and Desertification Impacts in almost all of the country, but serious situation for the provinces of Tete, Manica and Sofala. 1998 June: Desertification and drought affects six provinces (Maputo, Gaza, Inhambane, Tete, Manica, Sofala) 2000: Floods affect whole country Greatest impact in Gaza and districts in the Zambezi valley 2000, April: Drought in the north, floods in southern Manica 2002, March: Drought Impacts in central and southern regions 2003: Drought and cyclones between October and March Impacts in central and southern regions 2003: Cyclone Japhet Impact on some parts of the southern and central regions, particularly north of Impacts Anticipation of agricultural crop in Nampula; One and a half million people at risk Estimated above 100 deaths from hunger; 1994/95 agricultural campaign affected; 200 animals die in Tete province (in the districts of Changara, Moatize and Chiuta). 24,800 families affected by flooding in Mutarara. Forced migrations of populations as well as animals from their natural habitat to other areas of the country and to neighboring countries. In Inhambane, 60 000 people are affected; Lack of water for consumption. Access routes to critical areas of flooding in the province of Gaza are cut off; 307,000 people are affected in the north of Gaza 700 people died and half a million lost their homes 3,000 ha of crop in Mussurize, Sussundenga and Machaze affected. 4,000 families in the districts of Sussundenga and Mussorize in a state of emergency. More than 500,000 people affected (114,000 families) across country, of which 60,000 in central and south. 2.6 million people affected; Lack of food affects 1.5 million people; In the province of Gaza nearly 60,000 suffer from hunger and 25 000 ha of maize have affected in Chokwe district Destruction of Massangena hospital in the north of the province of Gaza.

349 World Bank, Economics of Adapting to Climate Change, Mozambique, 2010

279

the province of Gaza (in Massangena) 2004: Drought Impact on the provinces of Gaza, Inhambane, Manica and Sofala 2005: Drought Impact on southern and central regions 2007: Cyclone Fvio Occurrence in Inhambane

659 000 people affected. 1,142,250 people affected in seven provinces. About 183,000 people were affected by the Cyclone; Food crisis due to rupture of stock in the warehouses of the National Disaster Management Institute (INGC); 6,000 workers in the hotel industry were without jobs ; 42 touristic establishments in Inhambane destroyed by the cyclone; Destruction of 20,800 hectares of crops. Food crisis causes a rupture of stock in the warehouses of the INGC. In 2007, flooding in Mozambique killed at least 29 people and affected 285,000 people, the worst since the 2000- 2001 floods. 144,280 people affected. . 25 000 people affected by floods in the districts of Caia, Morrumbala, Mutarrara and Mopeia. Heavy rains in Zambia, Zimbabwe and Malawi caused flooding in Mozambique that displaced tens of thousands of people and destroyed almost 100,000 hectares of crops. As a result of the floods and consecutive droughts in 2002/03, 2003/4 and 2007/08, the World Food Programme placed 300,000 people under food assistance 785 families affected; The province of Inhambane considered the most severely impacted with about 193 000 families affected. In Chemba, 726 hectares of various crops were destroyed by the waters of the Zambezi; 4 deaths and 11,561 left homeless; 229 thousand hectares of various crops at risk.

2007: Floods Impact in the provinces of Sofala, Manica, Tete, Zambezia 2007, March: Drought Impact in the province of Gaza 2008: Floods Impacts in the central region of the country

2010 February: Drought Impacts in 61 districts across the country 2010: Floods impact Central region; Drought in the southern region

Table 8.5: Impacts of HMDs in Mozambique, 1995-2010

8.2.1.2 Through our community fieldwork conducted in Mozambique from March 20th May 1st 2011, we elicited impacts of HMDs in communities along the Mozambican coastline, and their CCA-DRR needs, using the Red Cross climate Vulnerability and Capacity Assessment (VCA) tool. In this section we present the critical community resilience-building needs that have emanated from our community study. 280 Zooming in: Impacts at the community-level

Case selection criteria:

Seven communities were selected at random, in three different sites along Mozambiques coastline from North (Pemba) to South (Inharrime). We chose to focus on coastal areas for these VCAs due to the triple vulnerability of coastal areas to all the major hazards that hit Mozambique. Thus all of the communities that were selected are low-income fishing or peasant communities, ocean-dependent and vulnerable to all three main hazards cyclones, drought and flooding. While some of the communities identified in the VCA analysis are not located along the actual coastline, they do rely on the coast for their livelihoods. Fig. 8.12 provides a visual representation of the location of the three sites where VCAs were conducted.

Figure 8.12: Geographic location of the selected VCAs sites along Mozambiques coastline. Credit: Daniel Zakarias.

281

Vulnerability in Pemba Municipality (urban, northern Mozambique) The city of Pemba is located on the northern coast of Mozambique, bordered by the Indian Ocean, whose waters and crystal white sands of its beaches provide beautiful scenery. The municipal area, with an area of 194 sq. km, occupies a peninsula that defines the Bay of Pemba, the third largest bay in the world. It is bounded on the east by the Indian Ocean, to the west and north by Pemba Bay and in the South by the District of Mecfi and the District of Pemba Metuge. The peninsula of Pemba is situated in tropical zone and has a hot and humid climate that is reinforced by its location on the shore of Pemba Bay and its proximity to the Mozambique Channel. The winds associated with the inter-tropical convergence zone typically occur in summer, mainly in January. In other periods, the atmospheric circulation is influenced by winds blowing from east and anti-cyclonic actions. The action of storms and winds is the main driver of the formation and maintenance of the shoreline through the deposition of sand. Additionally waves and ocean currents also influence the dynamics of the coast, through processes of confrontation or erosion, including wind, caused by rains, and waves of marine waters. The coast of Pemba Bay (the mouth and inland) is characterized by low cliffs alternating with small sandy beaches and mangroves. Predominate in the inner bay water areas, salt marshes and mangroves, while the peninsula's northern coast (mouth of the Bay) is composed of rocky areas composed of fossil coral, and small sandy beaches, culminating in the beautiful beach of Wimbe. The Indian Ocean has vast beaches of fine white sand and clear waters. The banks and reefs run along the entire coastline of the municipal territory. Tides are semi-diurnal with a maximum width of 4.5 meters (INAHINA, 2002). However, the movement of water in the bay has not been studied. Vulnerability of the District of Pebane (rural, central Mozambique) Pebane is situated at the north-eastern ocean front of Zambzia province. It covers a surface of 10.086 sq. km with a population trend of 135.275 habitants (in 1997); 282

168.602 in 2005 and 185,333 in 2007 (National Census, 2007). This population is young (nearly 44% of the population is below 15 years old), mainly feminine (51%) and mostly rural (only 6% of the total area of the district is urban). Climate in Pebane is predominatly rainy tropical with two distinct seasons, and average precipitation ranging 1.286 mm and the reference evapo-transpiration ranging 1.514 mm. High rainfall occur from December to April (nearly 75 to 80% of the total rainfall in the district), mostly influenced by the proximity with the sea. Its geomorphology is dominantly a sedimentary plain with sandy cover in the coast and consolidated precambric sediments in the interior, both intercepted by natural drainage lines where much recent sediments occur (marine and estuarine at the coast and aluvionary in the interior part of the district). The coastal strip also comprises yellow and whitish sandy soils locally modified by hidromorphism. There are two options to access Pebane, by car or by boats, although accessibility is very difficult. Water access is still scarce with some communities travelling nearly 12 km to fetch 25 litres of water. There are 111 schools and 11 health establishments. Main economic activities at the district of Pebane are agriculture (small-scale rain-fed subsistence agriculture), small-scale fisheries and tourism. Traditional leadership is secured by the representatives of the government at the local level such as secretaries and chiefs of districts, heads of groups of villages, heads of villages and other persons in the community respected and legitimated by their social, economic, cultural and religious roles. In traditional leadership is no division of labor and responsibilities between the different community leaders. Thus, the secretaries have the task of mobilizing the community for social and economic tasks and the traditional leaders dealing mainly of traditional aspects, such as ceremonies, rites and social conflicts. 283

The district of Pebane is vulnerable to natural hazards (droughts, floods, cyclones). Erosion, uncontrolled bush fires and deforestation of mangroves for charcoal production are the major threats to the environmental sustainability of local resources. Several measures have been developed at local level to reverse the situation and improve the resilience of communities. Apart from preventing forest fires, planting of casuarina trees along the seafront and beach accesses, as well as surveying in the area where erosion occurs, sensitization of communities to spare part of their surplus production, practice crop drought- resistant, choose riverbanks for agricultural production, promotion and improvement of cashew, production and distribution of cassava cuttings with short cycle and rot resistant, as well as the purchase and distribution of bananas and pineapples are some actions being implemented. Vulnerability of Zvora (rural, southern Mozambique) Inharrime district (where Zvora is located) is situated at the southern border of Inhambane province, covering an area of 2,744 sq. km, with an estimated population of 97, 950 inhabitants in 2007. The population is young (45% of the total population is under 15 years of age), mostly female (approximately 54% of total population) and with rural characteristics (only 5% of the total area of the district is urban). Tropical dry zones (inside) and moist zones (near shore) dominate the climate of the district. The coastline, with permeable soil and favorable conditions for agriculture and livestock, has average temperatures between 18 and 33 C, while the interior features sandy and clay soils, average annual rainfall between 1000 and 1200 mm and high temperatures causing water shortages. The district is accessible only by the road through the National Road 1 (across the district), while access to the interior and the coastline is via dirt roads. There are 60 district schools and 6 clinics. The main economic activities are agriculture, fisheries, tourism and small-scale. 284

Aggregation of priority DRR-CCA needs across 7 target communities:

From our fieldwork in Pemba, Pebane and Zavorra emanated a clear picture of the priority needs of multiply exposed coastal communities in Mozambique. Table 8.6 summarizes the priority DRR-CCA needs emanating from the seven target communities surveyed along the Mozambican coastline. As we see in the next section, many of these priority needs are addressed by the National CCA-DRR policy.
Table Key: 1. Chuiba (Pemba site) 2. Paquitequete (Pemba site) 3. Chuibuabuare (Pemba site) 4. Macuacuane (Pebane site) 5. Quichanga (Pebane site) 6. Malaua/Porto (Pebane site) 7. Sihane (Inharrime site)

Intervention 1.Facilitate/ Enable transition to alternative climate- resilient livelihoods 2. Replant trees, grass and mangrove along coastline to thwart coastal erosion and break force of cyclones/strong winds 3. Tailored agricultural extension services to master/access agricultural techniques adapted to increased climate variability (drought & flood) 4. Provide training and equipment for construction of small- scale water management works for irrigation (water pumps, reservoirs) 5. Community Disaster Risk Management support (including establish/equip local Needed community DRR-CCA interventions (above local capacities to cope) 1 2 3 4 5 6 7

(to drought only)

(establis h)

(establi sh)

(establish )

(equip existing local DRM committe e &

(establish )

(establish )

285

committee for disaster risk management). 6. Identification & mapping of Climate risks to support planning and programming at local, district, municipal levels: 7. Establishment of a Sea Level Rise Monitoring/surveillan ce system 8. Decision-making planning informed by knowledge of risks (develop and implement urban climate risk management plan, public awareness campaigns) 10. Dig a new canal/drainage ways to take care of problem of stagnant waters and soil erosion 11. Administer cure for dying coconut trees, reservoir of nutrition, knowledge and cultural identity for the local community 12. Capacity-building of community leadership to develop community- based adaptation plan; awareness-raising programs. 13. Capacity-building of district leadership to carry out district- wide awareness- raising programs on CC risks

communit y radio)

Table 8.6. Mozambique community Vulnerability and Capacity Assessments: results of community prioritization of climate change adaptation interventions

286

8.2.2 National Policy Analysis We had found from our mapping that Mozambique was a country classified as a Disaster Averter. However, is this verified in practice? What has the government of Mozambique done to confront rising climate risks, and address local CCA-DRR priorities? How is Mozambique organized to deliver on the CCA-DRR challenge? 8.2.2.1 Five main organizations dominate Mozambiques institutional landscape for CCA- DRR, composed of a partnership of NGOs, development partners, UN agencies and civil society organizations. These are: the INGC, CCGC, INAM, CENOE and MICOA. INGC The National Institute of Disaster Management (INGC), established in 1999, coordinates disaster risk management activities in Mozambique. INGC operates under the Ministry of State Administration (MAE) and is mandated to coordinate emergencies, promote disaster prevention through population and government mobilization; protect human lives; ensure multisectoral coordination in disaster emergency; coordinate early warning systems; carry out public awareness; and re- utilize arid and semi-arid zones. INGC is responsible for coordinating disaster risk management at the national, provincial and district levels. Three regional emergency operation centers handle cyclones and droughts (Vilanculos), floods (Caia) and cyclones (Nacala). There are also four multiple use centers (CERUM) at the district level specializing in reducing vulnerability to droughts. At the community level, INGC acts through local committees for Disaster Risk Management that are empowered to deal with both disaster prevention and preparedness. CCGC 287 Institutional landscape of CCA-DRR governance in Mozambique: what are the rules of the game, who are the stakeholders?

