Social Impact of Discount Food Retail

FAO INVESTMENT CENTRE / EBRD COOPERATION PROGRAMME
REPORT SERIES N. 16 JUNE 2011
Poland, Bulgaria and Romania

Social Impact of Discount Food Retail in Remote Regions
FAO INVESTMENT CENTRE/EBRD COOPERATION PROGRAMME
REPORT SERIES N.16 JUNE 2011
Poland, Bulgaria and Romania

Social Impact of Discount Food Retail in Remote Regions
POLAND, BULGARIA AND ROMANIA: Social Impact of Discount Food Retail in Remote Regions
DISCLAIMER
The designations employed and the presentation ofmaterial inthis information product donot imply the expression ofany opinion whatsoever onthe part ofthe Food and Agriculture Organization ofthe United Nations (FAO) concerning the legal ordevelopment status ofany country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention ofspecific companies orproducts ofmanufacturers, whether ornot these have been patented, does not imply that these have been endorsed orrecommended byFAO inpreference to others of a similar nature that are not mentioned. The views expressed in this information product are those ofthe author(s) and donot necessarily reflect the views ofFAO. All rights reserved. Reproduction and dissemination ofmaterial inthis information product for educational orother non-commercial purposes are authorized without any prior written permission from the copyright holders provided the source isfully acknowledged. Reproduction ofmaterial inthis information product for resale orother commercial purposes isprohibited without written permission ofthe copyright holders. Applications for such permission should beaddressed to: Director Investment Centre Division FAO Viale delle Terme diCaracalla, 00153 Rome, Italy orby e-mail to: Investment-Centre@fao.org FAO 2011
ii
TABLE OF CONTENTS
Acknowledgements....................................................................................................................... vii Acronyms.and.Abbreviations........................................................................................................ viii Executive.Summary.......................................................................................................................ix 1.. Introduction. ............................................................................................................................1 . 1.1 Background ......................................................................................................................1 1.2 Objectives .........................................................................................................................1 1.3 Methodology ....................................................................................................................1 1.4 Organization .....................................................................................................................4 2.. Demographic.and.Socio-Economic.Analysis.ofRegions.inPoland,.Bulgaria.and.Romania.........5 2.1 Demographic Analysis ......................................................................................................5 2.1.1 Population & Change ..............................................................................................5 2.1.2 Population Density ..................................................................................................6 2.1.3 Age Structure ..........................................................................................................7 2.2 Socio-Economic Analysis .................................................................................................8 2.2.1 Income ....................................................................................................................8 2.2.2 Unemployment........................................................................................................9 2.2.3 Mobility and Car Ownership................................................................................. 10 2.2.4 Importance ofthe Agricultural Sector ................................................................... 11 2.3 Statistical Analysis .......................................................................................................... 13 2.4 Summary ........................................................................................................................ 17 3.. The.Transformation.ofthe.Retail.Industry.............................................................................. 18 3.1 Stages inthe Transformation ofthe Retail Industry ........................................................ 18 3.1.1 State Controlled Retailing During the Pre-liberalization Period......................... 19 3.1.2 Privation and Restructuring ofRetailing During Transition.................................. 21 3.1.3 Foreign Investment and Globalization ofRetailing ............................................... 22 4.. Recent.Developments.and.Effects............................................................................................ 25 4.1.1 Continued Growth ofModern Retail .................................................................... 25 4.1.2 Diversification and Growth ofHypermarkets and Discount Formats ................... 26 4.1.3 Retail Investments inSmall Towns and More Remote Areas ................................. 29 4.1.4 Continued Growth ofForeign Participation .......................................................... 30 4.1.5 Increasing Concentration ...................................................................................... 30 4.1.6 Shift Towards Private Standards ............................................................................ 32 5.. Impacts.onConsumers,.Producers,.Traditional.Retail.and.Rural.Society................................. 36 5.1 Impact onConsumers ..................................................................................................... 36 5.1.1 Impact onConsumer Choice and Shopping Behavior ........................................... 36 5.1.2 Impact onConsumer Prices .................................................................................. 37 5.1.3 Impact onProduct Quality .................................................................................... 42 5.1.4 Impact onProduct Variety .................................................................................... 43 5.1.5 Impact onDietary Trends ..................................................................................... 45
iii
5.1.6 Type ofConsumers................................................................................................ 49 5.1.7 The impact ofRural Infrastructure ....................................................................... 51 5.1.8 Impact onRural Services....................................................................................... 51 5.2 Impact onProducers....................................................................................................... 54 5.2.1 Changes inthe Supply Chain................................................................................. 55 5.2.2 Impact onProcessing Companies .......................................................................... 58 5.2.3 Impact onSmall Producers ................................................................................... 58 5.2.4 Impact onAgricultural Employment and Incomes................................................ 61 5.3 Impact onTraditional Retailers ...................................................................................... 66 5.4 Impact onEmployment .................................................................................................. 70 5.4.1 Impact onthe Type ofRetail Employment ............................................................ 70 5.4.2 Impact onthe Quality ofRetail Employment ........................................................ 71 5.5. Gender.Effects. ...................................................................................................................... 76 . 5.6. Impact.onSocial.Cohesion. .................................................................................................... 79 . 6.. References.............................................................................................................................. 81 7.. Appendix................................................................................................................................ 88 7.1 Regional Change inPopulation ...................................................................................... 88 7.2 Regional Population Density .......................................................................................... 95 7.3 Regional Age Structure ................................................................................................. 105 7.4 Regional GDP per capita .............................................................................................. 112 7.5 Regional Unemployment Rate ...................................................................................... 119 7.6 Regional Car Ownership Data ...................................................................................... 126 7.7 Regional Importance ofthe Agricultural Sector ........................................................... 131 7.8 Statistical Analysis: Principal Component Analysis ...................................................... 145 7.9 Foreign Direct Investment inthe Grocery Sales ............................................................ 148
LIST OF FIGURES
Figure 1: The centralized/state approach offood distribution during communism ......................20 Figure 2: Share of the modern retail sector in total and GDP/capita in Eastern Europe ............. 24 Figure 3: Share of the modern retail sector and reform progress in Eastern Europe ................... 24 Figure 4: The Main Shopping Places inselected Central and East European countries in2009 ......25 Figure 5: Share intotal grocery sales ofdifferent modern retail formats inPoland (20042009) .....27 Figure 6: Share intotal grocery sales ofdifferent modern retail formats inRomania (20042009) ..27 Figure 7: Share intotal grocery sales ofdifferent modern retail formats inBulgaria (20042009) ...28 Figure 8: Share offoreign retailers intotal grocery sales inPoland, Bulgaria and Romania ............31 Figure 9: Retailers Self Assessed Standards compared tothose ofthe Government .......................34 Figure 10: Development ofPrivate Labels inthe Low, Medium and Premium Price Brackets atTesco, Carrefour and E.Leclerc inPoland, March 2006 and March 2008 ..................34 Figure 11A: Share ofdifferent food categories inthe total fat consumption inBulgaria (%) ...........47 Figure 11B: Share ofdifferent food categories inthe total fat consumption inPoland (%)..............47
iv
Figure 11C: Share ofdifferent food categories inthe total fat consumption inRomania (%) ..........48 Figure 12: Agricultural Employment (inAWU Annual Working Units).....................................62 Figure 13: Agricultural Productivity: Wheat Yields .........................................................................62 Figure 14: Agricultural Labour Productivity (ALP) ........................................................................63 Figure 15: Evolution ofemployment inthe retail and wholesale sector inBulgaria, Romania and Poland .......................................................................................................72 Figure 16: Evolution ofself-employment inthe retail and wholesale sector inBulgaria, Romania and Poland .......................................................................................................72 Figure 17: Evolution ofwage employment inthe retail and wholesale sector inBulgaria, Romania and Poland .......................................................................................................73 Figure 18: Share ofself-employment inthe retail and wholesale sector inBulgaria, Romania and Poland. .................................................................................................. 73 .
LIST OF TABLES
Table 1: Population and Change inthe Period 20022008 .............................................................6 Table 2: Population Density inBulgaria, Poland and Romania (inhabitants per km) ...................6 Table 3: Age structure ofthe population inBulgaria, Poland and Romania (%; 2009) ..................7 Table 4: GDP per capita inBulgaria, Poland and Romania (19992008; euro) .............................8 Table 5: Unemployment inBulgaria, Poland and Romania (20022008; %)..................................9 Table 6: Agricultural GVA inBulgaria, Poland and Romania (%) .............................................. 11 Table 7: Employment inthe agricultural sector inBulgaria, Poland and Romania (%) ............... 12 Table 8A: Correlation between the indicators for rural regions inBulgaria ................................. 14 Table 8B: Correlation between the indicators for rural regions inPoland.................................... 15 Table 8C: Correlation between the indicators for rural regions inRomania ................................ 16 Table 9: Key characteristics ofthe three stages inthe transformation ofthe retail industry......... 19 Table 10: Number ofretail outlets and stores inPoland .............................................................. 21 Table 11: Size distribution ofstores inPoland ............................................................................. 22 Table 12: HYPERMARKETS: Value Sales, Outlets and Selling Space 20042009 ..................... 28 Table 13: DISCOUNTERS: Value Sales, Outlets and Selling Space 20042009 .......................... 29 Table 14: The spread ofmodern retail insmall cities and towns inPoland in2003 ...................... 30 Table 15: Share ofthe largest supermarkets intotal supermarket sales (%) ................................. 32 Table 15A: Modern retailer astheir main shopping place (% ofthe interviewees) ....................... 38 Table 16: Main shopping place in2010 (% ofthe interviewees) ................................................... 38 Table 17: Monthly expenditures per household member (ineuro) ............................................... 38 Table 18: Main reason toincrease the share oftotal purchase bought inthe modern retailer in2010 (% ofthe interviewees) ..................................................................................... 39 Table 19: Quality requirements requested bymodern retailers .................................................... 43 Table 20: Variety insupply inmodern retailers and traditional shops (number ofproducts) ....... 45 Table 21: Average consumption inthe case study countries (kcal/capita/day) .............................. 46 Table 22: Average fat consumption inthe case study countries (grams/capita/day) ..................... 46
v
Table 23: Sales ofprocessed food products inBulgaria (000 tonnes; 20042009)........................ 49 Table 24: Sales ofprocessed food products inPoland (tonnes; 20042009) ................................. 49 Table 25: Sales ofprocessed food products inRomania (tonnes; 20042009) .............................. 49 Table 26: Differences between consumers ofmodern retailers and traditional shops close tothe modern retailers ................................................................................................. 50 Table 27: Contract terms (% ofthe contracts that includes this).................................................. 57 Table 28: Number ofemployees infood processing companies, number ofcompanies and annual turnover (inMio Euro) .............................................................................. 58 Table 29: Employment inmodern retailers and traditional shops (number employees) ............... 71 Table 30: Percentage offemales employed inthe retail sector inself-employment and asemployee inthe first quarter of2010 (%)........................................................... 77
vi
ACKNOWLEDGEMENTS
This report presents the findings of a study conducted under the cooperation programme between the Food and Agriculture Organization of the United Nations (FAO) and the European Bank for Reconstruction and Development (EBRD). It was written by Professor Johan Swinnen and Kristine Van Herck, from the University of Leuven, under the supervision of Emmanuel Hidier, Senior Economist, Investment Centre Division, FAO. The report was reviewed by Jochen Gorth and Nadia Petkova, Bankers, EBRD Agribusiness team. The study draws, among other things, on a series of interviews carried out in Bulgaria, Poland, and Romania, as well as information collected in these three countries. Important contributions were made by Tomek Woek, Warsaw University, Poland; Christina Harizanova, University of National and World Economy, Sofia, Bulgaria; and Cornelia Alboiu, Institute of Agricultural Economics, Bucharest, Romania. The authors would like to extend their sincere thanks to all government and private sector representatives who kindly shared time with them for their kind assistance. The analysis and opinions presented in this report are the sole responsibility of its main authors, Johan Swinnen and Kristine Van Heck.
vii
ACRONYMS AND ABBREVIATIONS

ALP Agricultural Labour Productivity AWU Annual working units EBRD European Bank for Reconstruction and Development EU European Union FDI Foreigh direct investment GDP Gross domestic product GVA Gross value added MT Ministry oftrade OECD Organisation for Economic Co-operation and Development PCA Principal component analysis PR Predominantly rural regions PU Predominantly urban regions SR Significantly rural regions
viii
EXECUTIVE SUMMARY
Objectives & Methodology
Over the last years, the EBRD has supported through targeted investments the emergence of modern retailing systems in many transition countries. To enhance the economic and social impact of investments in the food retail sector it is important to obtain a better understanding ofthe socio-economic impact ofretail investments onmore remote regions ofthese countries. The first objective of the study is to summarise the conclusions on social impacts of discountand organised retail formats onthe more remote areas ofcountries with economies comparable toPoland, Romania and Bulgaria over alonger period oftime, whilst gathering general demographic and socio-economic data/indicators for each country and providing guidance on splitting these countries into geographic areas (i.e. from relatively disadvantaged to relatively well-off) to help identify locations ofinvestment. The second objective isto validate/qualify the conclusions from the first objective through astudy ofdiscount and organized retail formats operating inPoland, Romania and Bulgaria, and todraw inference and apply these conclusions tojudge the prospective social impact ofpotential investments inremote areas ofRomania and Bulgaria. The analysis and conclusions in this report draw upon a combination of different sources of information and insights. The first source is existing studies. The second source is statistical material from a variety of sources that could be collected through desk study work. The third source isstatistical data and qualitative information wehad collected inprevious studies which isrelevant for the current report. The fourth source isnewly collected data based onnew interviews and surveys. Anumber ofimportant methodological issues should bementioned: Literature review. The existing literature on the social impact of changes in the retail sector is rapidly growing. It yields important insights for the purpose of this study. At the same time both the specific issues addressed, the geographical focus, and the quality ofthe previous analyses and existing studies impose constraints onhow much conclusions can bereliable drawn for the purposes of the current study. In the process of writing this report, we paid special attention toexisting impact studies inBulgaria, Romania and Poland, orEastern Europe more generally. However, because there are only alimited number ofstudies available onthis region, weextended the literature review byproviding evidence from other transition, emerging, and even developing countries. Wewill draw conclusions from these studies tothe extent that wefeel confident that one could extrapolate findings onthe social impact ofchanges from these regions for the retail sector inBulgaria, Romania and Poland. Obviously, the relevance ofstudies from other regions will differ from issue toissue, and wewill treat them accordingly tothe extent possible. Interviews. We have complemented the literature review and data from existing sources with empirical findings based oninterviews done inthe course ofthis study with various stakeholders inthe countries, and inparticular inremote areas ofthe countries under consideration. More specifically, we interviewed and collected data from modern retailers (including discounters), traditional shopkeepers, consumers and producers and their organizations in remote regions in Bulgaria, Romania and Poland. Our interviews took place inJune-July 2010. Itshould beemphasized that,
ix
given the stringent time constraints and the budgetary constraints, inaddition tothe refusal ofsome ofthe key stakeholders tocooperate, itwas not possible tocollect statistically representative and/ orcomplete information. Wehave tried tokeep these constraints inmind when interpreting the data and the information and todraw conclusions ofwhich wewere relatively confident, given the various sources ofinformation atour disposal.
Demographic and Socio-Economic Analysis of Regions

Inthe second section ofthe report, weidentify key characteristics ofvarious rural regions across the countries geographic areas which should beuseful for discussions onwhere new investment inthe retail industry could belocated. For each demographic and socio-economic indicator the report includes avisual summary inthe form ofamap ofthe country based onthe NUTS3 classification. The indicators discussed inthe report are: changes inpopulation between 20022008; population density; share ofthe population older than 65years; GDP per capita; unemployment; car ownership and importance ofthe agricultural sector (interms ofGVA and employment). In general, we find that based on the demographic as well as the socio-economic indicators, Romania is situated between Poland and Bulgaria. For example, GDP per capita isthe highest inPoland and the lowest inBulgaria, while RomaniasGDP per capita isin between. Also interms ofdemographic factors, such aspopulation density and the share ofolder inhabitants inthe total population we can situate Romania between Poland and Bulgaria. The main exception is the share of the population employed in agriculture as agricultural employment is most important inRomania. Inall three countries there are major differences among regions. Based oncorrelation and principal component analysis wefind that ingeneral rural and remote regions have alower GDP per capita, higher unemployment and higher importance ofthe agricultural sector. Many ofthese regions also have alarger share ofold persons inthe total population and high out-migration. However, there are large differences within rural regions. The rural regions that are the worst off are: in Bulgaria, the mountainous regions close to the border with Romania (Vidin and Montana); inPoland, the regions close tothe border with Belarus and Ukraine (Lomzynski, Bialski and Chelmsko-Zamojski) and some regions incentral Poland (Sieradzki, Skierniewicki and Sandomiersko-jedrzejowski); inRomania, the regions close tothe border with Bulgaria (Giugiu, Calarasi and Olt). The rural regions that are the best off are: inBulgaria, the regions close tothe Black Sea (Varna and Burgas); in Poland, the regions close to the border with the Czech Republic (Opolski and legnicko-Glogowski and Bydgosko-Torunski) and the rural regions inthe predominantly urban voivodeship Slaskie (Czestochowski and Bielski); in Romania, the regions located in central Romania (Sibiu and Brasov) and close tothe border ofHungary (Timis) and close tothe Black Sea (Constanta).
The Transformation of the Retail Industry & Recent Developments

Over the past 15 years, the importance of the retail sector has rapidly increased in Bulgaria, Romania and Poland. Starting from almost 0 % inthe years after liberalization/privatization ofthe former state-controlled retail sector, the share of modern retail grew at least 30 % over recent
x
years, with the main growth being observed inthe last 5-8years. This recent growth was driven byacombination ofdemand and supply factors. The combination of these demand and supply factors resulted in three waves of development: (a)thefirst-wave countries are inthe northern half ofCentral Europe, including our case study country Poland, where the share ofmodern retail infood retail isgoing from aniche market inthe early 1990s to4550 % atpresent; (b) the second wave countries are the Balkan countries, including our case study countries Bulgaria and Romania, where the modern retail sector started togrow in the mid/late 1990s and reaches on average 3040 % today; (c) the third wave includes some of southern Central European and all the Eastern European countries. In this third wave area, modern retail growth started inthe late 1990s and early 2000s. There are several key features ofthe modern retail sector inthe case study countries that are crucial tounderstand their social impact. The most recent relevant developments are the following: Diversification and growth of hypermarkets and discount formats. Soon after the take-off of modern retail chains in the region, retailers started diversify the supermarket concept and invested in different formats, such as hypermarkets, discount stores and convenience stores toincrease market coverage, diversify product choice, lower prices and hence increase their market share. This diversification started already inthe beginning ofthe 2000s inthe first wave countries, such asPoland, and bythe mid-2000s the first wave countries tend beheavily hypermarketized and discountered. Inthe second wave countries, such asBulgaria and Romania, modern retailers only recently started todiversify indifferent formats (mid 2000s). Retail investments insmall towns and more remote areas. Inthe same way that foreign investors in the retail sector spread their activities from the relatively-saturated countries (or first wave countries) tothe less-saturated (or second wave and later also third wave countries) and from supermarkets to hypermarkets and discounters, retailers also spread their activities within one country. Initially, modern retail growth started inlarge cities targeting mainly the upper and middle income groups, but when this market gradually became saturated they expanded their focus tolowermiddle and lower income groups asthey started toinvest insmaller cities and even rural towns. Continued growth offoreign participation. Inrecent years, foreign retailers invested heavily inthe retail sector ofCentral and Eastern Europe because itoffered more opportunities for growth than the fairly saturated markets inWestern Europe asthey could benefit from soft local competition, higher mark-ups, growing markets and less strict spatial planning and employment regulations. The share ofFDI inthe retail market has been rapidly increasing and currently, itis 40 % inPoland ,35 % inRomania and 20 % inBulgaria. Increasing concentration. Consolidation, with rapid mergers and acquisitions by larger domestic and foreign chains, increased over time and in2009, the three-firm concentration ratio in supermarket sales was 14 % in Bulgaria, 19 % in Romania and 21 % in Poland. These figures are still low compared tosome northern orwestern European countries and therefore one should expect that inthe future the sector will concentrate even more and will reach the same level asin the EU15. Shift towards private standards. Inthe past years there has been ashift from informal standards oralack ofpublic standards tothe establishment ofmore stringent public and private standards
xi
related to quality and food safety. This is accelerated, and directed, by the entry of Western European chains and the progressive implementation of private standards used throughout the chain regardless ofthe country. Private labels are increasingly important inthe region. For example, in the interviewed store of Penny Market in Romania, the number of private-label products increased inthe last year with 20 %, from 500 products in2009 to600 products in2010.
Impact on Consumers, Producers, Traditional Retail and Rural Society