The Coordinating Council for Disaster Management (CCGC), chaired by the Prime Minister, ensures multi- sectoral coordination in disaster prevention, assistance to victims, and disaster rehabilitation. It receives advice from the Technical Council for Disaster Management (CTGC). The CTGC, composed of technical staff from sector Ministries represented in the CCGC, proposes technical responses to disasters, which are then submitted for analysis and approval to the CCGC. The CCGG decision is then forwarded to the operating body of INGC for action through its regional, provincial and district representatives. The CTGC is also active at the provincial level, where it advises the local INGC and the Provincial Government and conducts disaster evaluations. Sounding the Alarm: INAM and SETSAN Mozambique has a well-developed Early Warning System. INGC holds overall coordination responsibilities for the system, but monitoring is carried out by specialized agencies. Hence, the national Directorate of Water is responsible for flood forecasting, in collaboration with INGC and the National Institute of Meteorology (INAM). INAM and its regional centers are responsible for cyclone monitoring. Once southern Africa climate regional climate outlook Forum forecasts are released, the national institutions draw specialized forecasts, and INGC launches a contingency Plan preparation. The Technical Secretariat for Food Security and Nutrition platform (SETSAN) is responsible for the food security early warning system. SETSAN is composed of most ministerial institutions under the leadership of the Ministries of Agriculture and Health. It carries vulnerability surveys nationwide to assess community food insecurity and requirements for emergency relief. GTZ/Munich-Re and UNDP fund specialized flood early warning systems in the basins of Save and Licungo Rivers. The CENOE The National Emergency Operations Center, CENOE, under INGC, coordinates disaster response activities. A national civil Protection unit (UNAPROC) supports 288

CENOE to assist with search and rescue activities. INGC prepares annual Contingency Plans in a participatory manner involving central and regional government, donors, the UN System and civil society. The Plans are prepared following the issuance of the hydro- meteorological forecast by the Meteorological national institute in coordination the national Directorate of Water and consider four main hazards: floods, droughts, cyclones and earthquakes. They include a profile of the most vulnerable districts and priority needs. According to the scenarios established by the Contingency Plan, pre-positioning of goods takes place in the most vulnerable and least accessible areas. The early warning mechanism is refined and a national, regional and local simulation takes place, as a signal to launch Mozambiques disaster response. Training to local committees for risk Management is accelerated. In addition, CTGC weekly meetings are held to exchange information among disaster risk response stakeholders. The CENOE information team is activated to monitor information sharing among all disaster risk reduction institutions, including high-level decision makers who are members of the CCGC, chaired by the Prime Minister. INGC, UBDP, GTZ and INWENT are currently taking the lead in financing the strengthening and training of local risk management committees and the expansion of this network to other high-risk districts. MICOA The Ministry for the Coordination of Environmental affairs (MICOA) is the mandated UNFCCC focal point in Mozambique and designated national entity for climate change adaptation. MICOA finalized a National Adaptation Plan for Action (NAPA) in 2007. However, due to poor institutional integration with INGC, the country lead for HMDs, the largest impact of a changing climate in Mozambique, MICOA is not a major player on the institutional scene, and has poor visibility.

289


Figure 8.13: Mozambiques national institutional framework for the governance of climate- related disasters. Source: GFDRR, 2011.

8.2.2.2 In order to confront the multiple climate-related disasters that affect it, current as well as future, the Government of Mozambique (GoM) has undertaken vigorous steps in the area of CCA-DRR. It has also engaged serious measures to mainstream management of climate risks into existing development plans and strategies. 1999-2006: Adoption of National Master Plan for the Prevention and Mitigation of Natural Calamities Firstly, Mozambique adopted a national strategy in 2006: the National Master Plan for Prevention and Mitigation of Natural Calamities. The Master Plan followed the Disaster Management Policy of 1999 and became the countrys operative strategy for disaster risk management. It specifically emphasizes the links between development policies and disaster preparedness, prevention, mitigation and 290 Introspection into Mozambiques National Climate Disaster Management Policy

vulnerability reduction, addressing all of the elements of effective disaster management. Attention is paid to developing arid zones through the introduction of conservation agriculture and non-agricultural income generation activities, water supply and rainwater harvesting. For flood protection in risky areas, water resources infrastructure such as dams and dikes are included as key elements. Furthermore, the Ministry for the Coordination of Environmental Affairs (MICOA) finalized a National Adaptation Plan for Action (NAPA) in 2007. The plan, prepared by an inter-agency NAPA team, reviewed Mozambiques vulnerability to key hazards and identified four adaptation priorities: 1. Strengthening early warning systems; 2. Strengthening the capacity of farmers to deal with climate change; 3. Reduction of climate change impacts along the coastal zone; and 4. Water resources management. A national resettlement program coordinated by the INGC has also provided infrastructure for people to move from harms way, preceded by an assessment and identification of non-aedificendi zones across the country350. Despite all this progress, a number of critical institutional weaknesses remain: the 2008 Interim National Progress Report on the Implementation of the Hyogo Framework for Action cites weak institutional capacity to manage the relationship between Disaster Risk Management, and climate change and Environmental issues. The main capacity constraints are unresolved coordination issues between INGC and MICOA to address disasters as an environmental issue, and the fact that most line ministries lack a legal mandate to participate in the Master Plan (ISDR, 2009). A national disaster management law is in draft form, but has been awaiting ratification by parliament for a number of years. As a result the responsibilities of various government departments in disaster risk management are not yet clearly defined. Partially as a consequence of this, Mozambique continues to depend heavily

350 Expert interview, 2011

291

on international technical assistance to implement disaster risk management plans351. Mainstreaming of CCA-DRR into development plans: Disaster Risk Management is integrated, although not yet fully mainstreamed, into major development strategies in Mozambique. The Governments Five Year Plan (2005-2009) addresses some of the challenges related to disaster risk management and climate change adaptation. It identifies as priority objectives the reduction of number of human victims and amount of property loss, and it emphasizes a culture of prevention and mitigation. As part of the Plan, the Government committed to mapping zones at high risk, strengthening early warning systems, increasing resources for the prevention and mitigation of natural disasters, reinforcing capacities for inter-sector coordination, strengthening river basin management, establishing a database for information on climate change trends and impacts, promoting water storage systems in drought-prone areas, and increasing training and civic education352. The national Second Poverty Reduction Support Strategy (PARPA-II 2006-2009) recognizes disaster risk management as a cross cutting issue thereby acknowledging the need for a long-term strategy to reducing the vulnerability of communities and infrastructure exposed to extreme natural phenomena. Disasters are also part of the Medium-Term Fiscal Framework (MTFF). However, the priorities identified by the Master Plan for Disaster Prevention and Mitigation were not reflected in the PARPA II. With the release of the NAPA in 2007 and INGCs Study on Impact of Climate Change on Disaster Risk, adaptation strategies are expected to be much more closely mainstreamed into the next Poverty reduction

351 GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique
352 Ibidem

292

support strategy353. Disaster mitigation and enhanced resilience are specific objectives under the World Banks Country assistance Strategy. The Mozambique country Partnership strategy (2008-2011) specifies mitigation of risks from disasters and shocks as one of the objectives and enhanced capacity to respond to disasters as one of the outcomes under the pillar on sustainable and Broad-Based Growth. The establishment of early warning and emergency preparedness systems is specified as a goal. The Country Partnership also recognizes that future economic growth depends on the prevention of a major natural disaster. The Joint staff Advisory note, commenting on the PRSP progress, indicates a need to integrate disaster risk management in sectoral plans at all levels, and strengthen inter-sectoral coordination. While it compliments Government efforts in mitigating the impacts of climate shocks in 2007, it recognizes the financial limitations of the Government in facing major disasters, and therefore recommends the establishment of a national Disaster Fund, including mechanisms for risk transfer354. Furthermore, the Government annually provides USD $3.5- $5 million to INGC for disaster risk management and response, which may be increased depending on the magnitude of a disaster. This is equivalent to about 0.2% of the annual national Budget. International donors also fund the Contingency Plan. Additional resources are also allocated to other sectors for disaster risk management activities, such as irrigation schemes, small dams, construction of ponds and environmental protection355. Finally, a noteworthy major achievement is Mozambiques success in decentralizing disaster risk management. In Mozambique since 2006, provinces and districts have
353 GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique

354 355

Ibidem Ibidem 293

gradually integrated disaster risk management into their annual plans and budgets. The Government allocates direct financing to provincial and district plans in accordance with the Decentralization law of 8/2003. District land use plans have been developed by local governments (districts) with the support of provincial governments and integrated into District Development strategic Plans. However, provincial INGC delegations are still considered to be weak and need considerable support and capacity building to respond adequately to the numerous disaster risk management challenges. Furthermore, Mozambique has spearheaded one of the most noteworthy instances of community based disaster risk management practices in Africa: community-based DRM project in the Buzi District, with the establishment of a functional community-based earl warning-early action system and inclusion of traditional knowledge and community engagement, which has become a model. Finally, equipped and capacitated local disaster risk management committees have been established in all provinces across the country, as well as in most districts, the most basic circumscriptions in the country, albeit with various capacity levels. Despite these challenges, disaster risk management and adaptation to climate change have unquestionably become a central issue to economic development in Mozambique, and are expected to continue to grow in importance in the future356. A comprehensive program to reduce disaster risk and vulnerability Finally, Mozambique has developed a comprehensive national program to reduce disaster risk and factors of vulnerability to HMDs, with multiple projects ongoing. Table 8.7 reviews the major national programs on CCA-DRR ongoing, along with the funding donors and main objectives of each program component.

356 GFDRR, Disaster Risk Management Programs For Priority Countries Africa- Mozambique 294


Table 8.7: Ongoing projects and organizations in Mozambiques institutional framework on DRR-CCA. Source: GFDRR, 2009

Indeed, after completing a substantial number of background studies, the Government is now placing high priority on piloting risk reduction on key sectors. These include: (a) Flood protection measures such as dams, settlement protection dikes, and increased drainage in transport infra- structure. The Massingir dam recently rehabilitated in the limpopo river prevented floods in 2008 which could have affected chkwe and Xai-Xai cities, and small towns along the river basin. The dam was also perceived as having protected the largest irrigation scheme in the country in chkwe. Protection dikes have also been popular measures to protect settlements from floods: during 2007 and 2008, for example, dikes along the Zambezi were seen to have protected Luabo and Marromeu towns from inundation. New technologies for road construction using drifts and drainage have also reduced road cuts during floods. As a result, trade has become less affected and traffic has been re-established in the immediate post-disaster period, stabilizing food security and access to markets. 295

(b) Water management in arid lands, including the construction of small retention dams and ponds to increase water availability for irrigation purposes and for cattle in dry land areas. Investments in pilot adaptation measures for water management in arid lands, however, are not yet as developed as in flood-prone areas. (c) Coastal erosion control measures. These pilots are still incipient and closely linked to coastal inundation control. They are urgently needed in vulnerable coastal cities such as Beira, Maputo, Inhambane, and Quelimane. Given that the vulnerability of certain low-lying areas may leave no choice but to consider retreat, involvement of urban communities in participatory mapping is considered essential. (d) Social infrastructure using safe norms. Once hazard risk management guidelines are incorporated into building codes and infrastructure safety standards, the Government would like to promote safe pilots - such as model houses, schools, health centers and other social infrastructure. These would serve to show to communities and the private sector how their infrastructure can be protected against common hazards. As all the above points demonstrate, Mozambique has adopted a resolute stance to forestall the impacts of HMDs on its economic growth and communities. Why did it engage on such a course of action, while many of its neighbors (see red countries on fig. 5.6) do not even have as much as a nationally approved operative strategy for disaster risk management? We turn to this question in the next sections.

8.2.2.3 Overall Policy picture: Mozambique, an effective Disaster Averter By all accounts, our review of Mozambiques national plans to govern rising HMDs reveals that Mozambique indeed qualifies for all of the attributes initially hypothesized to characterize a Disaster Averter (see table 8.8). Mozambique has in practice equipped itself with one of the most sophisticated institutional frameworks on the continent to address rising high-magnitude HMDs that the country is faced

296

with. For the purposes of our comparative study, we can then proceed to use the country case as an adequate representative of the Green policy group.
Indicators of the Disaster Averters policy type A national Disaster Management Strategy or law has been adopted, with clearly defined and appropriated procedures and responsibilities for disaster response: The National Master Plan for Prevention and Mitigation of Natural Calamities (adopted 2006) In the national budget a dedicated budget line exists for disaster preparedness and emergency response: Government provides USD
$3.5-5million/year to INGC for disaster risk management and response (about 0.2% of annual national Budget)

Present? (Present: Absent: X)

A mandated agency for disaster management exists, able to coordinate across ministries and administrations and effectively command disaster response activities during an Emergency: The National Institute of Disaster Management (INGC) An annual Contingency plan is elaborated, and delineates the procedures, actions and responsibilities to be carried out in the advent of all possible disasters: The National Contingency Plan Regular drills of these disaster response procedures take place: Yes No more than three (3) days to deliver needed relief to disaster zones Disasters risks are treated as a crosscutting, transversal threat to development, inscribed in national Poverty Reduction Strategy Papers (PRSPs): CCA-DRR explicitly inscribed in the Governments 5-Year Plan (2005-2009) and in the Poverty Reduction Support Strategy (PARPA-II) Climate and meteorological risk information is used to inform disaster planning: Functional Early Warning System by INAM and SETSAN National/ sub-national contingency plan(s) based on seasonal climate/weather forecasts: Annual Contingency Plan elaborated Functional multi-hazard Early Warning Early Action systems exist Long-term disaster risk reduction plans are elaborated based on sound scientific research (risk identification, elaboration and enforcement of resilient building codes, mapping of risk zones, legislative bans on building/human settlement in risk zones, etc.): National Master Plan, National Resettlement Plan

297

Dedicated and sufficient resources are allocated for the implementation of DRR plans at all administrative levels: since 2006,
provinces and districts have integrated disaster risk management into annual plans and budgets. Government allocates direct financing to provincial and district plans according to Decentralization law 8/2003

Overall: Countries in this category take the threat of disasters very seriously and invest in building the countrys resilience to disasters, managing disaster risk with all means at hand. A country in this category has experienced a paradigm shift in its strategy to address disasters and transitioned from disaster Preparedness to Disaster Prevention. Country shifted the focus away from the disaster event itself towards the underlying risk factors that create disasters. Country in this category focuses on pre-empting disasters by reducing disaster risk and vulnerability so that Hazards do not turn into disasters, establishing solid weather forecasts, early warning systems, risk/vulnerability mapping, and other relevant tools/initiatives for climate risk management. Table 8.8: Mozambique, a Climate Disaster Averter indeed.