Investments bymodern retail companies and discounters have important effects onspecific groups insociety. Inthe report, wedistinguish between the effect onconsumers, producers and traditional retailers, gender effects and the impact onsocial cohesion. Impact onConsumers Inthe literature different consumer effects are identified: the impact ofmodern retail onconsumer choice ofshops, onprices, onquality, onvariety, ondiet trends, onthe profile ofconsumers and onrural services. The report discusses each ofthese aspects indetail. Consumer.Choice. The arrival ofmodern retailers and discounters has major effects onconsumer behavior, in particular in that consumers are increasingly and significantly moving away from shopping intraditional retail outlets toshopping insupermarkets, hypermarkets and discounters ifthey are available. This isnot only the case inurban regions, but also inrural regions the modern retailer became the most important place where consumers dotheir grocery shopping. Nevertheless there are important differences between product categories: in the early stage of modern retail development, consumers buy mainly processed and staple food products inthe modern retailer. In an intermediate stage, they also buy semi-processed food products, such as meat and dairy, inthe modern retailer. Finally, inthe most advanced stage, consumers also buy fruits and vegetables inthe modern retailer. Prices. Price isan important reason for rural consumers toswitch from shopping intraditional shops toshopping inmodern retailers. Price differences between modern and traditional retailers largely depend on the product type and the level of modernization of the procurement system. Ingeneral wedistinguish three stages: (1) Early stage: Prices offered in modern retail formats are equal or higher (compared totraditional retailers) for processed food products and higher for fresh food produce. (2) Intermediate (but still early) stage: Processed food ischeaper inmodern retailers, but the results for fresh produce are mixed. This isbecause retailers typically first introduce changes intheir procurement strategies for processed food products (and rapidly afterwards also for semi-processed food products such asmeat and dairy). The changes inprocurement strategies substantially reduce transaction costs and incase ofhigh competition between retailers, retailers will pass onlower transaction costs totheir customers, which will result inlower consumer prices for these products. (3) Advanced stage: Food prices tend tobe generally lower (compared totraditional retailers) for almost all food products, including fresh produce, because modern retailers also adopted the changes inprocurement strategies for these products.
xii
Overall our results seem to suggest that prices in modern retailers in Poland, Bulgaria and Romania are lower compared totraditional shops and that this isan important driver behind the success ofmodern retailers and especially for discount stores for whom the low price policy isparticularly important. Quality. Ingeneral, the quality offood products sold bymodern retailers ishigher than intraditional shops and higher quality and food safety standards are animportant reason for consumers tostart purchasing some products (mainly meat and dairy products) in modern retail outlets. Also for fruits and vegetables, producers indicate that the quality requirements that modern retailers impose are more stringent than those of other trading partners (e.g. traditional wholesalers). However, several studies have argued that the superior quality ofproducts from modern retailers isdebatable and that itis important tomake adistinction between true quality (based onhealth and safety standards) and esthetic quality (based on standards for size, shape and colour). Modern retailers have been criticized totake mainly the latter into account and toreject alarge percentage ofproducts that are produced according tothe health and safety standards, but donot fulfill the modern retailers esthetic requirements. We found mixed evidence of such practices. For example, in Bulgaria, the representative of the producer organization indicated that an important benefit from contracting with the modern retailer, isthat they buy the entire production, including the products which are classified assecond class ornon-standardized. The first class products are sold inthe fruit and vegetable department ofthe modern retailer, while the second class and unstandardized products are used inprepared dishes. Also the managers ofthe modern retail stores that weinterviewed, indicate that they never reject deliveries because of esthetic quality standards. However, 24 % ofthe producers mention that the modern retailer towhom they delivered in2010 rejected atleast once adelivery because of esthetic characteristics while their products were fresh and fulfilled all quality standards. Variety. Modern retailers in general offer more variety in food products. Moreover, we find animportant increase inthe variety offered byboth modern retailers and traditional shops, which indicates that, through increased competition with the modern retail sector, there are positive spillover effects onthe variety offered bytraditional shopkeepers. Dietary. Trends.. Dietary trends in the region show mixed developments. In the past years, fat consumption inBulgaria, Poland and Romania has increased, but when weconsider the type offat that is consumed, we find an increase in the intake of vegetable fat (unsaturated, healthy fat) atthe expense ofanimal fat (saturated, unhealthy fat). Atthe same time, there isan increase inthe consumption of processed food products, which generally include more trans fats, a type of fat ofwhich excessive consumption can lead toserious health risks. However, wehave noevidence ofthe extent towhich these evolutions are related tothe emergence ofmodern retail chains and not toan increase inpurchasing power. Type. of Consumers.. In general, we find no significant difference in age, gender, education orhousehold size between consumers visiting modern retailers and those visiting the traditional shop close tomodern retailer. Rural. Infrastructure.. Also rural consumers that live relatively far from the modern retailer ordiscounter (15-30km) buy atleast some oftheir groceries (mainly processed, staple food and
xiii
beverages) in a modern retailer or discounter, indicating that these consumers are not a priori excluded from the benefits (lower prices, more variety,) that modern retailers can offer. Nevertheless, rural consumers living far from the modern retailer indicate that they would like tobuy more inmodern retailers, but they are mobility constrained indoing soas they donot have acar and public transport isinsufficient. Especially, older and poorer households are found tobe excluded from the benefits that the modern retail sector can offer. Rural Services..While.modern retailers offer their consumers loyalty cards and brochures, traditional retail shops offer other services which modern retailers and discounters typically do not offer, such asproviding consumer credit and delivering products athome. These services appear tobe especially important for older people which are less mobile. Given the mobility constraints, the demographic characteristics (ageing) and the socio-economic characteristics (poorer) ofthe rural population, providing such services isan important contribution tothe local community, but also an asset in the survival strategy of traditional shops in the most remote areas (see also section onthe impact ontraditional retailers). Recommendations tomaximize positive and mitigate adverse social impacts onconsumers Improvement ofrural infrastructure and public transport facilities. This iscrucial tomake sure that all rural inhabitants can benefit from the changes inthe retail sector. Inaddition, investment inrural infrastructure will make rural regions more attractive for modern retail investment. This investment should come from the government, but inthe meanwhile, modern retailers and local governments could engage inprivate-public partnerships toprovide some services and initiatives that increase access tomodern retailers for rural consumers which are currently excluded from the potential benefits that modern retailers may offer. For example, they could provide abus service connecting the village tothe shop. Promote healthy food products. Food consumption isan individual choice and modern retailers are not responsible for their customers. However, the modern retailer may promote the consumption of healthy products, for example by increasing the visibility healthy products or by introducing certain food and safety standards inthe production oftheir private-label products. Inaddition, they could also provide more information onhealthy food products and the potential risks ofan unhealthy diet toincrease consumer awareness. However, not only the modern retailer can provide information, also the government can play animportant role inincreasing consumer awareness, for example, byfinancing TVspots that promote the consumption offruits and vegetables. Stimulate agood balance between true quality and esthetic quality requirements. Currently, producers indicate that modern retailers sometimes reject their deliveries which are produced according to the health and safety standards, but do not fulfill the modern retailers esthetic requirements. Modern retailers could engage to buy a part of the production that does not fit their esthetic requirements (but for example atalower price). These products could besold with adiscount orcould beused inthe preparation ofpre-cooked dishes. Impact onProducers The introduction of modern retail chains induced major changes in the product procurement systems and modernized procurement systems emerged, which translated inimportant changes inthe relationship with local producers.
xiv
Inorder toreduce transaction costs, generate economies ofscale, work with fewer (specialized) wholesalers and have atight control onquality ofthe product, modern retailers prefer towork with asystem where there isone centralized buying office for one product category and one orseveral distribution centers over the country tosupply local stores. Nevertheless, wefound that there are different stages inthe development ofthe centralized distribution system depending onthe product category. Inthe case study countries, wefound that ingeneral, the supply ofprocessed and semiprocessed products is centralized through a distribution center, while the supply of fruits and vegetables isoften still organized atthe store level. For similar reasons aswhy modern retailers prefer towork with alimited number ofwholesalers (reduced transaction costs, tighter control on quality, ), wholesalers (and modern retailers) prefer tocontract with alimited number ofproducers ora preferred supplier system. Usually these contracts include detailed conditions onprice, frequency and quantity ofdelivery and food safety and quality standards that need to be respected. Sometimes these contracts also include farm assistance programs, such asassistance intransport and packaging and advance payments. However, the most important benefit for producers from contracting with specialized wholesalers (or modern retailers) is that they always work with written contracts, which is a way to reduce market risk and have guaranteed sales. In the past years, there was a rapid increase in product variety for all food products, including processed food, such as milk, bread and sugar. However, this is no guarantee that local food processors benefited from these changes. Ingeneral, wefound that the number offood processing enterprises has declined, but atthe same time the size ofthe surviving enterprises, both interms ofemployment and especially interms ofturnover has increased. However, itis unclear towhat extent wecan relate these evolutions tothe changes inthe modern retail sector. Akey concern inthe debate onthe welfare implications ofthese changes isthat the emergence and rapid spread ofmodern supply chains will push alarge share offarmers and inparticular the poor, small farmers, out ofthe market asretailers prefer tocontract with larger and wealthier farmers. However, the impact ofmodern retail investments and the associated standards onsmall producers ismixed. Ingeneral, modern retailers indicate that they prefer topurchase fresh products, such asfruits and vegetables, from large legal entities. When this isnot possible, they supply from small farmers through aspecialized wholesaler. However, there are several constraints for (small) farmers when delivering toamodern retailer: Procurement mechanisms vary between retailers. Some modern retailers are found to pay only three weeks oreven one month after the delivery ofthe products (trade credit), which can be problematic for farmers that do not have a financial buffer to overcome this period. Others pay more promptly. Nevertheless, the producers and their representatives report that procurement mechanisms vary between retailers and infact, our producer survey indicates that in17 % ofthe sales toamodern retailer, the retailer pays apart ofthe price inadvance and the remaining part atthe time ofdelivery. Inorder tobe allowed todeliver tosome large retailers farmers have topay anentrance fee, socalled access charges, which isoften too high for small farmers. Already in2008, the employer organizations and trade unions in Romania, indicated that these access charges make it very difficult for small farmers to deliver to modern retailers because they cannot supply sufficient quantities (the access charges are substantially higher for producers that are only able tooffer
xv
small quantities). Asaresult, the Romanian government imposed aban onsuch practices. However, our interviews with producers and their representatives, indicate that practices such asasking local producers for access charges are used byat least some retailers toexclude small producers from getting acontract with amodern retailer. Small farmers are often not able tocomply tothe quality, packaging and sorting requirements of the modern retailers. Producer interviews indicate that modern retailers impose are more stringent quality requirements than for example traditional wholesalers. Important tonote with respect toquality requirements, isthat wedo not find asignificant difference between the quality requirements that supermarkets and discounters impose ontheir suppliers. However, none of interviewed stakeholders indicate that it is impossible for small farmers toproduce for amodern retailer. All emphasized the important role that producer organizations can play inovercoming the above mentioned constraints. Currently, these organizations already help farmers toconnect tothe market byoffering them assistance programs, such extension services and storage facilities, and establishing contacts between farmers and modern retailers. Inthe past years, while agricultural employment has decreased, productivity and income ofthose that remained infarming, increased. Much ofthis isdue toEuropean integration. European integration relates first to the EU accession and the associated changes in policy (increased subsidies and public standards for food quality and safety). Second, European integration also induced massive foreign investments and trade integration prior toaccession which made technology, capital, etc. available tothe food chains inthese countries. Atthe same time, atleast part ofthe type ofproduct requirements which are now imposed bythe retail chains were already introduced with the global integration ofthese countries over the past years. Inthis sense the imposition ofmore stringent product requirements bymodern retailers ismore anevolution ofaprocess that has started over the past 15years rather than asudden dramatic change with the arrival of new investments. There are two opposing effects. On the one hand, byenhancing the requirements for producers itis likely tohave enhanced production constraints and therefore reinforced incentives for the less skilled and low productive farmers toeither stop farming or (continue to) produce for subsistence purposes orfor local informal markets. Onthe other hand, itwill have reinforced incentives for others to upgrade the quality of their produce and inthis way continue toor enter production for the formal higher-quality, and higher-value, markets. Finally, when discussing the impact ofretail investments onsmall and large farmers itis important tokeep inmind that for most ofthe products which are sold inmodern retailers and discount stores the relationship with farms isindirect atbest. Itis primarily inthe area offresh fruits and vegetables that there isapotential direct relationship between the modern retailer and the farm. For all other products, such asmilk, bread, processed fruits and vegetables, wine, sugar, etc. the impact of retail investments on farms is indirect at most. In these cases, supermarkets and discounters purchase their supplies from food processing companies, such asdairy companies, sugar processors, etc. Most, if not all of these companies, are already operating according to standards imposed byretailers inthe major (urban orforeign) consumer areas and following EUrules. Hence for most ofthese sectors the impact ofretailer and discounter investments onlocal agricultural producers may bequite limited.
xvi
Recommendations tomaximize positive and mitigate adverse social impacts onproducers Prevent unfair trading practices and increase producers awareness on their legal rights. From our interviews it appeared that asymmetry in bargaining power between modern retailers and their producers may lead tounfair trading practices, such asthe existence ofa access charges inRomania and Bulgaria. There could bean commission established that analyzes the existence and use of classic cartels, joint purchasing agreements (buying alliances) and private labels. For such practices a careful balancing of efficiency enhancing and potentially anti-competitive effects isneeded and acase-by-case analysis based onthe specificities oflocal market conditions isnecessary inorder toestablish the existence ofpossible competitive harm. Establish a code ofgood practice (public orprivate). Inorder toresolve the tension between modern retailers and their suppliers, they could engage in a code of best practice. The terms of such a code could include as main elements: (1) compliance with contracts by retailers and suppliers; (2) equal treatment of suppliers; (3) prompt payments; (4) banning of unfair trading practices. There are already some countries that introduced such acode. In2002, aprivate sector code was encouraged bythe competition commission inthe UK (and later itwas made mandatory). Also inArgentina there exists a code ofbest practice. Encourage the development of producer organizations. Modern retailers demand a constant delivery of products and because sometimes small producers cannot provide the demanded quantity, modern retailers may prefer towork with one large producer instead ofseveral small ones. Inorder toovercome this problem farmers must form groups (orproducer organizations) tohelp their members toconnect tothe market byoffering them assistance programs, such asextension services and storage facilities, and byestablishing contacts between farmers and modern retailers However, the willingness to cooperate is still weak in most Eastern European countries. First, itwill beimportant that the government creates aclear legal framework inwhich such activities are possible. For example, inBulgaria, the person incharge ofthe producer organization indicated that the producer organization asalegal entity isnot allowed tobulk the production ofdifferent suppliers, but only to establish contacts between individual producers and modern retailers. Second, there should be a commission that analyzes the existence of joint commercialization agreements and analyzes onacase-by-case analysis whether there exists apotential competitive harm. Encourage innovative private-public actions which increase farmers access tocredit. Anessential problem for (small) farmers isthat they donot have access tocapital tomake investments, which are needed tofulfill the quality requirements that modern retailers impose ontheir suppliers. Insome cases, modern retailers are intermediating between commercial banks and farmers (e.g. offering bank loan guarantees inBulgaria). However, there isascope for innovative private-public actions which increase farmers access to capital. For example, in Mexico there is the financier rural program. This program allows the supplier to get immediate cash instead of having a waiting period for payment and the bank then invoices the payment from the modern retailers. Increase price transparency inthe supply chain. This will allow usto comparison ofprices paid bydifferent stakeholders and provide more information onthe price margin that each stakeholder takes. Second, also for farmers such a tool is informative as currently they have only limited information onprices.
xvii
Impact onTraditional Retailers There is some evidence that the emergence of modern retailers has a negative impact on the survival oftraditional shops. Wefound that, ingeneral, gross receipts oftraditional shopkeepers have declined since the modern retailer was established in the region, which they themselves relate toincreased competition with the modern retailer. Inaddition, wealso find that onaverage the number ofemployees inatraditional shop has decreased since the establishment ofmodern retailers. These findings indicate that modern retailers are putting traditional shopkeepers increasingly under pressure and inthe future weexpect that the share oftraditional shops will decrease even more. Recommendations tomaximize positive and mitigate adverse social impacts ontraditional retailers. Advise traditional shops tooffer services and products that modern retailers donot offer. Increased competition from modern retailers may negatively affect gross receipts (and hence business survival) oftraditional shops. Inorder tomitigate these effects, traditional shopkeepers can focus onproviding certain services that modern retailers donot provide, such ashome delivery, credit provision, Another option is to specialize in certain high quality niche products or offering freshly prepared dishes that are not (or less) available in the modern retailer. Modern retailers can mitigate the impact they have ontraditional shops byproviding advice totraditional shops onoffering new services. For example, Metro Group isplanning tooffer such advisory services totraditional shops when they are setting upcash and carry shops inKazakhstan. Upgrade human capital as this can be a driving factor behind the emergence of new business activities. There will be shopkeepers who are not able to face the competition with the modern retailer. However, this should not necessary beanegative evolution asthe emergence ofamodern retailer can unleash a process of creative destruction. This theory, which was first described bySchumpeter, indicates that there are inventions (e.g. emergence ofamodern retailer) that result inbusiness failures incertain sectors (e.g. traditional shopkeepers), but despite these failures may lead tonet gains because ofthe positive impacts oneconomic activity inother sectors. Inorder tofacilitate this transition, itwill beimportant toupgrade human capital inrural regions asthis can beone ofthe driving factors behind the emergence ofnew business activities. The government could provide support measures for job reconversion, but also modern retailers can provide training programs totrain local people for ajob inaretail outlet. Impact onEmployment Besides job destruction through increased competition, modern retailers can also play animportant role injob creation inrural regions. Since their establishment, employment inmodern retail outlets has increased rapidly. Moreover, wefind some indications that job creation bymodern retailers exceeds job destruction of traditional shops. However, there was a shift from self-employment towage employment. Ingeneral, wefind that both modern retailers and traditional shops have long opening hours and are open onSaturdays and Sundays. These long opening hours are also reflected inlong working hours, especially for self-employed traditional shopkeepers. Also for employees inmodern retail chains, wefind long working hours and inpractice, employees often work longer than officially allowed. Astudy onPoland reports frequent violations ofworking time regulation, such asimposing long