8.3 Incentives Analysis: which factors prompted institutional change for effective CCA in Mozambique? How were impacts of a changing climate managed in practice between 1995-2010? What factors explain Mozambiques policy discontinuity in its management of climate hazards? What was Mozambiques pathway of institutional change to address emerging climate risks? What does this reveal about the overall pathway of institutional change to address the new CCA challenge in Africa? These questions reside at the heart of our investigation of Mozambique. To answer them, we first retrace the Government of Mozambique (GoM)s responses to HMDs that made landfall in the country during 1995-2010, proxy for a changing climates impacts in this highly disaster prone country of the East African coast. 298

We then proceed with an analysis of government policies towards HMDs during the period of study, as these began to increase, focusing on the identification of critical break years during which change occurred in public policy towards hydro- meteorological disasters. Finally, we derive from our analysis the factors that may explain breaks in the continuity of public policies and institutions towards the management of HMDs. Our analysis rests on a thorough event analysis based on 78 newspaper clips that matched our keyword search on two of the countrys major newspapers (Noticias and J. do Domingo), catalogued in chronological order from 1995 to 2010. We then confronted the information from our events analysis with insights from iterative interviews with nine experts on DRR-CCA in Mozambique, who were pivotal actors in the countrys transition to Disaster Averters in the decade of the 2000s. We also consulted the secondary literature at great length, in search of areas of agreement and mutual corroboration. Only the findings confirmed by both sources of information (newsclips and expert interviews) are presented in the following sections. 8.3.1 Process tracing: Review of government responses to HMDs, 1995-2010

A quick glance over the table summarizing governments responses to HMDs (see table 8.9, Annex III) enables us to assert that multiple major HMDs have made landfall over Mozambique during 1995 2010 (see fig 8.14). The government of Mozambique (GoM), however, provided differentiated responses to these disasters.

299

1994: Cyclone Nadia (Nampula hit)

1997: Floods (Zambezi and Chire rivers); Droughts affect whole country, esp. Tete, Manica and Sofala

2000: worst vloods in history of Mozambique

2003: Drought and Cyclone Japhet (Central and South regions)

2005: Drought (in Central and South regions)

2008: Flooding

1995: Drought (Tete affected)

1998: Drought (6 provinces affected)

2001: Drought (Central and South regions affected)

2004: Drought affects 4 provinces

2007: Flood in 4 provinces , Cyclone Flavio in Inhamban e and Drought in Gaza

2010: Drought in 61 districts, Flooding in Central and South regions), storm and rains

Figure 8.14: Timeline of major HMDs in Mozambique since 1994: A major HMD hits at least once every other year, with multiple HMDs frequent within the same year.

We can generally divide GoMs responses to climate-related disasters into three distinct periods. Period 1 (pre-1997): Firefighters response During this period, the Department of Prevention and Combat of Natural Calamities (DPCCN), mandated Disaster Management unit, sits under the Ministry of Foreign affairs, a very revealing institutional arrangement. The DPCCNs main role is to distribute disaster relief funds and goods to disaster victims across country. Period 2: Transition towards a new institutional equilibrium (199799) 1997 marks the break year in Mozambqiues governance of HMDs. Following serious floods in the Zambezi and Chire rivers, and droughts of 1997, the Government distributes seeds and tools for the next agricultural season 1997/1998 (J. domingo, 7/27/97). The Assembly of the Republic then announced the measure

300

of creating food reserve stocks operating a minimum amount of 95 000 tonnes of maize and 5000 of beans. (J. Noticias, 11/22/97). The seeds of change in public policy towards HMDs are planted then, and will yield a solid tree capable of governing DRR-CCA in the next period. Period 3 (1999 onwards): Institutionalization of change in Mozambiques management of HMDs Indeed, 1999 marks the institutionalization of change in Mozambiques management of HMDs, with the dissolution of the DPCCN and creation of the INGC. In 2000, under Chisanos government, ex-ante efforts to prevent HMD impacts begin in Zambezia first. Then pro-active nationwide endeavors to prevent and mitigate HMD impacts take off in earnest beginning in 2001. Indeed, the GoM poses solid acts to this aid, with a few examples being: the approval and funding of a national contingency plan in end December 2011, the identification of risk zones and assessment of vulnerability to drought (the SETSAN study), local awareness building and empowerment on what to do when a disaster occurs by INGC, adoption of the national action plan against drought and desertification, the elaboration and adoption of the national contingency plan of 2005-2006, systematic pre-season risk screening, and starting in 2005 the of more structural measures such as infractustrure revamping to reduce vulnerability. The list goes on. In 2006, with the change of regime in favor of President Guebuza, we witness a continuation and deepening of period 3, culminating into the final adoption of the national Master plan for prevention and mitigation of disasters. 301

8.3.2 Analysis: explaining institutional change and policy discontinuity in Mozambique Mozambiques process of institutional change in its governance of HMDs began in 1997. 1997 indeed marks the break year when the seeds of change were planted, with the Parliament posing the first public act of disaster prevention: creation of food reserves to enable disaster victims to have a recourse in case a disaster comes their way, whatever the nature of the disaster. This marks a sharp difference with the practices of the then country lead entity on Disaster Management, the Department of Prevention and Combat of Natural Calamities (DPCCN), which sat under the Ministry of Foreign affairs with the main role of distributing disaster relief funds and goods to disaster victims across country. Parliaments decision was the first public sign of the desire for change in DRR-CCA governance. Two years later, in 1999, we see the institutionalization of this desire for change in DRR-CCA governance. Indeed, with the creation of the National Institute for Disaster Management (INGC) in 1999, President Chisanos Government decides to become very serious about HMDs. The new DM unit is placed under the Ministry of State Administration (MAE), no longer Foreign Affairs. In 2006, the president appoints a technically very able man as director of the new INGC (Paulo Zucula). Under Zuculas apt leadership and expertise, INGC is significantly revamped to fulfill its role as the country lead on DRR-CCA. We interpret from Zuculas nomination a strong political commitment at the highest levels of the Executive to make INGC functional indeed, Mr. Zucula, an agricultural expert, was tasked to lead INGC and come up with a functional vision. He was also given adequate resources to do so357 . Indeed starting in 2006, the national transition towards the pro-active anticipation and prevention of high-impact climate disasters is crafted and steered steadily but firmly. Zucula posed the following important acts. First under his leadership, the
357 Expert Interview, 2011

302

National Master plan on Prevention and Mitigation (DRR Master Plan) was adopted. Then, the Technical Counsel for DRM (the CTGC), an inter-ministerial framework, was established. CTGC still serves today at the Coordination Counsel for DM, an Inter-ministerial counsel of ministers, chaired by the Prime Minister. It is important to note that the INGC was revamped after 2006 also partly as a result of the Hyogo Framework for Action (HFA) impetus and funding streams following the Hyogo World Conference on DRR (expert interview, May 2011). Hyogo was a global confirmation that what Mozambique had already begun since 1999, was the way forward. The GoM engaged fully on this course thereafter. Mozambiques National Hyogo platform was established following the conference. Development of the new strategy towards more efficient management of disaster risks was also underpinned by adequate experience-sharing and peer-learning from other countries that had already successfully developed national disaster prevention and mitigation plans and early warning systems, as well as by extensive staff training and capacity building. Indeed interviewed experts mentioned multiple exchange visits in the early 2000s, upon the creation of INGC, to Latin American and Asian countries in order to learn how other countries also hit by multiple HMDs managed to successfully confront this problem (expert interviews, May 2011). Furthermore, a multitude of scientific studies on Mozambiques vulnerability to Climate Change and disaster related hazards were conducted by the Government of Mozambique, in alliance with external partners, consultants and technical assistants from all over the world. This explains why in the course of the present research a plethora of scientific studies on Mozambiques vulnerability to climate risks was readily available, unlike the other countries of Africa. Thus, knowledge truly drove and informed national planning and decision-making as Mozambique made its initial steps towards developing its strategy on disaster prevention and mitigation. 303

Finally, responsibilities for DDR were decentralized since 2006, devolving responsibility to provincial and sub-provincial delegations of INGC, and empowering local citizens to confront rising HMDs and reduce their vulnerability to them, all the way down to the local Committees for DRR established in each village of the country. We witnessed the existence of these local committees during our community field surveys. Although some of the local committees for DRR are troubled with severe capacity issues, many are fully functional and their members are active agents of Early warning early action in their communities. When Guebuza was elected to power in 2005, he continued his predecessors committed efforts towards more effective management of climate-related risks. By then, change was already institutionalized; INGC was an autonomous endogenously grown and well functioning institution, already with a budget line of from the State budget. INGC had most of all come to life in response to a national need. Though Guebuza politicizes disaster prevention funds to a certain extent, uses them as leverage for patronage, the DRR-CCA system still works insulated from political pressures358. In 2006, under Guebuza, the Decentralization Strategy was also launched, with UNDP support to the Ministry of Planning and Development. Gebuza gave the decentralization strategy all of its impulse; many purport in the aim of using the funds from it as political lever (expert interview, 2011). However the programs achievements remain impressive. Decentralization was under-funded until the programs inception. Today, 7million meticais (approx. $256,000 USD) are sent to each district, paid for by donor budget support, with the hope that districts will use this money to implement local development plans, including DRR-CCA. All in all, Mozambiques process of institutional change and adaptation to meet the new national challenge of rising HMDs appears to have been done right. The country
358 Expert interviews, May 2011

304

proceeded to address its challenge of rising HMDs by effective coordination (with INGC institutionalized to play the coordinating role on disaster management) and proper decentralization (expert interview, April 2011). All of the means, will and determination at the highest levels of the State were brought to bear to support the countrys new directional shift towards effective disaster prevention. From then on, Mozambique embarked on a stellar progress towards formulating and implementing its national plan for disaster mitigation and prevention: formulating contingency plans, dedicating resources to it and conducting annual simulation exercises or drills, constituting one of the rare success cases and countries to do so in Africa. A most unlikely candidate it was, emerging from twenty years of civil war, but Mozambique is well positioned today as one of the countries with the most sophisticated, integrated and functional institutional frameworks to govern rising climate risks and related disasters, spanning timescales (from disaster anticipation, to response and rehabilitation) as well as geographical scales (from national to local levels). High dependence on donor funding and support to run the national machinery on disaster prevention (notably the Early Warning Systems), however, is a major constraint, which the country still has to overcome359 . 8.3.3 Conclusion: Identification of relevant Independent Variables From our analysis of Mozambiques dazzling ascension to the category of a disaster averter in less than ten years, between 1997 and 2006, what factors seem to have been determinant in the countrys change of institutions to govern HMDs? The following competing variables appear relevant from the Mozambican case:
359 Expert interviews, May 2011

305

1- Increase in the recurrence of high-impacts HMDs starting in 1994 Indeed, Mozambique was faced with a sudden rise in the number and intensity of the major HMDs beginning in 1994, with up to three major disasters impacting its coast and mainland simultaneously in a year (see fig. 8.14). This sheer increase in the lethal and destructive force of HMDs placed the issue high on the national agenda. In turn it led to: - - Development of a national desire for change in the governance of rising climate risks in 1997 Institutionalization of this national desire for change staring in 1999, with notably the creation of a mandated, funded and capacitated National Institute for the Management of Disasters (INGC) and then in 2006, the acceleration and finalization of institutional reforms to better manage HMDs making landfall in the country; - There were thus two moments of change in Mozambiques pathway towards Effective climate disaster management: first the 1997 -99 period of abrupt institutional change or institutional revolution which led to the creation of the INGC, followed by a period of incremental changes and adjustments leading up to the adoption of 2006 National master plan. 2- Donor and partners support First, support came through the UN system (UN-ISDR and then through UNDP), as well as the various NGOs and UN agencies that flocked to Mozambique after the 2000 flood event and rescue operation. Indeed Mozambiques national plan was and remains almost entirely donor funded (expert interview, 2011). 3- And lastly, the enlightened leadership of President Joaquim Chisano. Let us now review each of these factors, in light of our findings from previous cases. Agent-based explanations 1) Enlightened Leadership of Chisano as driver of 1999 institutional revolution 306

Indeed in the words of one of the interviewed experts, the key to all that has happened in Mozambique and is still to happen is political will and leadership. This comes from the times of Chissano, who won the Mo Ibrahim Prize for good leadership360. As nicely put by another expert interviewed: People in Mozambique have suffered a lot during the years of the civil war and with recurrent disasters. In one way or another we know somebody who has been affected and that makes hearts bleed. Before the war, mistakes were made about the economic model chosen for the development of the country, but the government had genuine intentions to develop the country (authors note: under Samora Machels leadership). If you look at the strategic plans developed then, you will be amazed with the vision, but of course they were never materialized. Today, things have changed a lot, but there are few people who still want to genuinely push things forward. If you have such people making decisions, then you can move mountains (expert interview, May 2011). Chisano, the broker of the peace deals that extirpated Mozambique out of its bloody 20-year civil war, was, in all likelihood, one of such. Freedom warriors of Mozambiques independence indeed had a high sense of national duty, and a strong desire to answer peoples concerns and pull the country out of poverty after years of suffering under the war. FRELIMOs position as overseer of the nation also played a role. In any case, political determination was high to lift the country out of poverty. There was a genuine desire to now live, and make the country live on despite all the adversities of humans and Nature. Thus the history of suffering was important in generating leaders who had enough of the war and desolation, and wanted to live, and the country to live on. This could potentially explain why Chisano was so committed to the CCA-DRR agenda. Tired of seeing the country brought back by so many disasters, after having fought for this
360 Expert interviews, May 2011

307

country, it follows that he could pioneer a national strategy to reduce disaster impacts on the citizenry and embark on road to more effective management of disasters to reduce toll of disasters on human activities, livelihoods and national welfare. The concrete acts posed by Chisano during his stewardship of the country between 1996-2006 make it very plausible as well that his leadership was the determining factor in the countrys transition towards disaster prevention. He demonstrated irreproachable political leadership by establishing the INGC and mandating the drafting of the countrys disaster Master plan, finalized after his term, in 2006. Chisanos successor completed the process paved by Chisano by appointed Zucula as the head of the new INGC. According to one of the experts interviewed, who was a public servant under Chisanos lead, Chisano in his stay at the helm of Mozambique, had one merit: he appointed people according to their Technical capacity, not by political power. He was very supportive of DRR-CCA. He also delegated a lot of power to good Prime Ministers, whom he handpicked because they were capable/powerful and stepped up to the task in times of emergency. Today ministers are only appointed because of because of political loyalty361 This agent-based explanation, placing the spotlight on President Chisanos admittedly apt leadership of the countrys institutional adaptation between 1997 and 2005 to meet the new challenge of CCA, has to be tested against however the more structural levers that may have precipitated Mozambiques transition to the Disaster averter category. Path-dependent explanations (Accidental factors) 2) Abrupt rise in the propensity of disasters
361 Expert interviews, May 2011