xviii
sequences ofworking days, working longer shifts than official shifts and difficulties intaking breaks during the day. There are also important differences inthe work done byemployees inmodern retailers and selfemployed traditional shopkeepers. Traditional shopkeepers usually combine a series of tasks, from filling the shelves to bookkeeping, for which they need a large variety of skills, while employees inthe modern retailers have more specific tasks (e.g. filling the shelves, keeping the inventories, ). Currently, inthe capital and larger cities retailers inPoland, Bulgaria and Romania have difficulties tofind suitably qualified workers. Most ofthem try toimprove the human capital oftheir work force byoffering attractive remuneration and organizing training toimprove the human capital oftheir work force. However, insmall towns and rural areas, the minimum wage isthe standard, especially as far as low skilled sales agents are concerned. Moreover, in rural Romania, there isalarge population ofworkers who are not legally employed and receive aremuneration below the minimum wage. InPoland, modern retailers are found topay the minimum wage and the rest ispaid under the counter. However, weare not able tocompare earnings inamodern retailer with earnings of traditional shopkeepers as these were reluctant to tell us about their earnings. Moreover, itwould beincorrect tocompare earnings ofaself-employed shopkeeper and earnings of, for example, asalaried cashier, asthey may have different responsibilities and skills. The majority ofthe workforce inthe modern retailers are women (see section onGender Effects), which implies that the that many ofthe employment effects disproportionately affect either women ormen, depending onwhether they are self-employed orbeing hired asan employee inthe modern retail sector. Recommendations tomaximize positive and mitigate adverse social impacts onemployment. Encourage the establishment of regular internal audits. Modern retailers should comply with local employment legislations (working time, minimum wage, ). Currently, there are studies that signal violations ofthis legislation inRomania and Poland. The government plays animportant role indetecting these practices. However, also the modern retailer should take his responsibility. Itis for example possible that the general management isnot aware ofmalpractices inindividual stores ofthe chain (e.g. incase offranchising). Inthat case the modern retailer can set upan internal audit todetect malpractices inthe individual stores and take appropriate measures toavoid future violations ofthe employment legislation. Increase awareness among employees onlegal rights. Besides setting upan internal audit system, modern retailers should encourage (orat least not obstruct) actions that increase awareness among the employees with respect tolabour legislation. Increase labour market flexibility. Inorder tohelp employees, which are mainly women, combine their work and family life, modern retailers could offer their staff the possibility to work parttime, but in addition they could introduce shift patterns, which are communicated a long time upfront inorder toallow employees toarrange possible family obligations. Inaddition, modern retailers could also organize the provision ofchild care facilities (atcost price). The latter will only beprofitable inlarger plants.
xix
Gender Effects The main gender effect from discounter and modern retail investments isindirect through the shift from employment insmall shops tolarger retail outlets, and the associated working conditions. Mainly women benefit from this shift from self-employment towage employment asmainly women are employed asemployees inthe retail sector. Hence, this implies that many ofthe employment effects disproportionately affect either women ormen. The welfare effect for the women employed bythe modern retailer depends onwhat these women were doing before being employed. Ingeneral, most ofthe women that are employed aslower staff (e.g. cashiers) are young and low skilled and they were unemployed orstill atschool when sending their application tothe modern retailer. For these women employment inamodern retailer might have apositive welfare effect inspite ofthe long working hours and low wages. For those previously employed asself-employed shopkeepers, the welfare effects are less obvious and wage employment may imply astep down. However, itis difficult tocompare working conditions and wages inselfemployment and wage employment. Another concern related to gender inequality in modern retailer is that women get no (or less) opportunities than men tobe promoted into store supervision ormanagement. Asthe management ofthe modern retailers was reluctant toanswer this question and prevented from usfrom interviewing their staff, we have only some ad hoc evidence on this matter. This indicates that the problem israther limited. First, several managers ofmodern retail stores that weinterviewed were women, which isafirst indication that gender discrimination israther limited. Second, inaPolish case study onthe working conditions offemale workers insupermarkets/ hypermarkets, the authors also find noevidence ofgender discrimination for jobs instore supervision ormanagement. Nevertheless, the women inthe study mention that itis not uncommon that during the recruitment process questions about marital status, number and age ofchildren and planned pregnancy are asked, but they have noinformation onthe extent towhich employers use this information when selecting acandidate. Recommendations tomaximize positive and mitigate adverse gender impacts Increase awareness on legal rights. Although we have no evidence on gender discrimination bymodern retailers, itis important that there isan employee (for example amember ofthe labour union) which women can approach with complaints and questions about their legal rights. Impact onSocial Cohesion Several studies have analyzed the impact ofmodern retail investments onthe social infrastructure of a community, both from an empirical and a theoretical point of view. However, the impact of modern retail investments on social capital is still unclear. Some studies have pointed at the negative impact of modern retail investments and the associated decline in small local shops on the social infrastructure of regions. They argue that small, local shops and their owners play animportant role insocial relationships, norms and trust inacommunity. Inaddition, the disappearance oftraditional shops also affects the supporting industry within communities that serves these traditional shops. However, other studies have pointed out that there isalso apositive effect ofmodern retail investments onsocial capital aslower prices and more convenient shopping inmodern retailers lead toan increase inthe money and time available for social capital producing products/ activities. Wehave noevidence onthe net effect ofinvestments onsocial capital inthe countries under consideration ingeneral and noclear evidence either way came out ofour interviews.
xx
Recommendations tomaximize positive and mitigate adverse impacts onsocial cohesion Provide facilities for social interaction. Modern retail investments and the associated decline insmall local shops may have anegative impact onsocial interactions within aregion. However, modern retailers are usually located inanew commercial center which inmost cases accommodates facilities where individuals can gather tohave adrink orameal. Ifthese facilities are not provided bythird parties, modern retailers can themselves set upabar orarestaurant. Inaddition, modern retailers can provide for example benches inthe entrance hall and outside the store, where older (and younger) individuals can sit down without being obliged tobuy acoffee orameal.
xxi
1. Introduction
1.1 Background
Since its creation, the European Bank for Reconstruction and Development (the EBRD or the Bank) has invested inaseries ofprojects aiming atstrengthening the distribution offood products. Over the last years, the EBRD has supported through targeted investments the emergence ofmodern retailing systems inthe region, most noticeably inCentral and Eastern Europe, South Eastern Europe, the Western Balkans and Russia. In doing so, the Bank has visibly supported large retailers, such asthe Schwarz Group (Germany) inPoland, Romania and Bulgaria and Cora (France) inRomania and Bulgaria. Ithas also supported local retail groups. For the EBRD, the rationale for supporting investments inthe food retailing sector todate has mostly focussed onthe impact the emergence ofmodern retailers has onthe whole food supply chain. Toenhance the economic and transition impact ofthe Banksinvestments inthe food retail sector it would like to obtain a better understanding of the social impact of retail investments onthe more remote regions ofthese countries.
1.2 Objectives
The objective of this study is to review the potential social implications in these regions, based onan analysis ofthe social impact offood retailing system inother developing regions ofthe world and onprimary data available incountries ofoperation ofthe EBRD where food retailing systems have already started todevelop, and the outcome ofsome field trips. More specifically, the first objective ofthe study isto summarise the conclusions onsocial impacts ofdiscount- and organised retail formats onthe more remote areas ofcountries with economies comparable to Poland, Romania and Bulgaria over a longer period of time, whilst gathering general demographic and socio-economic data/indicators for each country and providing guidance onsplitting these countries into geographic areas (i.e. from relatively disadvantaged torelatively well-off) tohelp identify locations ofinvestment. The second objective isto validate/qualify the conclusions from the first objective through astudy ofdiscount and organized retail formats operating inPoland, Romania and Bulgaria, and todraw inference and apply these conclusions tojudge the prospective social impact ofthe Bankspotential investments onaremote area ofeach ofRomania and Bulgaria, whilst providing recommendations onhow EBRDsfood retail investments could maximise positive-, and mitigate any adverse social impacts.
1.3 Methodology
The analysis and conclusions in this report draw upon a combination of different sources of information and insights. The first source is existing studies. The second source is statistical material from a variety of sources that could be collected through desk study work. The third
1
source isstatistical data and qualitative information wehad collected inprevious studies which isrelevant for the current report. The fourth source isnewly collected data based onnew interviews and surveys. Anumber ofimportant methodological issues should bementioned upfront. First, the existing literature onthe social impact ofchanges inthe retail sector israpidly growing. Ityields important insights for the purpose ofthis study. Atthe same time both the specific issues addressed, the geographical focus, and the quality of the previous analyses and existing studies impose constraints onhow much conclusions can bereliable drawn for the purposes ofthe current study. Second, inthe process ofwriting this report, wepaid special attention toexisting impact studies in Bulgaria, Romania and Poland, or Eastern Europe more generally. However, because there are only a limited number of studies available on this region, we extended the literature review byproviding evidence from other transition, emerging, and even developing countries. Wewill draw conclusions from these studies tothe extent that wefeel confident that one can extrapolate findings onthe social impact ofchanges from these regions for the retail sector inBulgaria, Romania and Poland. Obviously, the relevance ofstudies from other regions will differ from issue toissue, and wewill treat them accordingly tothe extent possible. Third, wehave complemented the literature review and data from existing sources by empirical findings based on interviews done in the course of this study with various stakeholders in the countries, and inparticular inremote areas ofthe countries. More specifically, weinterviewed and collected data from modern retailers, traditional shopkeepers, consumers and producers and their organizations in Bulgaria, Romania and Poland. Our interviews took place in June-July 2010. Ineach country, werandomly selected arural region.1 The selected regions were respectively Ialomita inRomania, Sandomiersko-jedrzejowsko inPoland and Pazardhik and Burgas inBulgaria2. Ineach region, weinterviewed stakeholders insmall towns or ifpossible invillages. Fourth, itshould beemphasized that, given the stringent time and budgetary constraints, inaddition tothe refusal ofsome ofthe key stakeholders tocooperate (see below) itwas not possible tocollect statistically representative and/orcomplete information. Wehave tried tokeep these constraints inmind when interpreting the data and the information and todraw conclusions ofwhich wewere relatively confident, given the various sources ofinformation atour disposal. Fifth, specifically, ineach country, interviews were undertaken with the following stakeholders: (i) Retailers..Ineach country wecontacted aseries ofretailers and inparticular those which have stores located in remote areas. The survey included several open and closed questions on the organizational structure of the chain, the evolution of sales in the store, the variety ofproducts sold, the employment conditions, the procurement strategy (with aspecial focus onfresh produce),
1.- The regions are defined atthe NUTS3 level and inorder todetermine whether aregion isrural, weuse the OECD definition (see section 2.1.2). 2.- InBulgaria, weselected inaddition toPazardhik also Burgas, because wedid not find adiscount store that was willing tocooperate in Pazardhik. In Bulgaria, discount retail formats are only a very recent phenomenon and hence are less widespread in rural areas compared toPoland orRomania.
The reaction was mixed. Some, such asKaufland, refused tocooperate inall ofthe countries. Others were reluctant and only provided limited information. Some were more cooperative and provided more extensive information. Still, wewere able tointerview several retailers and discounters. More specifically, ineach country, weinterviewed the management oftwo supermarkets and two discount stores3, including atleast one foreign retailer with outlets inremote/rural regions ineach country. Nevertheless their willingness tocooperate with the study, the shop managers ofthe interviewed stores asked usfor confidentiality onthe data and therefore weopt togive nodetailed information onwhich stores weinterviewed. (ii).Traditional.shopkeepers4..Wealso interviewed aseries oftraditional shops. Ineach country weinterviewed four traditional shopkeepers, half ofwhich are located inthe proximity ofthe modern retailer and half ofwhich are located atacertain distance from the store (1530km), such that there isless direct competition with the modern retailer. (iv).Consumers..Weinterviewed representatives ofconsumer organizations (except for Bulgaria, where wedid not find asuitable consumer organization) and aseries of60 consumers ineach country. The consumers were selected randomly when they left the shops (half ofthe interviewed consumers were shopping atamodern retailer and half were shopping inatraditional shop5). However, given the limited size of the sample and the absence of a well-designed statistical selection procedure6, the consumer sample may not be fully representative of the local population. (iv) Producers..Weinterviewed ineach country three producer organization specialized infruits and vegetables. The specific commodity specializations varied bycountry: Tomato and cucumber producers in Bulgaria: One organization is exclusively handling tomatoes and has 18 members, all legal entities. The organization is delivering tomatoes to the local market (1 % ofits production), the processing industry (80 %) and modern retailers7 (19 %). The second producer organization counts 43members, amix oflarge (32) and small household farmers (11), and is delivering tomatoes and cucumbers to the local market (30 %), traditional wholesalers (50 %), modern retailers (15 %) and the export market (5 %). The third producer organization is specialized in cucumbers and has 9 members, all small household farmers. The organization is delivering cucumbers to the local market (20 %), the processing industry (40 %), traditional wholesalers (25 %) and modern retailers (15 %). ; Tomato, water melon, cherry, apple, cabbage and potato producers inRomania: One organization isspecialized intomatoes and watermelons and counts 42members, all small household farmers.
3.- InPoland, wewere only able tointerview the management ofone discount store. Wecontacted the management ofseveral other stores ofdifferent retail chains (also inother regions) but none was willing tocooperate with the study. 4.- Inorder tohave some idea onhow the emergence ofmodern retail chains has affected traditional food supply chains, weinterviewed aseries traditional shopkeepers. Traditional shops are important competitors ofmodern retailers, but also open markets play major role inthe food supply chain inrural regions. Therefore wealso incorporate some insights from our producer interviews asalmost all producers that weinterviewed which are selling tomodern retailers, are also selling apart oftheir production onthe open market. 5.- From those that have been shopping intraditional shops weselected half ofthe interviewed consumers intraditional shops close tothe modern retailer and half ofthem intraditional shops far from the modern retailer. 6.- Weselected every fifth consumer that left the shop and that was willing tocooperate with the study. However, this methodology can yield biased results because oftwo reasons. First, only individuals willing toparticipate inaconsumer survey are inthe sample. Second, the consumers are selected atacertain time ofthe day which could bias the results asconsumers buying groceries onMonday morning may have adifferent profile than those visiting the shop onFriday evening. 7.- Inthe group modern retailers, weincluded specialized wholesalers working for amodern retailer.
The organization isdelivering tomatoes tothe local market (30 %), traditional wholesalers (50 %) and modern retailers (20 %) and water melons tothe local market (25 %), traditional wholesalers (40 %), modern retailers (15 %) and the export market (20 %). The second organization isspecialized incherries and apples and counts 42members, amix oflarge (4) and small household farmers (38). The organization isdelivering cherries tothe local market (5 %), the processing industry (90 %) and the export market (5 %), while the organization isdelivering apples tothe local market (96 %) and modern retailers (4 %). The third organization isspecialized incabbage and potatoes and counts 34members, all small household farmers. The organization isselling cabbage tothe local market (15 %), the processing industry (5 %), traditional wholesalers (35 %) and modern retailers (45 %) and potatoes tothe local market (30 %), traditional wholesalers (60 %) and modern retailers (10 %); Apple, pear, tomato and cucumber producers inPoland: One organization counts 35members, a mix of small (18) and large (17) farmers and is apples delivering to the processing industry (5 %), traditional wholesalers (30 %) and modern retailers (65 %). The second organization counts 21farmers, which all have their own cold storage facilities and isdelivering, among other types offruits, apples and pears tothe export market and modern retailers. The third organization counts 10farmers, with afarm size between 2and 6ha, and isselling tomatoes (90 % ofthe sales) and cucumbers (10 % ofthe sales) totraditional wholesalers and modern retailers. Inaddition, weinterviewed ineach country 15individual fruit and vegetable producers delivering tovarious trading partners, including the local market, traditional wholesalers, traditional shops and modern retailers (including discounters). We also used data from extensive survey work among producers in these countries which was collected inearlier studies inthe past years. Sixth, our interviews specifically targeted stakeholders inrural and remote areas, which makes the information complementary toexisting data, virtually all ofwhich isonly available atthe aggregate level orfor urban areas. Hence these interviews provided information which was especially useful for the purpose ofthis study and for comparative purposes.
1.4 Organization
The report isorganized infour parts. Section 2provides ananalysis ofdemographic and socioeconomic data of the three countries concerned. Section 3 analyzes the transformation of the discount and organized retail industry. Section 4discusses the rapid growth and recent developments in the retail sector in Poland, Bulgaria and Romania. Section 5 analyzes expected and actual social impacts ofdiscount and organized retail investments inremote areas ofthe three countries based onkey findings from the literature onthe relative impact ofmodern versus traditional retail ondifferent stakeholders inthe food system complemented byour own data. Anexecutive summary ofthe report isprovided upfront.
2. Demographic and Socio-Economic Analysis of Regions in Poland, Bulgaria and Romania

This section provides acomparative analysis ofregions inBulgaria, Romania and Poland toidentify key characteristics ofremote regions inthese countries. Despite some common features, there are major differences between regions inBulgaria, Romania and Poland interms ofdemographic and socio-economic development. The data presented in this section identify key characteristics of various regions across the countries geographic areas which should beuseful interms ofdiscussions onwhere new investment in the retail industry could be located. For each country, we present regional data on different demographic and socio-economic factors. For each indicator the report includes avisual summary inthe form ofamap ofthe country based onthe NUTS3 classification. These maps and further details, including adetailed list ofthe NUTS3 regions can befound inthe appendix ofthe report. The NUTS3 classification was chosen because itis commonly used byEurostat todivide acountry into different geographical regions.
2.1 Demographic Analysis

The demographic features of a region have an important impact on and are a consequence ofeconomic growth and living conditions inthe region, which inturn affect the type ofconsumers, the number ofconsumers and their expenditures inretail outlets inthe region. Therefore itwill beimportant totake the demographic features ofaregion into account when analyzing the social impact of retailers investments in a certain region. This section analyzes in turn differences inpopulation (change), population density and age distribution between regions. 2.1.1. Population.&.Change The country with the largest population ofthe three countries isPoland, which has slightly more than 38 million of inhabitants, while Bulgaria, with a population of approximately 7,5 million inhabitants, isthe country with the smallest population ofthe three countries. Inbetween isRomania with approximately 21,5million inhabitants (Table 1). All three countries experienced adecline inpopulation inthe period 2002 2008 because oflow fertility rates and high emigration levels. Especially Bulgaria witnessed an extremely decline in population as the total population declined by 3,2 % in the period 2002 2008, a dramatic change for anEU member state. InPoland, the decline inpopulation was much more modest and population only declined by0,3 % inthe same period (Table 1). Besides differences between countries, there are also substantial differences between regions (see Appendix 7.1). InBulgaria, population decline varied between 0,34 % inVarna to12 % inVidin inthe period 20022008. Population increased only inSofia, the capital ofBulgaria. The population decline inRomania was less extreme than inBulgaria. Nevertheless population decreased in most regions and population decline varied between 0,14 % in Sibiu and 6,9 %
5
Table.1:.Population.and.Change.inthe.Period.20022008 2002 Bulgaria Poland Romania 7.891.095 38.242.197 21.833.483 2008 7.640.238 38.115.641 21.528.627 %.change.in20022008 3,2 % 0,3 % 1,4 %
Source: Eurostat Online Database inTeleorman inthe period 20022008. The population increased only in7out of44 NUTS3 regions (Suceava (0,01 %), Brasov (0,12 %), Bucharesti (0,37 %), Constanta (0,82 %), Timis (1,70 %), Iasi (1,75 %) and Ilfov (7,6 %)). The decline in rural population and increase in urban population, as observed in Bulgaria and Romania, isin contrast with the situation inPoland. Inthe period 20022008, the overall population declined somewhat inPoland but atthe same time, there was significant population growth inthe rural regions around the big cities, while the big cities themselves faced adecline inpopulation. This corresponds with asubstantial migration from urban regions tomore rural regions inPoland inthis period. 2.1.2. Population.Density Population density is typically an important factor for retailers in determining where to invest innew retail outlets. After all, they need toreach sufficient consumers inorder tobe profitable. Aswe will document inmore detail insection 4.1.3, modern retailers typically first invest inthe regions with the highest population density, such asthe capital and the larger cities. Later, when these markets get saturated, they consider investing insmaller towns inmore remote and typically more rural regions. Table 2shows that there are substantial differences between the countries interms ofpopulation density. Poland is the most densely populated of the three countries, with a population density ofon average 122 inhabitants per square kilometre. InBulgaria, the population density was the lowest ofthe three countries and in2008 there lived onaverage 69persons per square kilometre. InRomania, population density was intermediate and onaverage there were 93inhabitants per square kilometre. Table.2:.Population.Density.inBulgaria,.Poland.and.Romania.(inhabitants.per.km) Bulgaria Poland Romania 2001 71,3 122,3 97,50 2002 70,9 122,3 94,8 2003 70,5 122,2 94,5 2004 70,1 122,1 94,3 2005 69,7 122,1 94,1 2006 69,4 122,0 93,9 2007 69,0 121,9 93,7 2008 68,8 121,9 93,1
Source: Eurostat Online Database However, besides differences between countries, there are also large differences between regions (see Appendix 7.2). InBulgaria, for example, population density varied between 36,5inhabitants
6
per square kilometre inthe regions outside Sofia, whereas inSofia itself, population density was much higher, namely 919 inhabitants per square kilometre. This is well below the population density ofthe capital region ofRomania, where there lived 10.504 inhabitants per square kilometre in 2007. However also in the rural regions population density in Romania was substantially higher than inBulgaria. InPoland, population density varied between 44inhabitants per square kilometre inSuwalski and 3.307 inhabitants per square kilometre inMiasto Warszawa, the capital region. OECD used these population density indicators todevelop adefinition for a rural region. Regions are classified inone ofthe three categories (OECD 1994; 2005): Predominantly rural regions (PR): if more than 50 % of the population is living in rural communes (with less than 150 inhabitants per square kilometre). However, when there isan urban centre with more than 200.000 inhabitants representing noless than 25 % ofthe population ina predominantly rural region, itis reclassified as significantly rural. Significantly rural regions (SR): if15 % to50 % ofthe population isliving inrural communes (with less than 150 inhabitants per square kilometre). However if there is an urban centre with more than 500.000 inhabitants representing noless than 25 % ofthe population ina significantly rural region, itis reclassified as predominantly urban. Predominantly urban regions (PU): ifless than 15 % ofthe population isliving inrural communes (with less than 150 inhabitants per square kilometre). Applying the OECD definition onthe NUTS3 regional level inthe three countries shows these three countries, looked from aterritorial perspective, are quite rural. The only regions that are classified aspredominantly urban are the capital regions and some regions inPoland (e.g. Krakow, Wroclaw,). All other regions are classified aspredominantly orsignificantly rural (see Appendix7.2). 2.1.3. Age.Structure The age structure ofthe population isan important demographic factor that retailers take inaccount asone should expect differences inconsumer behavior depending onthe age ofthe consumer. Ageing ofpopulation isan important issue inall industrialized countries. Table 3presents the age structure ofthe population inthe three countries. The share ofpopulation younger than 31years isbetween 35 % and 40 %: itis the highest inPoland where approximately 40 % ofthe population isyounger than 31years, while inBulgaria the population ison average older and only 35 % ofthe population isyounger than 31years. InRomania, 39 % ofthe population isyounger than 31years. Table.3:.Age.structure.ofthe.population.inBulgaria,.Poland.and.Romania.(%;.2009) <18year Bulgaria Poland Romania 17 % 19 % 19 % 18-30year 18 % 21 % 20 % 31-45year 22 % 20 % 23 % 46-65year 27 % 27 % 24 % >65year 16 % 13 % 14 %
Source: Eurostat Online Database