308

The sharp increase in the severity and frequency of HMDs since 1994, with multiple HMDs in a year, also constituted a key trigger in bringing about political will, and placing the issue of HMDs high up the political agenda. Needless to say that Mozambique has no choice, but to adapt. Where it is located, if it is not a tropical cyclone, it is a flood or drought (expert interview, May 2011). As such, it appears that it is those who are recurrently battered by disasters who tend to develop the frameworks to react, going from doing things on ad hoc basis every time there is a disaster to anticipating disasters and responding through a coordinated system. If this explanation holds, it is because Mozambique began to experience rising disasters in 1994 that in 1997 it began to consider alternative ways to tackle disasters, above and beyond mere response. 3) Donor support and funding Indeed Mozambiques national disaster prevention plan was and remains overwhelmingly entirely donor funded. Mozambique only covers 45% of its annual budget, the rest is donor funding. Donors in 1999 demanded a new tool to better manage disasters, and provided funding to enable this (see annex 8.1). The combination of genuinely determined leadership with trusting providing donors gave birth to new effective socially beneficial institutions, or rules of the game, to manage HMDs, from national to local levels. This however is a rare congruence of circumstances. Indeed, donor funding seldom recognizes the primal importance of DRR and links between DRR-CCA and development, preferring investments in classic development areas such as such as education, health, water and sanitation, infrastructure, agriculture, etc. Even as DRR is recognized as an urgent need, donors tend to shy away from investing in disaster preparedness as funds are more easily diverted when invested in such activities and they do not trust their recipient countries. Indeed: On the other side, donors are suspicious of governments and so they think that any money put for preparedness could be diverted to other pressing needs. This is why some of DRR activities are carried out by NGOs (who are 309

trusted by donors). Basically donors who support countries did not give the incentives for governments to be disaster averse. One may argue that it is not the donors fault, but most governments dance along the tunes played by donors.362 Equally rare is the presence of genuinely dedicated leaders. Indeed leaders seldom plan beyond 4-5 years of their electoral cycle. But when the combination of genuinely determined leaders and trusting donors obtains, funded socially beneficial policies, and institutions to implement them, ensue. This combination obtained in Mozambique; and, as eloquently stated by Patt, donors were ready to fund good projects in underserved areas, and the government was determined to get these funds and use them to reduce poverty (INGC, 2009). In all likelihood, in Mozambique the sheer disastrous extent of HMDs and their impacts, notably since 1994, made the case for urgently placing DRR on the development agenda. High frequency of multiple high-impact HMDs created a national demand for effective and socially beneficial institutions to manage them, which in turn drove Mozambiques institutional revolution in disaster management, supported by external funding. 8.3.4 Mozambiques pathway of Institutional Change in DRR-CCA

Amidst the multiple possible explanations for Mozambiques institutional change to meet the new challenge of rising HMDs, it appears that the most determinant variable is the abrupt increase in the propensity of HMDs in 1994. The presence of committed leadership was important to lead on the issue, however the sheer recurrence of high-impact HMDs made the effective management of climate-related risks a national necessity, which in turn placed the issue high on the national agenda
362 Expert interviews, April 2011

310

and made it rise to prominence. Thus, Mozambique had no choice but to adapt adapt the way it manages climate disasters with a higher emphasis on prevention, efficiency in response and disaster risk reduction in rehabilitation or perish. Donor trust in Mozambiques leadership and support for the 1999 DM institutional reform also played a key role in driving the initiative forward, giving the GoM the means to carry to fruition its newfound desire for new and better ways of managing HMDs. Thus, when committed leadership in Mozambique was confronted with the national issue of intensifying large-scale HMDs, and donor support was present, this congruence of circumstances led to the birth of an effective, socially beneficial institution to manage HMDs.

311


PART IV:

DISCUSSION & CONCLUSIONS

312

 CHAPTER 9: CROSS-COUNTRY COMPARATIVE ANALYSIS: What Triggers Institutional Change/Adaptation To Meet The New Challenge Of Climate Change Adaptation? What do we know about the Governance of Climate Change risks and disasters from our three country cases in Africa? All three countries reviewed, Senegal, Uganda and Mozambique, are competitive electoral democracies, where elections sanctified by international observers, are held every 5 to 7 years. In the context of hypertrophied executive branches across our three country cases that can create new districts at a whim to secure additional electoral wins (as was the case in Uganda in 2009-10 and in Senegal in 2010) and which, in most cases, are not checked by counter-powers neither in the National Assembly nor in the judiciary branches which the Executive controls, and where, furthermore, the Party (e.g.: Frelimo) is the State, it is evident that public goods delivery in general, and delivery of effective disaster management in particular, has to come first and foremost through the will of political leaders. This brought us to ask the critical question: what generates political will to tackle rising climate risks? 9.1. Trajectories of institutional change in the face of rising HMDs Compared In Senegal In Senegal, the government only responds to disasters, and inefficiently so (Unprepared Firefighters). In times of national climate crises between 1995-2010, the State most often abdicated and allowed NGOs and the UN system, better organized, to take over, despite the existence on paper of National Disaster 313

Response Plans (the Plan ORSEC) and procedures. Donor pressure was present, mainly from the World Bank after the 2009 flooding in Dakar for the third consecutive year that external financial support was sought to respond to disasters, but this factor was not the impetus for significant change. The only instance of public good provision occurred when the flood afflicted citizens of Dakars suburbs, a key electoral bastion, sanctioned the ruling party through the ballot at local elections in 2009, bringing to power opposition mayors and leaders, and maintained the pressure through street protests. This local demand for effective governance of flood disasters led to the delivery starting in 2010 of a meaningful flood combat program for Dakar. Such public good provision was only forthcoming for Dakars suburbs however, and was not generalized into a national flood prevention program. We learn from the Senegal case that donor pressure or support alone is not sufficient to trigger institutional change. Impetus for institutional change has to emanate from within and has to be a strong local demand. It has to be endogenous, and locally grown. Our review of the systematic factors that thwarted meaningful action on HMDs in Senegal suggests that, in order for substantive institutional change to take place in the governance of HMDs, the following conditions needed to obtain (see fig. 9.1): - - Local / national demand for improved CCA-DRR Donor support

314

LOCAL demand for effective vlood management

Donor support

Political will to address CCA-DRR

Fig. 9.1: Suggested triggers of institutional change from Senegal case

What makes Uganda implement, where Senegal does not? From the Ugandan case, another causal factor was added to our analysis of the triggers for institutional change in the face of climate-related disasters: the frequency and severity of HMDs hitting the country. Indeed in Uganda, HMDs increased by a staggering 7.7% between the period of 1960 1994 and 1995 2011. A direct result of this was the desire for improved organizational forms to tackle disasters more effectively. A first attempt at institutional reform in the governance of disasters was attempted in Uganda in 1995, following the severe earthquake of 1994, a non-climate related event that still shook the nation to the importance of apt disaster preparedness, with constitutional provisions established for the creation of a Commission on Disaster Preparedness and Management. However the momentum for such an initiative was lost after three years without any disaster hitting the country between 1995-1998. Efforts at meaningful changes in the rules of the game to address disasters (climate or otherwise) only took off in earnest in 1999 again, when the country had to address a massive internally displaced peoples (IDP) problem, bequeathed by the a decade of civil war in Northern Uganda. After 1999, a new 315 In Uganda

national institution was established, merger between pre-existing institutions, creating the Ministry of Disaster Preparedness, Management and Refugees, with a dedicated department on Disaster Preparedness and Management (DDPM), sitting under the Office of the Prime Minister to ensure effective cross-ministerial coordination on the now prominent issue of disaster management. 1999 in this sense marks Ugandas institutional revolution to govern disasters properly, one consolidated in the decade that followed through incremental changes and adjustments for more effective delivery on the CCA challenge. All these institutional adaptations culminated into to the creation of the NECOC, dedicated disaster response operations center to play a leadership role during disaster relief operations with assimilated pre-rehearsed procedures, and the adoption of the National Plan on Disaster Preparedness and Management in 2010. Both Senegal and Ugandan cases thus appear to suggest that the country-driven- ness of institutional change is fundamental, and that key is a countrys appropriation of the desire to adopt effective rules of the game to govern the new issue of Climate Change. This desire under both instances is, itself, driven by the sheer increase in the frequency of HMDs (at the localized level in Senegals Dakar suburb and nationally in Uganda), which places CCA-DRR and the search for more effective institutional arrangements to manage and prevent the impacts of HDMs as a priority on the national agenda. But it becomes clear that national leaders first had to experience the effects of not addressing HMDs properly and not climate-proofing development grains from the rising impacts of climate risks (climate-compatible development), before they began to offer seriously engage with the battle of HMDs. In the context of Ugandas hypertrophied Executive branch, where the same president, Yoweri Musevini, has been in power since 1986, this finding is not surprising. In a patrimonial system revolved around personal rule, the one leader and executive manager of the patrimony needs to perceive the dangers and risks to the patrimony caused by HMDs, before he can react decisively and place the issue high on the national agenda. 316

In Uganda, we purport that national will to tackle rising climate risks was built by the rise in frequency and severity of high-impacts HMDs floods, droughts and landslides in the 2000s decade, which increased by a staggering 7.7% relative to the 1960 1994 period. As we will see shortly, this pattern was also replicated in Mozambique, where multiple HMDs also took the country by surprise, and the State was left with no choice but to confront the problem, and devise new institutions or rules of the game, to handle multiple HMDs.
Country driven-ness (response to rising HMDs)

Effective Policy on Disaster Preparedness & Prevention

Support of development partners

Fig. 9.2: Suggested triggers of Ugandas transition to a Prepared Firefighter

Finally, in Mozambique we discovered that an additional factor, agent-based this time, confirming Grindles hypothesis of institutional change, was important in the formation of political will to tackle rising HMDs: leadership. Indeed the Mozambican case demonstrated that the presence of a dedicated leader, namely President Joaquim Chisano who ruled the country from the end of the civil war in 1992 to 2005, was also a possible determinant of institutional change important to consider. Abdoulaye Wades pro-active response the 2005 national flooding crisis in Senegal supports this theory. Five years into his electoral honeymoon with the Senegalese 317

In Mozambique

people, and faced with the sheer extent of the 2005 floods which affected 200,000 people, killed 30 and displaced 1,700 households, Wades government took the 2005 flooding very seriously, interrupting Government vacations, making a blazing declaration of a national state of crisis and endeavoring the construction of a new city for the relocation of Dakars recurrently flooded. However all the hustle and bustle over flood management stopped in 2005, and was not carried forward beyond that point. Senegal indeed did not become a better-prepared Firefighter as a result of the 2005 disaster, nor did it learn from its mistakes then. As such, we find that these agent-based explanations of institutional change are indeed subordinate to the more structural explanation of heightened frequency of HMDs, as a better explanation for institutional change and national evolution from one policy category (red) to another (yellow or green).
High frequency of multiple HMDs

Genuinely determined leadership

Effective socially benevicial instition to manage HMDs

Donor support

Fig. 9.3: Triggers of institutional change in Mozambique in DRR-CCA

Also from both Uganda and Mozambique, noteworthy is the common trend that once the country decided to become serious about preventing the destructive impacts of HMDs on national welfare and community livelihoods, those in charge of drafting the National policy proceeded to go and seek outside expertise to learn how to do this right and what institutions to devise to adeptly handle the new nation 318

issue. Thus the oft-times upheld argument of lack of expertise and knowledge about the new issue of Climate change and related risks does not seem to hold. When faced with a national issue of HMDs jeopardizing national welfare and the sustenance of the entire patrimony, Uganda and Mozambique decided to become serious about the management of climate-related risks. They embarked on meaningful institutional reform and adaptation to deliver efficiently on disaster preparedness in Uganda, and prevention in Mozambique. Again here, the lower propensity of disaster occurrence in Uganda relative to Mozambique appears as the most significant driver of why the former embarked on a course of disaster preparedness only, whereas the latter on preparedness and prevention. Will Uganda sustain its progression towards effective disaster risk management and reach the category of a Disaster Averter if HMDs do continue to rise? Time will have to tell. However what is clear is that given the choice between Adapt and Perish, these countries chose to Adapt. Senegal on the other had, experiencing much less disasters per annum, still has not experienced a national paradigmatic shift in its management of climate-related risks. Variable
HMD propensity 1960-2011 % change in average number of HMDs (1960-1994 vs. 1995-2010) Institutional Change to deliver

Senegal
26 +0.2

Uganda
29 +7.7

Mozambique
58 +1.1

None- Beginning in 1999 Beginning in 1999 Only instance of with the creation of with the creation of public good the Ministry for the National provision on CCA- Disaster Institute for Disaster DRR: in Dakar Preparedness, Management (INGC), suburbs, recurrently Management and finalized in 2006 hit by flooding since Refugees, finalized in with the adoption of 2000s. 2010 with the the National adoption of the MasterPlan on the National Plan for Management of Disaster Natural Calamities Preparedness

Table 9.1: Cross-country findings on the correlations between HMDs and institutional change to deliver on CCA-DDR: Countries have no choice but to Adapt or Perish.

319

Thus overall, our country cases reveal that: 1) The largest driver for institutional change in the governance of Climate Risks is national demand for new institutional framework to manage rising climate disaster impacts. This idea is expressed by our concept of country- drivenness of the demand for a national covenant, from local to national levels, to manage HMDs efficiently. This national demand or country- drivenneess appears to be triggered by the attainment of a tipping point in the rise in frequency and/or magnitude of HMDs. In countries that have experienced more intense HMDs, there is a demand to manage disasters better; 2) Secondary independent variables are donor pressure and an agent-based explanation (the presence of a dedicated leader). As such, government will try to shirk its constitutional responsibility to deliver effective DRR-CCA, until it can no longer be avoided. Faced with the ultimatum of having to, political leaders will adapt, and seriously confront Climate Change and devise new rules of the game to address, and manage, efficiently the issue of HMDs, and climate impacts more generally. This determines a course of gradual institutional adaptation as countries are increasingly faced with more frequent and high-impact climate-related risks and disasters. It bodes negatively though of countries abilities to engage in ex-ante action until the projected climate impacts have started to hit them. This leads to a theory that countries in Africa, and elsewhere, will not take climate impacts seriously until these begin to manifest themselves. They will not learn form the age- old wisdom that prevention is better than cure. However when the projected predicament will hit them, then countries will scramble to get on their feet and seriously begin to tackle the problem, finally implementing all of the known recommendations for addressing climate impacts. 320

Faced with the ultimate choice of Adapting or Perishing: they will finally choose to Adapt. However, not before the ultimatum gets down to that. 9.2 Discussion of results Our findings from Senegal, Uganda, and Mozambique leave us with interesting prospects about why governments engage, or not, with the rising issue of CCA. 9.2.1 Why not? Let us begin with the reasons why governments appear to NOT engage with CCA- DRR. Indeed, it appears most countries have been reactive to disasters because: 1) They are always straggling with financial resources. Donor funding is critical to support willing countries to transition towards anticipating climate risks and opportunities and preventing their negative impacts on vulnerable communities. Our 3 case studies reviewed in this thesis provide ample evidence for the urgency of supporting such countries with technical assistance to conduct scientific assessments of risks and vulnerability, as well as funding to craft and implement national climate disaster mitigation and prevention plans of action. With CC projections for increased variability in the climate system of Africa and already witnessed more frequent HMDs, this need is of utmost importance for countries of Africa. However, it has appeared from this thesis that paramount, is the willingness of countries (and their national leaders) to embark on such a course of action. Our case countries teach us that countries need to first want to reform their policies and re- craft their institutions and coordination roles, towards a more effective management and mitigation of HMDs. 321

In Mozambique such a willingness came from the countrys history of suffering during the 20 years of civil war, which gave rise to leaders in the post-war era who were determined to live, and make the country live on, and extirpate it out of poverty, destruction and desolation. Thus President Chisano, broker of the 1992 peace accord, was also the pioneer of Mozambiques HMD institutional change. Under his political leadership and vision for the country, evidenced through his creation of the INGC, appointment of a competent technical person to spearhead the countrys new institutional/policy reform towards the management of HMDs and brokering of strategic partnerships with donors and partner countries in the implementation of the national strategy for DRR-CCA, Mozambique made significant strides towards overcoming its vulnerability to disasters and properly managing disaster risks and opportunities, in a manner meaningful for local communities at the frontlines of a changing climate, making it comfortably sit today as one of the leading countries in Africa in terms of its progressive policy towards CCA-DRR. An unlikely candidate it was after 20 years of civil war, but as nicely summarized by one of the interviewed experts: If you have people who still want to genuinely push things forward making decisions, then you can move mountains. 2) Most governments in Africa are however Itinerant. They plan only for the 4 or 5 years of their mandates. Abiding by electoral motives and caught in a logic of power consolidation, political leaders most often try to (i) please their constituencies by doing things that they can show (ribbon cutting) and (ii) do things that will please their partners by addressing first degree development challenges such as health, education, infrastructure, etc. They dont make the link between these issues and disasters or climate change adaptation. With this short sighted-ness though, politicians get the support from their constituencies and the legitimacy from partners and so they live on. There are very few countries with long-term development plans and/or strategies. Because of that they often tend to do things that would account to improving the Human Development Index, MDGs and other internationally agreed targets. 322

Unfortunately, these targets do not establish a clear link with DRR. As we have shown in our cross-country classification, which served the basis of our case selection in this research, most countries in their Poverty Reduction Strategy Papers (PRSP) make a very faint reference to disaster risk management of climate change adaptation. Finally there seems to be a positive relationship between HMD frequency and implementation of optimal and socially beneficial policies to address HMDs. Indeed a quick survey of the six countries in Africa that are disaster averters (see fig. 5.6) shows that most of them indeed have in place the right institutions and frameworks to mitigate and prevent disasters, relative to 26 disasters in Senegal or 27 for Uganda. Thus, the rise in disaster frequency emerges as the msost critical driver of CCA-DRR is pushed on the agenda, creating the national demand for effective CCA-DRR policies that enables that push. In Mozambique for instance, it is this sheer national need for policies to prevent the ample devastating impacts of the droughts, floods and cyclone that hit the country intensively since 1994 that triggered the institutionalization of disaster preparedness, and led to the creation of a dedicated national Institute (the INGC) in 1999, endogenously devised and grown to cater to this new national need. Over the next five years, the INGC grew to become the integrated and well-functioning organization, coordinator of Mozambiques new institutional framework to manage HMDs. As such, we can surmise that it is indeed those who are recurrently battered by disasters who tend to develop the frameworks to react363, going from doing things on ad hoc basis every time there is a disaster to anticipating disasters and responding through a coordinated system. And the frameworks they develop, one
363 Expert interviews, April 2011

323

could add, have a higher likelihood of being durable because country-driven and endogenously grown. Both Northian and Boonian thinkers agree on this critical point in institutional change: It needs time to soak, and grow. New institutions have to grow out of a national demand for them, either accidental or voluntary, that mandates their existence and creates the circumstances for agreeing on new rules of the game to manage the newly expressed societal need. Picking on the example of Mozambique again that successfully revamped its institutions for CRM, increasing HMDs put the need for a national strategy to manage HMDs on the agenda as a new national need. Perhaps as countries get increasingly negatively impacted by HMDs, they too will create the conditions necessary for a new national consensus on climate disaster risk management to obtain. 9.2.2 Why then? Let us now turn to the pre-conditions for meaningful institutional adaptation to meet the CCA challenge. Thus, the necessary conditions and most critical triggers of institutional change on CCA-DRR appear to be from our cross-country comparative assessment of the determinants of institutional change for climate change adaptation, arranged by order of importance: 1) The change in the frequency high-impact climate hazards (High disaster frequency/magnitude) 2) A new climate risks/disaster management institutional framework becoming a nationally accepted urgent need and demand (country-driven-ness); 3) Donor and development partners technical and financial support to carry forward countrys desire to formulate a new national institutional framework 4) The presence of a dedicated leader, insofar as it supports governance of all issues, CCA included. 324

9.3 Limits of the analysis How valid are our findings from our comparison of Senegal, Uganda and Mozambique? The following challenges to internal and external validity have to be taken into account, when building a general theory of the triggers of institutional adaptation to meet the new challenge of CC. Competing variable 1: If the single most driver was increased HMDs, then Ethiopia would have also had effective CCA-DRR institutions As table 9.2 evidences, Mozambique is not the only country suffering from a high propensity of HMDs. Madagascar, another Disaster Averter, still confirms our theory. However, Ethiopia as a matter of fact has been hit by a higher number of HMDs between 1960 and 2011. Why is Ethiopia then not a Disaster Averter, too, but a mere Prepared Firefighter?
Madagascar 6 0 6 45 57 14,25 Mozambique 12 0 29 17 58 14,5 South Africa 8 2 30 21 61 15,25 Tanzania Uni Rep 9 0 31 4 44 11

Drought Extreme temperature Flood Storm TOTAL AVERAGE

Ethiopia 14 0 49 0 63 15,75

Morocco 5 1 28 2 36 9

Niger 8 0 15 1 24 6

Senegal 6 0 17 3 26 6,5

Uganda 9 0 16 4 29 7,25

1960-2011 Source: EM-DAT

Table 8.3: Mozambiques exposure to HMDs, as compared to a sample of other African

Two elements allow us to effectively counter this argument. For one, we contend that institutional change only occurs after a certain threshold of additionally in HMD impacts is reached, basically when HMDs begin to recur beyond a certain rate and multiple HMDs hit the country simultaneously or successively in a short period of 325

time (as was the case in Uganda and Mozambique between 1994 2010). The main driver remains then the rise in HMDs in recent years, since the mid-1990s to be more specific, and not just HMD propensity. Ethiopia however also suffered from a higher increase in HMDs between 1960-1994
and 1995-2010, with 1.3% while Mozambique a 1.1% increase. Thus, suffering from a

higher propensity of HMDs and a higher rise in HMDs, why did Ethiopia not embark in meaningful institutional reform in climate disaster management whereas Mozambique and Uganda did? The Ethiopian exception calls our attention to possible other drivers, and gives additional credence to the Agent-based explanation of the presence of good leadership and regime type. Indeed, whereas all of our reviewed cases semi- democracies where local populations and average citizens still have leverage on the governance of national affairs, Ethiopia remains strict under authoritarian rule. Competing variable 2: Regime type determines Indeed a glance at figure 5.2, the map of African countries according to national disaster management policy, confirms that most Disaster averting (green) countries appear to be competitive democracies, or some form of hybrid-democracy under consolidation: South Africa, Mozambique, Madagascar, Mauritius and Cape Verde. The notable exception that enervates the regime type explanation however is Algeria, a firm Disaster Averter, even preceding Madagascar, Cape Verde and Mauritius. Also, an overview of Yellow countries that engaged a process of institutional change to better deliver on disaster preparedness demonstrates that most countries under this category are actually not democratic countries, but rather under some form of autocratic rule: Morocco, Sudan (prior to the countrys scission), Tunisia (up until the closure of the classification ranking for this study) and Ethiopia. 326

 Thus, the regime type explanation cannot be the significant driver of institutional change in the face of rising HMDs. Competing variable 3: geo localization Could the clustering of Disaster Averter countries along the Southern-East African cyclone track be the single largest driver? Indeed, figure 5.2 also hints to a competing IV, that Disaster Averters are mainly clustered on the East-Southern African coast, on the cyclone track. As such, geo-localization on high HMD areas could be the true driver here. What of Algeria, however? The presence of Algeria, located as far North and West as possible away from the East-Southern Africa disaster hub debunks this claim, and makes us look elsewhere for more meaningful explanations. Competing variable 4: National fiscal capacity to understand CC risks and engage in meaningful action to prevent CC impacts at the national level Do the rich countries of Africa alone implement CCA plans? A look into the Mozambican case, a country with a GDP/capita low as $428, ranked, easily debunks this argument. Seemingly, if Mozambique managed to hoist itself as the most stellar and effective implementer of DRR-CCA policies on the continent, then national capacity is ruled out as possible driver of institutional change on CCA. 9.4 Towards a Theory of Institutional Adaptation to Confront Rising HMDs In Africa From our comparative study of the triggers of Institutional Change in Africa to meet the new challenges of Climate Change, it remains that three fundamental factors drive institutional change on CCA. 327

Our cases suggest that countries engage in meaningful CCA-DRR when the following conditions obtain: Increased frequency of national HMDs, creating a country driven-ness for Institutional change to meet the new CCA challenge + Available funding and trust from donors + to minimal extent, the supportive presence of leaders animated with a strong desire to safeguard country and its citizens (agent based explanation; see fig. 9.4).

Country driven- ness

Donor Support Effective Policy on Disaster Preparedness & Prevention

Fig. 9.4: Triggers of institutional change from our cross-Africa comparative study

Both Senegal and Ugandan cases suggest that the country-driven-ness of institutional change is fundamental, and that key is the countrys appropriation of the desire to adopt effective rules of the game to govern the new issue of Climate Change. National leaders need to first and foremost want to lead their country forward on a track of development proofed from impacts of climate risks, whether voluntarily or as a result of local pressures on them to do so. We have found that in all our cases these pressures had to be local/nationally-driven they cannot emanate from outside. The country needs to want this first, and it needs ot be justified in the context of countries, or else the process will not be sustainable. 328

In Senegal, country-drivenness on the CCA-DRR issue was forthcoming only in one area of the country, and it was built through an alignment of political incentives with policy imperatives, when the critical voting bastion of Dakars suburbs were hit by flooding over multiple years. In Uganda and Mozambique, national will to tackle rising climate risks was driven the rise in frequency of high HMDs (a +7.7 exponential rise in Uganda). In Uganda, the need and a drive for more efficiency to manage the increased disasters the country became faced with in 1998 onwards drove the countrys institutional revolution in CCA-DRR. With more HMDs in sight, there was a need for better more efficient institutional and organizational ways to manage them. This finding is also not in the context of Ugandas hypertrophied executive power, where the same president, Yoweri Musevini, has been in power since 1986. In a system revolved around one-man rule, the one leader and executive manager of the patrimony needs to perceive the dangers and risks to the patrimony caused by HMDs, before he can react decisively and place it high on the national agenda. In Mozambique, a dedicated leader (Joachim Chisano) played a critical role. There, just as in Uganda, leadership sent its technical staff abroad to learn from countries that had already succeeded in establishing effective institutional mechanisms to forestall the impacts of Climate Change. Indeed in both instances, those in charge of drafting the National policy even had to go and seek outside expertise to learn how to do this right and what institutions to devise to rightfully handle the new nation issue. When given the choice between Adapt and Perish, these countries chose to Adapt. This determines a course of gradual institutional adaptation as countries are increasingly faced with rising climate-related risks and disasters. Also it is important to note that institutional adaptation appears to occur in two different moments: 1) Institutional revolution: following abrupt rise in HMDs reaching a tipping point in 1999 in Mozambique and Uganda 329

2) Incremental institutional adaptations and adjustments This determines a sequential course of institutional change, driven by rising HMDs, as follows:

Moment 1: Institutional Revolution (after tipping point is reached in rising HMDs)

Moment 2: Institutional Consolidation: Incremental changes and adjustments to proposed Institution

Moment 3: Agreement over and Adoption of New Institution by all players

Fig. 9.5: Course of Institutional Change, as gleaned from to the case of Institutional adaptation to meet the new CC challenge in Senegal, Uganda and Mozambique

This bodes negatively though of countries abilities to engage in ex-ante action until the projected climate impacts have started to hit them. This theory claims that countries in Africa, and elsewhere, will not take climate impacts seriously until these begin to manifest themselves. They will not learn form the age-old wisdom that prevention is better than cure. However when the projected predicament will hit them, then countries will scramble to get on their feet and seriously begin to tackle the problem, finally implementing all of the known recommendations for addressing climate impacts. Faced with the ultimate choice of Adapting or Perishing: they will finally choose to Adapt. However, not before the ultimatum gets down to that. For instance, will Uganda then sustain this course, and continue its transition up the spectrum of effective disaster risk management all the way to the green Disaster averter category? Will Senegal trigger nation-wide revamping of its institutions for disaster prevention as HMDs intensify? Only time will tell. If more and frequent intense disasters occur, and Uganda intensifies it efforts to transition towards ex- ante prevention on top of its existent preparedness, then our theory will be proven. 330

CHAPTER 10: PERSPECTIVES ON CCA-DRR IN AFRICA TOWARDS A THEORY OF CLIMATE CHANGE INSTITUTIONAL ADAPTATION My dissertation aimed to answer the fundamental question: how are countries in Africa governing rising HMDs (and by extension climate change), and what makes them adopt the institutions that they do to confront their changing climate? Our fundamental related question of interest is what can be done to curtail the impacts of CC on African countries, by countries, and ensure optimal governance of climate change risks and opportunities, in order to thwart their negative impacting of the development endeavor. A preliminary answer seems to be: build resilience to current climate variability, since variability is predicted to increase with CC. Preliminary findings however indicate that only a very small minority of Africas fifty-countries are doing so (see cross-continental map of African countries by CCA-DRR policy type on fig. 5.2). Finally we were interested in knowing why countries in Africa adopt the climate disaster risk management policies that they do. What are the incentives that prompt /deter states from adopting proven climate disaster risk reduction policies? Is it because there is no knowledge available on risks? No technical expertise? No additional financial resources to tackle climate risks? Or simply no willingness to govern climate change and avoid its detrimental consequences? We get to these reasons and detail them in our country Chapters 3-5. 10.1 Triggers of Institutional Change in the Governance of CCA: between North or Grindle, who was right? Our research first and foremost reveals that in many constitutions across Africa is 331

explicitly stated that the Republic shall be responsible to protect the physical integrity and ensure the security of its citizens. The Constitution of Senegal for instance, in its Act No. 2001-03 of 22 January 2001, part II on civil liberties and the human person, economic and social rights, and collective rights, Article 7, states that:
The human person is sacred. H/she is inviolable. The State has the obligation to respect and protect him/her. Every individual has the right to life, to liberty, to security (emphasis added), to free development of his/her personality, to physical integrity including to protection against all physical mutilation. The Senegalese people thus recognize the existence of human rights, inalienable and inviolable, as the basis of any human community, peace and justice in the world. All human beings are equal before the law. Men and women are equal in rights. In Senegal there is neither subject nor privileged place of birth, person or family. (Constitution of Senegal, 2001).