7
There is also variation in the age structure among regions (see Appendix 7.3). In Bulgaria for example, inthe northwestern region, Vidin, 23,9 % ofthe population isolder than 65years and inthe mountainous region, Kyustendil, 21,1 % ofthe population isolder than 65years. The only Bulgarian regions where the share ofthe population older than 65years issmaller than orequal to15 % are Sofia (14,8 %) and Blagoevgrad (14,8 %). In Romania, 14,9 % of the population is older than 65 years. However, there are large regional differences. Like inthe case ofBulgaria, the population inthe capital and the region close tothe capital isyounger than inthe rural regions. For example, inBucharest the share ofthe population that isolder than 65years in2009 is14,2 %, while inthe southern rural region, Teleorman, the share ofthe population older than 65years is21,6 %. Incontrast toBulgaria and Romania, the share ofthe population older than 65years issubstantially lower inPoland: in2009 only 13,5 % ofthe population isolder than 65years. There isalso alarge difference inthe regional variation ofage patterns inPoland. Unlike inBulgaria and Romania, the population inthe Polish capital and large cities issubstantially older than inthe rural regions. For example, inMiasto Warszawa, the region ofthe Polish capital, 17,2 % ofthe population isolder than 65years, while inPilski and Poznanski, two rural regions, respectively only 10,9 % and 10 % ofthe population isolder than 65years.
2.2 Socio-Economic Analysis

The socio-economic characteristics of a region, such as income and unemployment rate, have important implications for retail investment inaregion because they affect the purchasing power ofits population. Inthis section, weanalyze regional differences inincome, unemployment, the role ofagriculture inthe economy and mobility. 2.2.1. Income Inthe period 1999 2008 there was avery strong growth inGDP per capita inall three countries. GDP per capita (atconstant 2007 prices) increased by63 % inBulgaria, by45 % inPoland and by66 % inRomania (Table 4). In 2008, GDP per capita in Bulgaria was the lowest of the three countries, namely 4.028 per capita, while inPoland GDP per capita was 8.610 per capita, more than double that ofBulgaria. RomaniasGDP per capita was inbetween: 6.226 per capita in2008. Table.4:.GDP.per.capita.inBulgaria,.Poland.and.Romania.(19992008;.euro*) 1999 Bulgaria Poland Romania 2469 5956 3740 2000 2602 6210 3830 2001 2708 6285 4047 2002 2830 6376 4253 2003 2971 6623 4476 2004 3169 6977 4856 2005 3366 7229 5057 2006 3579 7679 5456 2007 3800 8200 5800 2008 4028 8610 6226
*Constant 2007 prices Source: Eurostat Online Database

8
However, everywhere there are large regional disparities (see Appendix 7.4). Ingeneral, the regions in which the capitals and larger cities (e.g. Sofia, Warsaw, Bucharest, Plovdiv, Cluji, dz#, Wroclaw,) are located, are much richer. GDP per capita can beup tothree times higher than the national average. InBulgaria, the regions closer tothe border with Greece are richer (e.g. Blagoevgrad and Smolyan), whereas the regions close tothe border ofSerbia and Romania are poorer (e.g. Vidin and Montana). Inthe east, Burgas and Varna are two richer regions, because ofthe proximity ofthe Black sea. In Romania, GDP per capita in the rural areas varies between 2.500 per capita in Vasluim, aregion close tothe Moldavian border and 8.400 per capita inTimis, aregion close tothe border ofHungary. Also inPoland, there isalarge heterogeneity inGDP per capita across regions: the regions closer tothe German and Czech border are richer (e.g. Opolski and Bielski), whereas the regions close tothe border ofSlovakia, Ukraine and Belarus are poorer (e.g. Elecki, Lubelski and Przemyski). 2.2.2. Unemployment In the years after transition, liberalization and market reform policies caused a large decline inemployment and consequently asubstantial increase inunemployment. The increase inincome which started inthe second half ofthe 1990s coincided with adecline inthe unemployment rate. Over the period 2002 2008 there was asteady, but persistent decline inthe unemployment rate inall three countries (Table 5). In 2008, unemployment was the lowest in Bulgaria and Romania, where respectively 5,6 % and 5,8 % ofthe active population was unemployed8. InPoland, the unemployment rate was higher and approximately 7 % ofthe active population was still unemployed in2008. Table.5:.Unemployment.inBulgaria,.Poland.and.Romania.(20022008;.%) Bulgaria Poland Romania 2002 na 19,9 8,4 2003 13,7 19,6 7,0 2004 12,0 18,9 8,1 2005 10,1 17,7 7,2 2006 9,0 13,9 7,3 2007 6,9 9,6 6,4 2008 5,6 7,1 5,8
Source: Eurostat Online Database Ingeneral, inall three countries, the unemployment rate islarger inrural regions compared tourban regions (see Appendix 7.5). For example inBulgaria, the unemployment rate varies between 16,7 % inShumen, arural region inthe north-east ofBulgaria, and 2,5 % inthe capital region. InRomania, the region with the highest unemployment rate is the south eastern region Gorj (14,9 %) while inBucharest the unemployment rate isvery low (3,1 %) in2008. InPoland, overall unemployment rates are substantially higher. In22 regions unemployment rates were higher than 8 %, while only
8.- Note that these figures represent the official unemployment figures, which are only anapproximation tothe real unemployment figures. Some part ofthe labour force will bediscouraged from searching for employment and therefore classified as economically inactive or is working on a family farm and does not register as unemployed although not working at full capacity (hidden unemployment).
inthree regions (Trojmiejski, Gdanski and Miasto Poznan) less than 4 % ofthe active population was unemployed. 2.2.3. Mobility.and.Car.Ownership Besides income and employment, also the mobility ofpeople inregion and the ease with which consumers can visit amodern retail outlet will beimportant. However, data onmobility islimited. There are only comparable data onprivate car ownership across countries and regions and these are only available onNUTS 2level, which less detailed than the NUTS 3level that wepresented for all other indicators inthis report. Asindicator ofprivate mobility, weuse data oncar ownership asacomparative mobility indicator across countries and regions (see Appendix 7.6). The country where there are the most cars per hundred inhabitants isPoland. In2008, there were onaverage 42cars per hundred inhabitants. The number ofcars per hundred inhabitants varies between 37cars per hundred inhabitants inthe regions Podkarpackie, Podlaskie and WarminskoMazurskie inthe eastern part ofPoland and 49cars per hundred inhabitants inMazowieckie, the region ofthe capital Warsaw. InBulgaria, there are onaverage 31cars per hundred inhabitants in2008, but there isasubstantial variation across regions. Inthe northern regions, Severozapaden and Severen tsentralen, there are only onaverage 26cars per hundred inhabitants, while inthe capital region there are onaverage 41cars per hundred inhabitants. Car ownership inRomania ismuch more limited compared toBulgaria and onaverage there are only 19cars per hundred inhabitants in2008. Inthe north eastern region, which isthe region that isthe closest toMoldova, there are only 11cars per hundred inhabitants, while inthe capital region there are onaverage 47cars per hundred inhabitants. Unfortunately, there isonly adhoc evidence onpublic transport facilities and road infrastructure, especially inthe most remote regions. In 2007, a survey on the provision of basic services in Bulgaria shows that 82.700 inhabitants of760 rural settlements (orapproximately 4 % ofthe rural population9) have noaccess toadaily bus service tothe municipality centre (Ministry ofAgriculture and Food, 2007). Also inPoland and Romania, there isascope for important improvements inthe coverage ofthe public transport network (mainly buses), especially inregions where the road infrastructure isrelatively poor (Polish Ministry for Agriculture and Rural Development, 2007; Romanian Ministry ofAgriculture and Rural Development, 2007). Also in the other countries an important constraint for the development of rural public transport facilities isthe poor road infrastructure. InRomania, only avery low percentage ofthe communal roads was modernized in2006, varying between 0,28 % inthe Harghita and 25,15 % inthe Olt (five ofthe counties Botosani, Buza Covasna, Tulcea and Vrancea had none ofthe u, communal roads modernized) (Romanian Ministry of Agriculture and Rural Development,
9.- Calculations based onestimates ofthe rural population from the UNdatabase World Urbanization Prospects.
10
2007). InBulgaria, a2007 survey indicated the need toreconstruct more than 8.600 kmof roads and 4.900 kmof streets inrural municipalities and the reconstruction ofmunicipal roads and streets is rated as the highest priority by more than 50 % of the rural municipalities (Ministry ofAgriculture and Food, 2007). These findings are confirmed byour consumer survey. Most ofthe consumers use atheir own car (orthe car offriends orfamily) (62 % ofthe consumers) orgo onfoot (20 %) tothe modern retailer, while only 18 % ofthe consumers uses public transport. Nevertheless, 30 % ofthose that use their own car orgo onfoot indicate that they would use public transport ifthere was amore frequent connection between their house and the modern retailer. 2.2.4. Importance.ofthe.Agricultural.Sector Inall three countries, the agricultural sector still plays animportant role interms ofgross value added (GVA) and employment compared tothe situation inthe EU15. Hence, potential changes inlocal procurement systems bymodern retailers will not only have animportant impact onthe agricultural sector, but also on the overall economy. As we will show, agricultural employment remains very high insome ofthe rural regions, although there are rapid changes. Despite the steady decline inimportance ofthe agricultural sector inGVA over the period 2000 2007, the agricultural sector inall three countries still represents asignificant share inthe economy (and much higher than inthe EU15). InBulgaria and Romania, the share ofagriculture intotal GVA was more than 12 % in2000 but rapidly declined torespectively 6,2 % and 6,5 % by2007, while inPoland this is4,3 % (Table 6). Table.6:.Agricultural.GVA.inBulgaria,.Poland.and.Romania.(%) 2000 Bulgaria Poland Romania 13,9 5,0 12,1 2001 13,4 5,1 14,7 2002 12,2 4,5 12,6 2003 11,7 4,3 13,0 2004 11,0 5,0 14,1 2005 9,4 4,5 9,5 2006 8,5 4,3 8,8 2007 6,2 4,3 6,5
Source: Eurostat Online Database Unsurprisingly, the share ofagriculture inGVA isthe lowest inthe regions where the capital orother important large cities are located (see Appendix 7.7). Nevertheless, there isalarge heterogeneity inthe importance ofthe agricultural sector across regions inthe three countries: In Bulgaria, in 6 out of 26 rural regions the share of the agricultural sector is higher than 16 %: Kardzhali (25,1 %), Silistra (24,3 %), Razgrad (19,3 %), Vidin (19,1 %), Dobrich (18,0 %) and Montana (17,0 %). The only rural region were the share ofthe agricultural sector inGVA islower than 4 % isthe coastal region, Varna, which isdominated bytouristic services. Also inRomania, the agricultural sector still plays animportant role inthe economy ofthe rural regions. The share ofagriculture inGVA varies between 3,5 % inPrahova and 18,3 % inCovasna. In Poland, the agricultural sector is less important than in Bulgaria and Romania. Only intworegions the agricultural sector represented more than 16 % ofGVA: Ostrolecko-siedlecki
11
(19,0 %) and Lomzynski (17,0 %), while in12 regions, the share ofagriculture inGVA isless than 4 %, varying between 0,1 % inTrojmiejski and 3,8 % inJeleniogrski. While the contribution ofagriculture toGVA has declined substantially, interms ofemployment, the agricultural sector still represents alarge share intotal employment inall three countries see Table 7. InRomania, the share ofthe agricultural sector intotal employment isthe highest ofthe three countries and more than 30 % ofthe active population isemployed inagriculture, but also inBulgaria and Poland, the share ofagriculture intotal employment ishigh, respectively 19,7 % and 14,7 %. These shares are huge ifone compares them with the EU15 oreven EU27 average. For example, in2007, agricultural employment inthe EU15 was only 3,4 % and inthe EU27 itwas 5,8 % (Eurostat online database). The very large difference between the share ofGVA and the share ofemployment indicates the low productivity ofagriculture inthese countries. However, italso suggests that the introduction ofthe modern food and retail sectors may have important effects. Table.7:.Employment.inthe.agricultural.sector.inBulgaria,.Poland.and.Romania.(%) 2000 Bulgaria Poland Romania 24,4 26,3 46,2 2001 24,1 19,1 45,7 2002 23,9 19,3 35,4 2003 23,1 18,4 37,6 2004 22,3 18,0 33,3 2005 21,4 17,4 33,3 2006 20,4 15,8 30,6 2007 19,7 14,7 30,3
Source: Eurostat Online Database There isalso substantial variation between regions (see maps inAppendix 7.7). Unsurprisingly, the share ofagriculture intotal employment increases with the degree ofrurality. In rural Bulgaria, the share of agricultural employment in total employment in 2007 varies between 11,1 % inthe coastal region ofVarna and 52,1 % inthe central region ofYambol. InRomania, the share ofthe agricultural employment isvery high and in15 regions more than 40 % ofthe active population isworking inthe agricultural sector. Especially, inthe regions close tothe border ofMoldova and Bulgaria, the share ofagricultural employment intotal employment isvery high. In Poland, the average share of the population working in agriculture is much lower than inBulgaria and Romania. Nevertheless, inseveral regions ineastern Poland the share ofagricultural employment intotal employment ishigher than 40 %. InWestern Poland, agricultural employment isless important.
12
2.3 Statistical Analysis