That of Uganda mandates the State to ensure that an efficient machinery is in place to address all hazards and risks that may lead to calamitous disasters for community (National Policy for Preparedness and Management of Disasters, 2010). An interpretation of these texts, in light of a national climate crisis, brings to light the States primal responsibility to ensure security, well-being and integrity of all of its citizens in peace time as well as in time of crisis. Both these texts make it a constitutional responsibility of the State to provide relief to victims in times of national disasters, or otherwise address disasters for all citizens, regardless of their sex, origin or birthplace. In light of Climate Change, with projected impacts high (refer back to our chapter 3), the States responsibility to protect and ensure the security of its citizens during times of climate crisis becomes even more primordial. Indeed, in times of climate crisis, the State, strong, accountable and effective, is needed more than at any another time. In this dissertation, we posited disaster prevention as a public good, at the same level as education or health. We looked into the systemic reasons that prevented 332

governments from delivering on this public good. The insights we derived from our cross-country investigation are valuable for all the scholars attempting to understand why CCA-DRR policy, and public goods delivery in general, is not forthcoming from many governments in Africa. This thesis furthermore offered a contextualized understanding of what it will take to move countries to address Climate Change seriously, before its worst projected impacts begin to materialize. The gloomy conclusion we reached is that countries wont, until they begin to experience and see the projected impacts of a changing climate for themselves. They will not learn from the age-old wisdom that prevention is better than cure, until the feared impacts of Climate Change begin to materialize. Then, CCA will become a national urgency, and move up the national agenda. When left only with a choice between to Adapt or to Perish, countries will choose to Adapt. The main reasons that explain why countries do not take any meaningful action on Climate Change, from the Senegal case mainly, which lies at the bottom of our spectrum of DRR-CCA along with 33 other countries in Africa (see chapter 5), are as follows: 1) Difference in time horizons between climate change and electoral cycles; 2) Political rationale not aligned with climate adaptation decision-making imperatives for effective DRR-CCA; 3) Corruption and lack of transparency: we indeed saw from the Senegal case how funds dedicated for disaster relief were even siphoned away. Our cases from Uganda and Mozambique suggest as well that when combined, a number of factors can also spur countries to place the prevention and preparedness for Climate change risks high up on the agenda, confronting them seriously to avoid their worst impacts. These factors include: 1) Increased frequency and/or magnitude of national HMDs, creating a country driven-ness for Institutional change to meet the new CCA challenge; 333

2) Donor financial and technical support: Available funding and trust from donors; 3) The supportive presence of leaders animated with a strong desire to safeguard country and its citizens (agent based explanation; in itself at times an accident of history). Another positive story also emanates from Mozambique: the power of the agent. Despite confirmation of path dependence, the only salvation coming from accidents that shift a countrys course (through increased HMDs mainly, but also a an internal war as was the case in Uganda that bequeathed a massive issue of internal refugees to be addressed), when an agent of the system, dedicated to the issue, is by chance at the helm of the republic, then meaningful institutional change can take place. But the systems have no mechanisms inherent in them to enable citizens at the base to demand leaders to cater to their fundamental CCA-DRR needs, as evidenced from our accountability analysis, which revealed a discrepancy between community needs and government interventions in all countries surveyed, except in Mozambique. However one could also extrapolate that, to a certain, such agents dedicated and patriotic are also the fruit of conjectures, or accidents of history in the words of North, as Mozambiques Joaquim Chisano was the fruit of a generation of warriors who had to fight for the countrys independence, peace and stable development, and as such assumedly cared for the concerns of their people. Thus, the big man theory brought in by the Mozambique case and hypothesized by Grindle only goes to prove that indeed good leadership does help, for all good governance, including that of climate related risks. But it is not the single largest driver of institutional revolution, adaptation and change to meet the new challenge of Climate Change. Rising HMDs are.

334

As such, Norths theory of institutional change is supported. Change appears to be indeed incremental, it takes a revolution or a natural disaster (or rather the rise in the propensity of their occurrence) to bring about a severe break in a countrys institutional continuity. This confirms Norths insight that natural disasters, along with war, may be the fastest routes to institutional change (North, 1990). They, better than any other route, achieve fast reconfigurations of a nations institutional mechanisms, or rules of the game, to achieve collective goals. In this instance Norths theory was tested on the CCA issue. Our comparative African study of the drivers of institutional change/adaptation in the face of rising HMDs confirms that there are generally two moments of change: 1) A moment of Institutional revolution, when the a tipping point is attained in the propensity of rising disasters hitting the country, discontinuing a countrys institutional pathway; 2) Followed by incremental changes and adjustments in the new proposed institution, or rules of the game, towards the consolidation of the new path in which the country is set by the incipient factor. This is the stage of small changes at the margins where organizations/players interact within the new institutional framework to define and chisel an institutional equilibrium that serves their interests.364 This seems to confirm that indeed, institutional change is path dependent. It took a national disaster, or rather a rise in their propensity, to set the country off track, onto a new track. The example hailed as the top of the class in Africa through this research, Mozambique, in reality had no choice, but to adapt. As such, it appears that it is those who are recurrently battered by disasters who tend to develop the frameworks to react, going from doing things on an ad-hoc basis every time there is a disaster to anticipating disasters and responding through a coordinated system. 364 North, Institutions, Institutional Change and Economic Performance, 6 335

 Otherwise, change in institutional frameworks to govern any given issue is incremental, and very much path dependent, unless set off track by an accident of history, or conjecture, such as a natural disaster. Our Senegal case, that remains mired in the paradigm of plans of paper that are never implemented when a disaster strikes, proves this adequately. It confirms that path dependence is a reality, and unless accidents of history occur to sidetrack the country and set it on a new path of more effective DRR-CCA, these will take very long to materialize. We see herein then the power of history, through which the past is linked to the present and to the future, through the continuity of institutions. Conjectures can create negative institutional dependence (e.g.: colonialism and its institutional legacies of economic dependence), but they can just as well break societies out of their lethargy, setting them on a new track of improved effectiveness and more socially efficient means to deliver on societal goals. These are the lessons from this thesis. Specifically for CCA, the lesson is that the more CC risks will increase, the more countries will be forced to adapt. As such, there is no need to badger national leaders with the gospel of CCA-DRR and the imperative of adaptation. Demand will have to come from within, endogenously, and emanate from their constituencies in democratic settings and patrimonies in non-democratic ones. In all likelihood, such local demand for improved CC governance will follow from the increase in the frequency/magnitude of climate disasters that are likely to be hitting nations in the years ahead. Till then, local communities will have to continue to bear the brunt of CC impacts without the assistance of their governments. NGOs have to keep accompanying them in their efforts to adapt to a fast changing climate. 336

10.2 Implications Implications for proponents of CCA-DRR in Africa In light of our findings, the value added of CCA-DRR proponents in Africa, and perhaps elsewhere too, is to work on creating demand from within for improved climate governance. This indeed, appears as the most direct route to get governance to respond to local priorities and uptake policy recommendations on CCA-DRR. Although global blueprints proved instrumental in establishing national policy frameworks auspicious for effective DRR-CCA policies (ISDR serving as an notable illustration of these successes), the type of substantive institutional change needed to meaningful confront rising climate-related risks and infuse life into existing national DM organizations cannot be effected from outside. It will be local citizens who will have to put CCA-DRR on the national agenda. External proponents can only support from the rims. How does one support the creation of local demand? One valuable way is for advocates to invest in processes that strengthen local agency and country driven- ness, in the aim of getting to the outcome of improved governance of climate risks and opportunities at the national level. In so doing, their focus will have to be on processes, new ways of doing business and operating as a system (business unusual), rather than on mere outputs. Empowering local DRR-CCA champions and stimulating multi-stakeholder dialogue processes365 around a national covenant new rules of the game so to say- to manage rising climate risks offer examples of how creating local demand for change in CCA-DRR governance can be achieved in Africa.

365

Refer to dialogue process example held in Northern Senegal, 2009: http://www.youtube.com/watch?v=87Hl66K7nnE

337

Another large area of need is in technical expertise, needed to comprehend and confront an issue as complex as CC. Scientific capacity development will be specifically needed for risk identification and understanding of CC hazards and impacts, as well as for devising locally-adequate institutional architecture to coordinate disaster prevention and preparedness, training and capacity building across all levels of government. But governments will have to come to the realization by themselves of the stakes of proper national institutions for DRR-CCA. Then they will seek such support, as Mozambique and Uganda did when they were ready to embark on meaningful institutional change. Country driven-ness is a pre- requisite. Implications for donors Donor support, from each of the three cases, appeared as an important factor. Indeed, faced with the new challenge of CCA-DRR, most countries in Africa at the outset will not know how to address the issue. When countries become willing to transition up the spectrum of effective climate disaster risk management however, donor and development partners support, both technical and financial, will be critical. At that stage, donor assistance to desirous countries following from an endogenous demand for more effective CC governance will be instrumental to accompany the latter in transforming their national desire for improved CC management into an efficient, and contextually-relevant national policy on CCA- DRR. In so doing, the emphasis on contextual relevance is essential. Indeed, donor support will have to be tailored to the specific national context of each beneficiary country. One-fits-all cookie-cutter policy prescriptions for DRR-CCA promotion, such as the ones currently promoted by global DRR efforts (see ISDR, 2007366), will
366 ISDR 2007, Hyogo Framework for Action 2005-2015: Building the resilience of nations and communities to

disasters; and UNISDR, Guidelines: National Platforms for Disaster Risk Reduction.

338

simply not do. They will not be useful for the countries donors are seeking to support to better address climate related-risks. As this thesis has demonstrated, African countries lie at different stages on the spectrum of effective disaster risk management (see fig. 5.6); they have achieved differential progress in implementing the Hyogo goals. As such, it is important that donors, if they seek to be useful, tailor their policy support based on whether a country is an unprepared vs. prepared firefighter vs. disaster averter; perhaps even creating agile, nimble mechanisms that allow for rapid deployment of support in the aftermath of particularly noteworthy disasters, so as to nurture the conditions for institutional change appropriate for the national context. Implication for national actors: Pushing for Governance change from within The most important implication of our findings is, perhaps, for government planners, public officials, development organizations and agents who see the new threats posed by CC, but find themselves hands-tied in a non-responsive system. Governance is the single largest determinant of whether a country will deliver on local CCA-DRR priorities. How to build political will in Africa to deliver on the public good of effective climate-disaster prevention, when counter-powers and checks on the power of the Executive are missing? It appears from our research that an answer can be to build local pressure mechanisms, national demand and country driven- ness for change in governance. When the need is there, patent and urgent, jeopardizing electoral gains and the patrimony, leaders cannot shirk away anymore from their constitutional responsibilities towards citizens. It will thus be up to public officials within the system, also, who are concerned with CC governance, to make it appealing to their national decision-makers to invest in prevention rather then cure. The latter have a number of cards in their hands: they can stress the rising importance of CC in conflicts and political change (see Strauss 339

Center work on Climate and Security)367. They can highlight the political costs of inaction in the face of more frequent CC hazards and accumulating impacts (e.g.: the political costs of floods in Senegal). They can underscore the gains from climate- proofing the patrimony and embarking on climate-compatible development, harnessing the multiple climate financing opportunities afforded in the new global governance of CC. Even more importantly, they will have to prove the value of climate-disaster prevention, commissioning research that builds hard evidence on the value of prevention, in the national context, and demonstrates the personal and collective dangers associated with non-adaptation to projected CC impacts. These officials within the system will be the finest allies of forces from outside seeking to promote DRR-CCA, and the most effective spokespeople of local circumscriptions at the frontlines of CC impacts, in the instances when the latter have no voice. It will be up to organizational players to build and keep pushing for momentum from within. They have the largest potential to influence policy from within the policy system. To these internal change agents the organizations or players in Norths description, it will be important to stress however that institutional change is incremental, it comes in piecemeal wins punctuated by occasional set-backs. They will have to arm themselves with persistence in their attempts to change the governance of CCA. 10.3 Proposed Way Forward to Enable Countries to Prevent rather than Cure In light of our final prediction of countries continued lethargy until projected CC impacts begin to materialize, what measures could be devised by policy proponents to pro-actively encourage countries to not Wait and see?
367 Climate Change and African Political Stability (CCAPS),

http://ccaps.strausscenter.org/research/constitutional-design-and-conflict-management

340

 Valuate costs of Prevention vs. costs of Curing Projected CC Impacts A way forward would be to gain a better understanding of how CC will concretely impact African countries, with more robust projections for impacts. In a context of multiple development priorities and urgencies calling for attention, one can make a stronger case for why a country needs to devote scarce resources to pro-active DRR- CCA when there are robust findings on the severity of the exact impacts projected to hit the country. In this vein, sturdier and more systematic scientific assessments of CC hazards, impacts, likely new trends and changed windows of opportunity are an urgent necessity at the level of every country, to make the case for Prevention rather than Cure. However, as defined earlier in this treatise CC impacts on society are tributary to two factors: the hazard itself, but also the adaptive capacity of the community, society or human system that is hit by the climate hazard. Identify and Strengthen local Adaptive Capacity Characterizing African countries adaptive capacity in the face of HMDs remains a tricky endeavor. Vulnerability to HMDs can only be assessed using proxies of overall inherent or social vulnerability assumed to compose vulnerability to HMDs, such as level of poverty, access to health, education and food security. However evidence on vulnerability at the sub-national level in is rare. Even in a country such as Mozambique, which represents the top of the crop in our Africa wide classification of countries, information related to the exposure of Mozambique is available at a geographically disaggregated level, but the same does not hold for sensitivity. Most data available on sensitivity is focused at the national level, or zooms in on a particular local case study, providing no basis for comparison. A thorough vulnerability mapping of the country of Mozambique requires access to detailed and reliable datasets on both exposure and sensitivity. 341

 Learn from your neighbors mistakes: the Domino Effect Alternatively, the demonstration effect throughout history has always had a powerful pedagogic effect. Thus demonstrating to countries that new risks that have emerged elsewhere in places that had never experienced them before (e.g.: landslides in Uganda beginning in 2010), could be a powerful medium to convince other countries that new emerging risks could also overtake them as a result of CC, and that measures can be taken today before these disasters compromise development gains, in the hope of a domino effect across the continent in the creation and implementation of effective CCA-DRR institutional frameworks. How to go about the process of demonstration to countries? Cross-country experiential learning will have to be facilitated through the use of innovative tools to communicate to policy-makers across different settings (participatory video serving as a prime example; see InsightShare368), while harnessing the new tools of modern technology, such as social media, to inform and mobilize the increasingly urban citizenry across Africa around the critical urgency of CC governance. Incentivize Planning from the Future Finally, planning from the future will have to be incentivized. Governing is the art of planning. In light of the likely 21st century world modulated by CC, with its host of new threats and reshaped opportunities, planning becomes urgent, now more than ever before. Planning from the perspective of what the country, and broader global context, will be in years 2025, 2050 and 2100, and setting the country on track to meet those future challenges, now. This is what planning from the future entails (Humanitarian Futures Program, 2012)369 . Such pro-active anticipatory development planning is not a luxury, no; but rather a vital element of getting ready for a future world where only the fittest, and best prepared, will survive.

368 InsightShare. http://www.insightshare.org/pv/pv-nutshell 369 Humanitarian Futures Program. http://www.humanitarianfutures.org/

342

Encouraging policy planning around CCA-DRR as an overall crosscutting issue is thus an imperative, and so is mainstreaming the issue into longer-term forward- looking development plans. The Poverty Reduction Strategy Papers (PSRPs), now recast across the continent as National Development plans and Visions, offer a strategic outlet for such planning. Placing funded CCA technical advisors in planning ministries, establishing competitive awards for the most climate-ready country, and funding Futures Planning National Centers also constitute examples of how planning from the future can be incentivized. However, this is a challenge larger than Climate Change. As demonstrated in this thesis, all incentives converge to make elected officials address the priorities of today, leaving them often with very little space to develop a vision for their countries, if they were ever inclined to do so. In this context, it becomes difficult to plan for, let alone from, the future. In Senegal for instance, Plan is nothing more than a small department ensconced within the almighty Ministry of Economy and Finance (MEF). Planning is hence reduced to mere economic projections, rather than a cross-cutting visioning exercise to set ones country on track to meeting the countrys future needs and challenges be they demographic, social, economic or institutional. Perhaps, Climate Change opens an opportunity for countries to break this pattern. Indeed, in the end, CC also opens a tremendous opportunity for Africa. An opportunity for States across the continent to reform their institutions and modus operandi, enabling better cross-sector coordination for improved service delivery in CCA as well as other development priorities deemed of national importance. Securing sturdier pressure channels from the base, paving the way for more downwardly accountable governance and integration of local priorities into national plans and strategies. Our analysis has shown the large discrepancy between local priorities and national funded plans on CCA-DRR, corroborating what Boone labeled 343

as the chasm between city and countryside in Africa. Overall, an opportunity to promote the definition of a new national covenant or social contract a new Institution in its essential definition since the winds of Independences blew over Africa370, centered this time on serving the priority needs of all citizens in the polity. When pandemonium breaks loose with CC, perhaps then, countries will finally awaken, and implement a new paradigm for addressing disasters, future threats, challenges and possibilities, organizing themselves in more effective ways to deliver on societal needs in 2025, 2050 and 2100. Making it possible for the Future to be bright and endurable for their citizens. Let us only hope that some of the above- proposed recommendations will be applied, so as to thwart us from reaching the point of hitting rock bottom. Planning from the future is an imperative in the CC world of the 21st century. We can not control the future, but we can change it by planning for it. 10.4 Limitations We were limited in our analysis by three major constraints. Firstly, more experts need to be interviewed in each country to ensure the internal validity of attained results for each country case, on which rests our cross-country analysis of the triggers of institutional change on CCA-DRR. Furthermore, the external validity of our results remains unascertained. Indeed, our suggested findings from this research have to now be tested across a larger swathe of countries within the same policy group, as well as over countries across policy groups, in order to validate the findings for each policy group. This will enable comparisons between a larger number of country cases from the three different policy groups up the effective DRM spectrum we delineated in this thesis: the Unprepared Firefighters, the Prepared

370 Were borrowing here the words of writer Ahmadou Krumah, employed in his novel Les Soleils des

Indpendances (Ahmadou Kourouma, Presses de Montral, 1968). There is some likelihood also to British Prime, McMillan,s historic speech in Cape Town when he visited South Africa in the 1960s, during the heights of Apartheid, in reference to the rising force of the counter black renaissance movement in the country.

344

Firefighters and Disaster Averters. At the same time, it needs to be recognized that the classification of Africa by DM policy type proposed in this treatise, which underpinned our country case selection, is only valid up until September 2010, when the classification task for this research was completed, using data that sometimes dated back to the early 2000s. Many new developments may have occurred (the adoption of a new legislation on CCA-DRR, the creation of a DRM organization) but not be captured in our present classification. Thus, any future studies aiming to extend the conclusions from this thesis, if they cannot update the classification data by themselves, will have to be mindful of this limitation, inherent in every classification endeavor that is work in constant progress. Secondly, our understanding and characterization of projected Climate Change impacts in Senegal and Uganda was crippled by the absence of reliable and rigorous scientific assessments of national vulnerability to Climate Change, and projected CC impacts, compared to those readily available for Mozambique. This thwarted our ability to provide a clear characterization of Senegal and Ugandas climate profile. Finally, the noted skew towards increased HMDs in recent years, which we rested our analysis on throughout this research, needs to be now be further investigated and ascertained across Africa, and other low income settings of the world. Indeed, the upward trend in HMDs we noted could be attributed to better disaster reporting in the age of improved Information technology systems we live in. However all sources we utilized converge to point to a break point in the evolution of HMDs in Africa (global datasets, national assessment such as DesInventar, and community accounts of HMDs). As such, the noted trend in HMDs seems to be more than a spurious conclusion resulting from better bookkeeping. Disasters are on the rise. This conclusion now has to be extended, tested and more widely recognized as a scientific fact we have to address today. 345

At last, a note on the limits of our predictive model is called for. We concluded this treatise with the supremacy of path-dependency above all other causal factors explaining institutional change/continuity. We underscored the strong linkages in which the past is linked to the present and the future through the continuity of societys institutions371, lest a powerful accident of history intervenes to discontinue the countrys institutional pathway, setting it off track, on a new institutional track. No-one is to predict, however, other accidents of history the future has in store. We stand by our conclusions in so far as the only foreseeable accident of history is the one of rising HMDs, which has been observed in a number of countries already in Africa, which other countries are likely to face in the near future. The future works in mysterious ways however, which a predictive model can always fall short of fully grasping. This thesis nonetheless offers a clear picture of what it will take for governments in Africa to truly rise to the mettle if HMDs continue to rise, and assume the Governance of Climate Change risks (and opportunities) in their countries, surviving in a future 21st century that is projected to be increasingly conditioned by changes in our collective climate system. Will States Adapt or Perish? Studies from the future will have to tell. But we already have enough elements as of now to craft the future, and ensure that national economies in Africa are set on a track of development that is both mindful and fortified against the vicissitudes of a changing climate. Arame Tall Dakar, April 4 2012 371 North, Institutions, Institutional Change and Economic Performance, vii 346

Bibliography Acemoglu, Daron, Simon Johnson, and James Robinson, The Colonial Origins of Comparative Development: An Empirical Investigation, American Economic Review 91 (2001): 1369-1401 Acemoglu, Daron and James Robinson, Political Losers As a Barrier to Economic Development AEA Papers and Proceedings 90 (2000): 126-130 Agrawal, A., and Nicolas Perrin, Climate Adaptation, Local Institutions, and Rural Livelihoods, IFPRI working paper # W08I-6 (2008), Accessed April 04, 2012. http://sitemaker.umich.edu/ifri/files/w08i6_agrawal_perrin.pdf Agrawal, A., The Role of Local Institutions in adapting to Climate Change, Paper prepared for the Social Development Department. Washington D.C.: World Bank, 2008. Ake, C., Democracy and development in Africa. Washington, D.C: Brookings Institute, 1996. Almond, Gabriel A., and Sydney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton: Princeton University Press, 1963. Amin, Samir, Imperialism and Unequal Development. New York: Monthly Review, 1977. Bates, Robert. Markets and States in Tropical Africa: the political basis of agricultural policies. Berkeley: University of California Press, 1981. Benaouda H., A. El Ouali and A. Saloui, Rapport final du projet CRDI/INRA : Mcanismes dadaptation aux changements climatiques des communauts rurales dans deux cosystmes contrasts en plaine et montagne du Maroc . Settat: Institut National de la Recherche Agronomique, 2008. Betts-Symonds, G. (2003) Best practices in disaster preparedness: Developing a Critical Pathway for More Effective VCA in the Red Cross / Red Crescent. Geneva: IFRC. Boone, C., Political Topographies of the African State: Territorial Authority and Institutional Choice. Cambridge: Cambridge University Press, 2003. Boudreau, Laura, Flood Management in Senegal: Past, Present, and Future, 2009 Braman, L., Early Warning, Early Action, an Evaluation, Report to the IFRC/Red Cross Red Crescent Climate Centre (2009), Accessed April 4, 2012. 347

http://www.climatecentre.org/downloads/File/ewea_an_evaluation_of_ifrc_west_a nd_central_africa.pdf Bryceson, D. F., The Scramble in Africa: Reorienting Rural Livelihoods. World Development (2002) 725739. Cane, M.A., S.E. Zebiak and S.C. Dolan, 1986: Experimental forecasts of El Nio. Nature, 321: 827832. Canziani, J.P. Palutikof, P.J. van der Linden, and C.E. Hanson (eds). Cambridge: Cambridge University Press, 2007. Cardoso, Fernando H., and Faletto Enzo. Dependency and Development in Latin America, trans. by Marjory M. Urguidi. Berkeley : University of California Press, 1979. Center for Research on the Epidemiology of Disasters (CRED), Annual Disaster Statistical Review, The numbers and trends 2007. Brussels: CRED, 2008. Chhibber, Ajay, and Rachid Laajaj. Disasters, Climate Change and Economic Development in Sub-Saharan Africa: Lessons and Future Directions (World Bank Independent Evaluation Group, Evaluation Brief 3). Washington DC: World Bank, 2007. Accessed April 4, 2012. http://siteresources.worldbank.org/INTOED/Resources/EvalBrief3.pdf. Civil Protection Department, Activits de la DPC 2001-2010 Coly, Adrien, Dr. Theme 2: Inondations Et Processus d'Urbanisation., 2001 Constitution of Senegal, 2001 Dai, A., P. J. Lamb, K. E. Trenberth, M. Hulme, P. D. Jones and P. Xie, 2004: The Recent Sahel drought is Real. Int. J. Climatol. 24:13231331. Dartmouth Flood Observatory, 2003: 2002 Global Register of Major Flood events. Dartmouth Flood Observatory, Hannover, NH, USA. Accessible online at: http://www.dartmouth.edu/artsci/geog/floods/ Denton, Fatima (ed.), Institutional Vulnerability to CC: Learning from lessons from across the Africa, CRDI: forthcoming. Dia, Aliou M., "Crues Et Inondations Dans Le Bassin Du Fleuve Sngal: Apport De La Tldtection Et Des SIG La Cartographie Des Zones Affectes Dans l'Estuaire 348

(Application Sur La Ville De Saint-Louis)", Diss., Universit de Marne la Valle: Institut Francilien des Gosciences, 2004 Diagne, K. Governance and Natural Disasters: addressing floods in Saint Louis, Senegal. Environment and Urbanization 19 (2) 2007: 552-562. Diouf, Sidate. Catastrophes Naturelles Et Leurs Effets: Le Contexte Sngalais. Ministre de l'Intrieur, Direction de la Protection Civile, 2005 Ekeh, P., Colonialism and the two publics in Africa, Comparative Studies in Society and History 17 (1975): 91-112 Ellis, Frank, Stephen Devereux and Philip White. Social Protection in Africa. Northampton: Edward Elgar Publishing, 2009. Encyclopdia Britannica Online, 2011 Janicot, S., S. Trzaska, and I. Poccard, Summer Sahel-ENSO teleconnection and decadal time scale SST variations, Climate Dynamics 18 3-4 (2001): 303-32 Korecha, D. and A. G. Barnston, Predictability of June-September rainfall in Ethiopia, Mon. Wea. Rev. 135 (2007): 628-650. Expert interview with URCS field coordinator, February 2012 Expert interview with Uganda Red Cross Society chief officer, October 2011 Expert interview with Abdoul Azziz Seck, 6 January 2011 Expert interview with Commandant, Civil Protection Department, Jan-March 2011 Expert interview with former Department Civil Protection-UNDP project head, 2011 Food and Agricultural Organization of the United Nations, FAO Statistical Yearbook 2009 (Rome: FAO, 2009). Accessed April 04, 2012. http://www.fao.org/economic/the-statistics-division-ess/publications- studies/statistical-yearbook/fao-statistical-yearbook-2009/en/ Geddes, Barbara. Politician's Dilemma: Building State Capacity in Latin America. Berkeley: University of California Press, 1994. Giannini, A., M. Biasutti and M. M. Verstraete. A Climate Model-Based Review of 349