Inthis section, weanalyze how the different indicators relate toeach other, which will allow usto give some indication where the establishment of discount stores will have the most important impact onthe rural population. Weanalyzed the correlation ofvarious indicators. Inaddition tothe correlation analysis, wealso analyzed the data with a more advanced statistical analysis, a Principal Component Analysis (PCA). The results ofthis analysis are consistent with the results ofthe correlation analysis and yield similar conclusions. The results ofthe PCA are discussed indetail inAppendix 7.8. The correlation between the different indicators for the regions inBulgaria, Poland and Romania are summarized inrespectively Table 8A, Table 8Band Table 8C. Based onthese results, wefind that ingeneral rural and remote regions have alower GDP per capita, higher unemployment and higher importance ofthe agricultural sector. Many ofthese regions also have alarger share ofold persons inthe total population and high out-migration. These are characteristics ofrural and remote regions ingeneral compared tomore urban areas. There are differences between countries. When we consider the main indicators that modern retailers and discounters are most likely totake inaccount when investing inacertain region (GDP per capita, mobility and population density), wefind substantial differences between Romania and the other two countries. For example, while wefound only alimited correlation between GDP per capita and the number ofcars per hundred inhabitants inPoland and Bulgaria, wefind alarge negative correlation in Romania. In addition, we also find a large negative correlation between the share ofolder inhabitants and GDP per capita inRomania, while inthe other two countries this correlation issmaller. Finally, wealso find anegative correlation between the number ofcars per hundred inhabitants and population density in Romania, while in the other two countries, there isapositive correlation. Also for Poland, wefind animportant difference inthe indicators compared to the other countries. While in Bulgaria and Romania, inhabitants of more densely populated regions are ingeneral younger, the opposite holds for Poland, where inhabitants ofmore densely populated regions are ingeneral older. The rural regions that are the worst off are: inBulgaria, the mountanous regions close tothe border with Romania (Vidin and Montana); inPoland, the regions close tothe border with Belarus and Ukraine (Lomzynski, Bialski and Chelmsko-Zamojski) and some regions inthe middle ofPoland (Sieradzki, Skierniewicki and Sandomiersko-jedrzejowski); inRomania, the regions close tothe border with Bulgaria (Giugiu , Calarasi and Olt). The rural regions that are the best off are: inBulgaria, the regions close tothe Black Sea (Varna and Burgas); inPoland, the regions close tothe border with the Czech Republic (Opolski and legnickoGlogowski and Bydgosko-Torunski) and, the rural regions inthe predominantly urban voivodeship Slaskie (Czestochowski and Bielski); inRomania, the regions located incentral Romania (Sibiu and Brasov) and close tothe border ofHungary (Timis) and tothe Black Sea (Constanta).
13
Table.8A:.Correlation.between.the.indicators.for.rural.regions.inBulgaria Population. density Unemployment Population. >.65years GDP. per.capita Cars. per.100. inhabitants Agri.. GVA Agri.. employment
Change. inpopulation
Change inpopulation
1.000
1.000
0.5664 0.289
1.000
Population density Population > 65years 0.520

1.000
0.748 0.104 0.046

1.000
GDP per capita 0.472 0.132 0.558 0.476 0.117 0.627 0.049 0.791 0.648 0.278 0.019 0.213 0.003 0.571
0.473
Unemployment
0.368
14
0.175
1.000
0.422
0.142 0.067
1.000
Cars per 100 inhabitants Agricultural GVA Agricultural employment
0.347
0.579
1.000
Source: Own Calculations based onEurostat Online Database
Table.8B:.Correlation.between.the.indicators.for.rural.regions.inPoland Population. density Unemployment Population. >.65years GDP. per.capita Cars. per.100. inhabitants Agri.. GVA Agri.. employment
Change inpopulation
1.000
1.000
0.091 0.326
1.000
Population density Population > 65years 0.551 0.313

1.000 0.262 0.071 0.093
0.660 0.054 0.044 0.270 0.111 0.352 0.638 0.488 0.264 0.244 1.000
GDP per capita
0.110
Unemployment 0.021 0.424 0.391
0.446
15
0.345
1.000 0.171 0.153

1.000
0.217
0.238
0.718
1.000
Table.8C:.Correlation.between.the.indicators.for.rural.regions.inRomania Population. density Unemployment Population. >.65years GDP. per.capita Cars. per.100. inhabitants Agri.. GVA Agri.. employment
Change inpopulation 1.000 0.146

1.000
1.000
0.519
Population density Population > 65years 0.327

1.000
0.563 0.434 0.132

1.000
GDP per capita 0.030 0.274 0.415 0.199 0.500 0.759 0.153 0.612 0.214 0.164 0.189 0.694 0.205 0.239
0.631
Unemployment
0.077
16
0.570
1.000
0.259
0.310 0.541
1.000 0.300
1.000
0.394
2.4 Summary
Despite some common features, there are major differences between regions inBulgaria, Romania and Poland in terms of demographic and socio-economic development. This section used acomparative analysis ofregions inBulgaria, Romania and Poland toidentify key characteristics ofregions inthese countries. The data presented in this section identify key characteristics of various regions across the countries geographic areas which should beuseful interms ofdiscussions onwhere new investment inthe retail industry could belocated. For each country, wepresented regional data ondifferent demographic and socio-economic factors. For each indicator the appendix ofthe report includes avisual summary inthe form ofamap ofthe country based onthe NUTS3 classification. In general, we find that based on the demographic as well as the socio economic indicators, Romania is situated between Poland and Bulgaria. For example, GDP per capita isthe highest inPoland and the lowest inBulgaria, while RomaniasGDP per capita isin between. Also interms ofdemographic factors, such aspopulation density and the share ofolder inhabitants inthe total population wecan situated Romania between Poland and Bulgaria. The main exception isregarding the share ofthe population employed inagriculture asin Romania agricultural employment is by far more important than in other two countries. In Poland and Bulgaria, agricultural employment isrespectively 14,7 % and 19,7 % in2007, while inRomania, this was substantially higher (30,3 %). Inall three countries there are major differences among regions. Based oncorrelation and principal component analysis wefind that ingeneral rural and remote regions have alower GDP per capita, higher unemployment and higher importance ofthe agricultural sector. Many ofthese regions also have alarger share ofold persons inthe total population and high out-migration. These are the characteristics ofrural and remote regions ingeneral compared tomore urban areas. However, there are large differences within rural regions. The rural regions that are the worst off are: inBulgaria, the mountanous regions close tothe border with Romania (Vidin and Montana); inPoland, the regions close tothe border with Belarus and Ukraine (Lomzynski, Bialski and Chelmsko-Zamojski) and some regions inthe middle ofPoland (Sieradzki, Skierniewicki and Sandomiersko-jedrzejowski); inRomania, the regions close tothe border with Bulgaria (Giugiu , Calarasi and Olt). The rural regions that are the best off are: inBulgaria, the regions close tothe Black Sea (Varna and Burgas); inPoland, the regions close tothe border with the Czech Republic (Opolski and legnickoGlogowski and Bydgosko-Torunski) and, the rural regions inthe predominantly urban voivodeship Slaskie (Czestochowski and Bielski); inRomania, the regions located incentral Romania (Sibiu and Brasov) and close tothe border ofHungary (Timis) and the Black Sea (Constanta).
17
3. The Transformation of the Retail Industry

In the past years, dramatic changes occurred in the retail industry in Poland, Bulgaria and Romania like inmost transition and developing countries. Itevolved from apublic governed distribution system toan increasingly modern and international industry. These changes are not only important from an academic perspective. Infact, tounderstand the current and future effects ofthese changes itis important toput the evolutions and changes into ahistorical perspective. Therefore webriefly discuss the transformation ofthe retail industry over the past two decades and identify the causes ofdifferences among countries and their implications for our current study.
3.1 Stages in the Transformation of the Retail Industry