Drought In The Sahel: Desertication, The Re-Greening And Climate Change, Global and Planetary Change 64 (2008): 119-12. Giannini, A., R. Saravanan, and P. Chang, Oceanic Forcing of Sahel Rainfall on Interannual to Interdecadal Time Scales, Science 302 7 (2003) Giannini et al., 2008 Government of Uganda (GoU), Uganda National Policy for Preparedness and Management of Disasters, 2010 GoU, Peace, Recovery, and Development Plan for Northern Uganda, 2007; GoU and UNDP, Millennium Development Goals: Ugandas Progress Report, 2007 GoU, National Policy for Disaster Preparedness and Management, September 2010. GoU, National Adaptation Plan of Action (NAPA), 2007. GoU, Vision 2025 GoU. National earthquake disaster committee report 1994, GoU. Preliminary report on earthquake disaster in Kabarole, Bundibugyo and Kasese districts, Ministry of Labour and Social Affairs, Republic of Uganda Grindle, Merillee S. Audacious reforms: Institutional invention and Democracy in Latin America. Baltimore: Johns Hopkins University Press, 2000. Gunder Frank, Andre. Latin America: Underdevelopment and Revolution. New York : Monthly Review Press, 1970. Hansen, J., M. Sato, P. Kharecha, D. Beerling, R. Berner, V. Masson-Delmotte, M. Pagani, M. Raymo, D. L. Royer, and J. C. Zachos. Target atmospheric CO2: Where should humanity aim?, (2008): Hastings, David A. Africas climate observed: perspectives on monitoring and management of floods, drought and desertification, In Climate Change and Africa. Ed. Pak Sum Low. Cambridge University Press (2005) Heike Shroder, xxx governance definition Heike, Shroder, Social Dimensions of Global Environmental Change, IHDP, University of East Anglia LOOK UP 350

Herbst, J. I. States and power in Africa : comparative lessons in authority and control. Princeton: Princeton University Press, 2000. Hulme M., R. Doherty, T. Ngara and M. New. Global Warming and African climate change: a reassessment, In Climate Change and Africa. Ed. Pak Sum Low. (Cambridge: Cambridge University Press, 2005). Hulme M., 2001: Climatic perspectives on Sahelian desiccation: 19731998, Global Environmental Change 11 (1): 19-29. Huntington, Samuel P. Political Order in Changing Societies, New Haven. (Yale University Press, 1968) IFRC- International Federation of Red Cross and Red Crescent Societies (2006). Guidelines for Vulnerability and Capacity Assessment (VCA). Available at www.ifrc.org/what/disasters/dp/planning/vcaguidelines.asp Instituto Nacional de Gestao de Calamidades (INGC), Study on the Impact of Climate Change on Disaster Risk in Mozambique: Main Report, Asante, K., Brundrit, G., Epstein, P., Fernandes, A., Marques, M.R., Mavume, A , Metzger, M., Patt, A., Queface, A., Sanchez del Valle, R., Tadross, M., Brito, R. (eds.), 2009. Inter-Governmental Panel on Climate Change (IPCC). Climate Change 2007: Impacts, Adaptation and Vulnerability, Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M.L. Parry, O.F. Inter-Governmental Panel on Climate Change (IPCC). Managing the Risks of Extreme Events and Disasters to Advance Climate Change adaptation, Special Report of the Intergovernmental Panel on Climate Change. Edited by Christopher B. Field, Vicente Barros, Thomas F. Stocker and Qin Dahe. (Cambridge: Cambridge University Press, 2012.) Intergovernmental Panel on Climate Change, IPCC Special Report on Emissions Scenarios, Summary for PolicyMakers (, 2000) Inter-Governmental Panel on Climate Change (IPCC), 2001: Climate Change 2001: The Scientific Basis, Houghton, J.T., et al. (eds.), Cambridge Univ. Press, Cambridge, Cambridge, United Kingdom and New York, NY, USA. International Federation of the Red Cross/Red Crescent Societies (IFRC), World Disasters Report 2009: Focus on Early Warning, Early Action (Geneva: IFRC, 2009). Institut des Sciences de lEnvironnement, Student field trip report on Dakar floods, September 2009 351

Kohli, Atul, Peter Evans, Peter J. Katzenstein, Adam Przeworski, Susanne HoeberRudolph, James C. Scott and Theda Skocpol, The Role of Theory in Comparative Politics: A Symposium, World Politics 48-1 (1995), 1-49. Le Quotidien 15 May, 2009 Le Quotidien, June 9 2009 Le Quotidien, 30 June 2009 Le Quotidien, 26 August 2009 Le Populaire, Jan 6 2011 Le Soleil, July 2 2008- PRESAO forecast warns of heightened risk for an abo normat season Le Soleil, July 26 2010: 2005-2010 Et si lhistoire du relogement metait contee Le Soleil, 15 January 2010 Le Soleil, May 20 2010 Le Soleil, May 28 Le Soleil, July 26 2010 Linnerooth-Bayer, J., Suarez, P., Victor, M. and Mechler, R. (2009). Drought insurance for subsistence farmers in Malawi, Natural Hazards Observer 33 (5): 1-8 Lipset, Seymour M., "Some Social Requisites of Democracy", American Political Science Review 53 no. 1 (1959): 69-105 Mamdani, M. Citizen and subject: contemporary Africa and the legacy of late colonialism. Princeton: Princeton University Press, 1996. Mbale Region draft Climate Change framework Plan, 2009 McGranahan, G., D. Balk and B. Anderson, The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones, Environ Urban 19 2007: 1737. Ministry for the Coordination of Environmental Affairs (MICOA), National Adaptation Programme of Action Mozambique, 2007 Ministry of Planning and Finance, Anuario Estatistico 1994, National Direcorate of Statistics, 1994 352

 Mortimore, M. and W. M. Adams, Farmer adaptation, change, and crisis in the Sahel Global Environmental Change 11 2001: 49-57. Moyini, 2005 In NAPA, 2007 MRDPR (Ministry of Relief Disaster Preparedness and Refugees), National Policy for Disaster Preparedness and Management, 2010 National Directorate of Statistics, National Census, 2007 New Vision,. Land, FGM, Electoral Reform Bills And CHOGM Scandals End 2009 Parliament. January 1, 2010. http://allafrica.com/stories/201001041477.html NAPA Senegal, 2006 Nkomo J. C., A. O. Nyong and K. Kulindwa, THE IMPACTS OF CLIMATE CHANGE IN AFRICA.In The Stern Review on the Economics of Climate Change. July, 2006. Niang-Diop, Isabelle, Dr. "Les Apprentis Sorciers l'uvre Au Canal De Drivation", Walf Fadjri, October 16, 2003. North, Douglass. Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press, 1990 North, Douglass C. Structure and Change in Economic History (New York: Norton, 1981). 240 pp. North, Douglass C., and Barry R. Weingast, "Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England", Journal of Economic History 49 4 (1989): 803-832 National Academies of Science: Joint science academies statement: Global response to climate change. http://nationalacademies.org/onpi/06072005.pdf NOAA Earth System Research Laboratory, January 2012 http://www.esrl.noaa.gov/gmd/ccgg/trends/ ONeill, B.C, and M. Oppenheimer, Dangerous Climate Impacts and the Kyoto Protocol, Science 296 2002: 1972 Official website of the Senegalese government, http://www.gouv.sn/spip.php?article694 353

OFDA, Disaster History: Significant Data on Major Disasters Worldwide, 1990-1995, 1996 Office of Foreign Disaster Assistance. USAID, Washington, 265pp. Olson, Mancur, Dictatorship, Democracy, and Development, American Political Science Review 87 (1993): 567-76 Pearson Prentice Hall Inc., Climatology: An Atmospheric Science, Hidore, Oliver, Snow and Snow (eds.), 2010 Patt, A.G., Suarez, P. and Hess, U. How much do smallholder farmers understand and want insurance? Evidence from Africa. Global Environmental Change 20 1 2010: 153-161. Patt, A.G., Peterson, N, Carter. M, Velez, M., Hess, U. and Suarez, P. Making index insurance attractive to farmers. Mitigation and Adaptation Strategies for Global Change 14 2009: 737-757. Pelling, Mark and Ben Wisner. Disaster Risk Reduction: cases from Urban Africa. London: Earthscan, 2009. Pierre Englebert. State Legitimacy and Development in Africa. Boulder: Lynne Rienner Publishers, 2000. Rawls, John, Theory of Justice. Cambridge: Belknap Press of Harvard University Press, 1971. Ribot, Jesse, Vulnerability does not Fall from the Sky: Towards Multi-Scale, Pro- Poor Climate Policy, In Social Dimensions of Climate Change: Equity and Vulnerability in a Warming World, Mearns, Robin and Andrew Norton (eds.), 2010. Rustow, Dankwart A., and Kenneth P. Erickson (eds.), Comparative Political Dynamics. New York: Harper Collins, 1991. Scott, James, Weapons of the Weak: Everyday forms of peasant resistance. New Haven: Yale University Press, 1985. Sen, Amartya. Poverty and Famines: An Essay on Entitlement and Deprivation. 1981 Sens Amartya developments in Poverty as deprivation Sen, Amartya P.overty and Famines: An Essay on Entitlement and Deprivation Development as Freedom. 1981 Senegal Hyogo Framework for Action mid-term review report, 2009 354

 Stanley L. Engerman, and Kenneth L. Sokoloff, "Factor Endowments, Inequality, and Paths of Development among New World Economies," Economia 3 1 (2002): 41- 101. Stern, P. C. and W. E. Easterly. Making Climate Forecasts Matter. National Research Council, 175 pp. 1999 Suarez, P., and Linnerooth-Bayer, J., Microinsurance for local adaptation. Wiley Interdisciplinary Reviews: Climate Change 1 2 (2010): 271-278. Sud Quotidien, November 7 2008 Sud Quotidien, 3-4 April 2009 Sud Quotidien, July 7 2010- PM Ndn and premade houses Sud Quotidien, August 17 2010- AGRHYMET alert on river sngal Sud Quotidien, August 19 2010 Sud Quotidien, August 20 2010 Tadross, M., Suarez, P., Lotsch, A., Hachigonta, S., Mdoka, M., Unganai, L., Lucio, F., Kamdonyo, D. and Muchinda, M. (2009). Growing-season rainfall and scenarios of future change in southeast Africa: implications for cultivating maize. Climate Research 40: 147-161 Tall, Arame, Learning from the Pre-colonial Past, Berlin: Lambert Academic Publishing, 2011. Tall, Arame. Climate forecasting to serve communities in West Africa. Procedia Environmental Sciences 1 (2010): 421-431. Accessed online December 30, 2011. http://www.sciencedirect.com/science/journal/18780296] Ting, Mingfang, Yochanan Kushnir, Richard Seager, And Cuihua Li, Forced and Internal Twentieth-Century SST Trends in the North Atlantic, Journal of Climate (2009) 1469-81 Oxfam GB, Turning Up the Heat: Climate Change and Poverty in Uganda, 2009 Uganda." Encyclopdia Britannica. Encyclopdia Britannica Online. Encyclopdia Britannica Inc., 2011. Web. 30 Dec. 2011. <http://www.britannica.com/EBchecked/topic/612597/Uganda>. 355

UN Department of Humanitarian Affairs Situational report Uganda Earthquake Feb 1994 UNDP, Climate-resilient low-carbon growth strategies, 2011 UNECA, Africa Governance report, Addis Ababa: UNECA, 2005 UNFCCC, NAMAs and NAPs, 2011. www.unfccc.int UNISDR, Hyogo Framework for Action 2005-2015: Building the resilience of nations and communities to disasters (Geneva : United Nations, 2007). Accessed March 29, 2011. http://www.preventionweb.net/files/1037_hyogoframeworkforactionenglish.pdf United Nations Development Program. Human Development Report 2007/2008, Fighting climate change: Human solidarity in a divided world. New York: McMillan, 2007 United Nations, United Nations Framework Convention on Climate Change Fccc/Informal/84 Ge.05-62220 (E) 200705 (1992), accessed December 1, 2011. http://unfccc.int/resource/docs/convkp/conveng.pdf Uganda Constitution, 1995 Chapter 17 Article 249 in the 1995 Constitution US Department of State, Background Notes Uganda, 2011 http://www.state.gov/r/pa/ei/bgn/ US Department of State, Background Notes Mozambique, 2011 USAID Disaster Assistance Report, 2007 Climate and Development Network (CDKN) Framework, www.cdkn.org Williams, A. Park , and Chris Funk, A westward extension of the warm pool leads to a westward extension of the Walker circulation, drying eastern Africa, Climate Dynamics (2011) World Development Indicators 2011, http://data.worldbank.org/country/mozambique World Bank, Economics of Adapting to Climate Change, Synthesis Report, Washington DC: The World Bank, 2010. World Bank, Economics of Adapting to Climate Change, Mozambique, 2010 356

World Bank/Government of Senegal, Rapport dEvaluation des Besoins Post- Catastrophe Inondations Urbaines Dakar 2009, Dakar: World Bank, 2010 World Bank, Mozambique Country Brief, accessed December 01, 2011 World Bank, Senegal Country Brief, accessed December 01, 2011 http://go.worldbank.org/PO6JPCB5P0 World Bank, Uganda Country Brief, accessed December 01, 2011 World Bank, World Development Report 1992: The Role of the State, Washington DC: Oxford University Press, 1992 World Bank. World Development Report 2003: Sustainable Development in a Dynamic World, Washington D.C.: Oxford University Press, 2003. World Bank: World Development Report 2010: Sustainable Development in a Dynamic World, Washington D.C.: Oxford University Press, 2010. World Resources Institute (WRI) in collaboration with United Nations Development Program, United Nations Environment Program, and World Bank. World Resources 20102011: Decision Making in a Changing ClimateAdaptation Challenges and Choices, ed. P. S. Angell et al. (Washington, DC: WRI, 2011). World Meteorological Organization (WMO), accessed March 10, 2011. www.wmo.int Walfadjri, June 9 2009 Walfadjri, June 4 2009 Walfadjri, July 30 Walfadjri, June 22 2010

357