The transformation ofthe retail industry intransition (and developing) countries can beseparated inthree stages (Dries etal. 2004). The first period the Pre-liberalization Period isthe period when the retail sector was amostly state-controlled economic activity. The second period the Transition Period was characterized by major reforms, including privatization and market liberalization. The third period the Globalization Period isthe period which started inthe mid tolate 1990s and still continues today. Inthis period there was alarge take-off ofthe diffusion of supermarkets and other retail formats, driven by large amounts of foreign direct investment (FDI) from multinational retail companies. Before discussing the three stages of the transformation of the retail sector in detail, we want toemphasize that there are significant differences inthe timing inwhich countries have gone from the second stage tothe third. Within Central and Eastern Europe, there isnot alot ofvariation inthe timing ofthe transition from the Pre-liberalization Period tothe Transition Period: all countries started toliberalize their economies around 1990. However, there ismuch more variation in the transition to the next stage. Some countries entered the Globalization Period already inthe mid-1990s, while others entered only inthe beginning ofthe 2000s. Globally, the spread of modern retail started earliest in South America, East Asia (excluding China), north-Central Europe and South Africa; then inCentral America and Mexico, Southeast Asia, and south-central Europe; and most recently it is now rapidly growing in China, India, Russia, Vietnam, and also emerging inSouthern/Eastern Africa. For the transition region, countries such asPoland, the Czech Republic and Hungary, are first wave countries interms ofretail transformation, starting the globalization period around 1996. Balkan countries such asRomania and Bulgaria are part ofasecond wave, where retail globalization started inthe late 1990s. Russia and Ukraine belong tothe third wave ofcountries, where retail globalization started in2002. Wewill now briefly discuss the various stages. Table 9summarizes some key characteristics ofthe three stages.
18
Table.9:.Key.characteristics.ofthe.three.stages.inthe.transformation.ofthe.retail.industry Preliberalization Concentration inretail sector Dominant source ofcapital Share ofmodern retail Share oflarge multinationals Location ofmodern retail outlets Source: Dries etal. 2004 3.1.1. State.Controlled.Retailing.During.the.Pre-liberalization.Period During the Pre-liberalization Period, governments in transition countries managed all stages of the food production and distribution system (Rozelle and Swinnen, 2004). However, there issome variation between transition countries. Wedistinguish between countries that adopted the centralized/State approach and the decentralized/State-private mixed approach. Most of the transition countries followed the centralized/state approach and the government controlled the nature, volume, prices and margins of the products (Burt, 2006). The physical distribution of products from manufacturers to retailers was controlled by state-led wholesale organizations and both wholesale organizations and retailers were organized as geographic monopolies, with little orno competition among each other (Nowak, 1991; Seitz, 1992). Inmost countries, cooperative retailing inrural areas was permitted alongside the state controlled system, although itwas still under the auspices ofthe government (King, 1988). Spatial segregation between rural cooperative retailing and urban state controlled retailing, incombination with price controls, limited the competition between these two organizational forms (Krasny, 1992). Besides the state managed distribution channels, there existed also informal free markets inmost countries where farmers sold their products directly toconsumers (OECD, 2000a; OECD 2000b). Figure 1 gives an overview of the centralized/state approach to food distribution during communism. InBulgaria, which followed the centralized/state approach, the state purchased practically all harvested crops and animal products from farms and state trusts and organized food processing and distribution. Apart ofthe produce was bought bythe Central Cooperative Union, which was under the auspices ofthe state. The Central Cooperative Union controlled asubstantial number of retail outlets, particularly in villages, and operated open-air markets. Most food stores were state-owned, orco-operative property. Yet, anestimated 30 % ofall food sold was supplied through local markets, inmost cases, these were open-air markets where farmers sell what they cultivate ontheir own small plots (OECD, 2000a). InRomania, also acountry that adopted the centralized/state approach, all marketed output from the state and co-operative farms was procured and distributed bythe state. Private farmers were forced to sell part of their output through contracts with the state-managed distribution system ifthey wanted tohave access toinputs such ascertified seeds, animal feed, vaccines and
19
Transition Low Domestic Low Low Cities
Globalization High Foreign High High Everywhere
High Domestic Low Low
Figure.1:.The.centralized/state.approach.offood.distribution.during.communism
Imports Exports
Vertically integrated units Farm and downstream combined
State trading companies
State procurement
State processors
State retailers
STATE FARMS
CO-OPs
PRIVATE FARMS
Sale of produce at peasant markets
Source: Gorton etal. (1998) chemical fertilisers (OECD, 2000b). The Domestic Trade Department of the Ministry of Trade (MT) was responsible for the whole food distribution system. In each of 41 judets the MT had wholesale distribution enterprises (Intreprindere decomercializare curidicata) which inturn were organized into 421 retail commercial units, each providing food to officially authorized retail shops within their sales area. In total there were about 250.000 retail outlets controlled by the MT, including those owned bythe Domestic Trade Department and operated directly bythe MT, consumer co-operatives and special shops dealing inone product only. The local peasant markets had been tolerated since the 1960s and since there were nosubsidies involved and prices were less controlled than onformal markets, products sold through this channel were relatively expensive. The main products sold inthis way were fresh fruits, vegetables, dairy products and flowers. The share ofpeasant markets inthe total food sales consumed varied in1989 from about 2 % for wheat and rye products to27 % for fruit and fruit products (Word Bank, 1991). Some countries, such asHungary and Poland, followed the decentralized/ state private mixed approach and experimented with anumber offree market reforms which allowed animportant private orat least mixed component inthe retail system. This resulted inan increase inthe quality and range ofmerchandise aswell asan increase inthe competition asthere was anincrease inthe number ofbasic shops close tothe largest housing estates (Michalak, 2001). For example, Poland adopted the decentralized/State-private mixed model before transition such that besides government controlled sales instate enterprises and cooperatives, also private sales in small shops were allowed. However, due to their limited floor surface, private sales are only marginal compared tosales through state enterprises and cooperatives (Karasiewicz and Nowak, 2010). Most sales took place through state enterprises (41 % ofthe sales in1989) and cooperatives (54 % ofthe sales in1989). Hence, private sales represented only 5 % oftotal sales in1989. Interms ofthe number ofshops, the share ofthe private sector ismore important. In1989, they represented 19 % of the total number of shops, while state owned shops represented 18 % of the shops and cooperatives 63 % (Karasiewicz and Nowak, 2010).
20
Consumers
3.1.2. Privation.and.Restructuring.ofRetailing.During.Transition There are two waves oftransformation that altered the structure ofretailing inthe transition countries. Inthe transition period this was the privatization ofretail and wholesale companies (latter during the globalization period FDI liberalization will lead to the second wave of transformation in thesector). The privatization ofthe retail industry was generally associated with the breakdown ofthe large state owned retail enterprises into smaller separate units. For example: in Bulgaria, the number of retail outlets more than doubled between 1990 and 1995 (from 15.000 in1990 to44.000 in1995) (Bushnakova, 2003); inRomania, the retail sector largely expanded inthe first years after transition and between 1990 and 1994 the total number ofretail outlets rose from about 64.000 in1990 to183.000 in1994 (OECD, 2000b); inPoland the number ofretail outlets increased from 250.000 in1989 to850.000 outlets in1994 and the number ofstores increased from 152.100 in1989 to415.400 in1994 (Table 10) (Karasiewicz and Nowak, 2010). Table.10:.Number.ofretail.outlets.and.stores.inPoland Number.ofretail.outlets. (inthousands) 1989 1990 1995 2000 250 470 890 860 Number.ofstores. (inthousands) 152,1 237,4 425,6 432 Population. per.store 250 161 91 89
Source: Karasiewicz and Nowak, 2010 The privatization process is also reflected in the rapidly increasing share of private food sales. For example, by1994, private food sales represented 73 % and 74 % inrespectively, Bulgaria and Romania (OECD 2000a, 2000b). Inthe same year, this was already 90 % inPoland (Karasiewicz and Nowak, 2010). Most of the private shops developed from petty traders selling from street kiosks and stalls. The vast majority ofthese new outlets are small shops, with less than 50square meters offloor-space and awide variety ofproducts. Latter, these stores started tomerge and form small retail chains (Dries etal., 2004). Table 11shows the size distribution ofstores inPoland inthe transition period.
21
Table.11:.Size.distribution.ofstores.inPoland Year 1993 1994 1995 1996 1997 Number. ofstores. (inthousands) 380,6 415,4 425,6 405,6 424,4 <.50. sq.m. 347,7 383,1 391,3 369,9 387,9 51100. sq.m. 19,9 10,4 20,3 20,7 20,6 101200. sq.m. 8,2 8,1 8,6 8,9 9,2 201300. sq.m. 2,1 2,0 2,2 2,4 2,6 301400. sq.m. 0,9 0,9 1,0 1,1 1,3 >.400. sq.m. 1,9 2,0 2,2 2,6 2,9
Source: Dawson and Henley (2000) 3.1.3. Foreign.Investment.and.Globalization.ofRetailing In the transition period almost all investment was with domestic capital as there were only limited retail FDI inflows. In this period, foreign investors were reluctant to invest in the retail sector oftransition countries indicating that reforms and astable policy environment are crucial for investment and contracting asthis isneeded tosecure property rights and facilitate contract enforcement. Foreign investment inthe retail industry started inthe most economic advanced countries, that also moved the fastest in their reform progress. These are the so-called first wave countries, such as Poland. Already in the early 1990s, when the market was still very unstable, there were some pioneers that entered the Polish retailing market (Dawson and Henley, 2000). Examples ofthese pioneers inretailing are the Austrian retailer Billa, the Nordic discounter REMA 1000 and the German retailer Makro. Billa entered the Polish market in1990 and developed initially supermarkets in Warsaw and Bielsko-Biala. The discounter REMA 1000 entered the market in1993 through aform offranchise system totarget the discount food market. Makro opened its first store in1994 inWarsaw and has expanded considerably since then. After the investment of a few pioneers in the beginning of the 1990s, the modern retail sector inPoland started togrow rapidly inthe mid-1990s. In1995, the French retailers Leclerc, Auchan, and Docks deFrance opened several stores inPoland and in1996 they were followed byCasino. Also in1995, Jeronimo Martins, from Portugal, entered joint ventures tooperate achain oflocal discount food stores and a cash and carry chain. In the same year, the British retailer Tesco purchased achain ofsmall supermarkets. Besides, investment insupermarkets and hypermarkets, also foreign discounters emerged when in1995 Metro and Tengelman entered with their discount formats ofrespectively Tip and Plus (Dawson and Henley, 2000). In the countries that were less advanced in the reform process, such as Bulgaria and Romania, ittook much longer until the first foreign pioneers invested inthe retail sector. InBulgaria and Romania, the rapid growth of the retail sector only started at the end of the 1990s and in the beginning ofthe 2000s. InBulgaria, only atthe end ofthe 1990s the first modern supermarkets and hypermarkets were established, mostly in Sofia and in the bigger cities, such as Plovdiv and Pleven. Some of these
22
investments inthe retail sector were financed byforeign capital (e.g., Metro (German) and Ena (Greek)). However, the bulk ofearly investment inthe retail sector was ofBulgarian origin and in the smaller towns and villages the stores of the Co-operative Union continued to operate (OECD, 2000a). The first multinational toinvest inthe retail sector was the Metro Group, who invested in1999. The next year, in2000, the Austrian retail company, Billa, opened its first shops inBulgaria and later in2005 and 2006 also respectively T-Market and Kaufland invested inthe Bulgarian retail sector. InRomania, economic instability and slow reforms slowed down foreign investment inthe retail sector and only in 1997, the first foreign investors entered the market, when Kolos (Danish) and Metro (German) opened their first supermarkets in the large cities. In the same year, also aBelgian-Romanian joint venture (LaFourmi) opened five supermarkets inBucharest (OECD, 2000b). However, overall there was only little foreign investment and most food sales were taking place through the traditional channels. This isalso reflected inthe four-firm concentration ratio inretailing, which was very low inRomania (about 1 %) in1998, compared to60 % inthe UKand 42 % inGermany atthe beginning ofthe 1990s (Gorton etal., 1998). After the pioneer investments ofMetro and Kolos inthe mid-1990s, the next multinational toinvest inthe Romanian retail market was the Austrian retailer, Billa, who entered the market in1999. Billa ispart ofthe German Rewe trade group and subsequently, the German group introduced its cash & carry stores, Selgros (1999), and the discount stores, XXL Mega Discount and Penny Market (in2005), inRomania. In2001, Carrefour invested inthe Romanian market and opened its first hypermarkets inBucharest, but rapidly expanded its activities toother large cities inthe country.
Box 1: Determinants ofthe speeD ofthe retail inDustry transition The retail transformation was driven by a combination of demand and supply factors (Reardon etal. 2003). Demand isdriven by (1) increasing per capita incomes; (2) urbanization, which improved access topublic transport and led tomore employment opportunities for women (and hence an increase in the opportunity cost of womens time which increased the demand for processed and packaged food); (3) areduction ofeffective food prices for consumers because ofsupermarket chainsmass procurement and efficient merchandising. Onthe supply side, increased investment inthe retail sector was driven bytwo determinants: (1) policy interventions such aspublic investments, market liberalization, trade liberalization and FDI liberalization; (2) FDI and competitive domestic investments fueled byagro-food industry entrepreneurs seeking economies ofscale, scope, and specialization. The relationship between the demand/ supply factors and the diffusion of modern retail isreflected inthe correlation between, onthe one hand, the share ofmodern retail intotal retail, and onthe other hand, the income level ofthe country and the progress ofreforms, which isillustrated byFigures 2and 3 (data for Bulgaria, Croatia, Czech Republic, Hungary, Poland, Romania, Russia, Slovakia, Ukraine).
23
Figure.2:.Share.ofthe.modern.retail.sector.intotal.and.GDP/capita.inEastern.Europe
60 Share of modern retail on total (%) 50 40 30 20 10 0
R Sq. = 0,82
1998 2002
1000
2000
3000
4000
5000
6000
7000
GPD/capita (USD)
Source: Dries etal. (2004) Figure.3:.Share.ofthe.modern.retail.sector.and.reform.progress.inEastern.Europe

60 Share of modern retail in total (%) 50 40 30 20 10 0 2 2,5 3 EBRD Reform Index 3,5 4 R Sq. = 0,79 1998 2002
Source: Dries etal. (2004) Figure 2shows that there was astrong positive correlation between the share ofmodern retail intotal retail and the income level ofthe country. Figure 3shows that there was astrong positive correlation between the extent ofreforms and the growth ofthe modern retail sector. These data suggest that there appears tobe aminimum level ofincome and reform and that once beyond this level, the modern retail sector grows exponentially. For the first wave countries, this process started inthe mid/late 1990s, while in second wave and third wave countries itstarted inrespectively 2000 and 2002.
24
4. Recent Developments and Effects

4.1.1. Continued.Growth.ofModern.Retail In the past years the share of modern retail has rapidly expanded. Figure 4 indicates that the modern retail sector (supermarkets, hypermarkets and discount stores) isnow the main shopping place for asignificant proportion ofthe population inEastern Europe. Throughout the first wave countries and inthe urban areas ofthe second wave and third wave countries, the majority ofthe respondents indicated that modern retailers were their main shopping place. For example, inthe Czech Republic 80 % ofthe respondents indicate that amodern retailer istheir main shopping place and also inMoskow (Russia) approximately 80 % ofthe population indicated that supermarkets, hypermarkets ordiscount stores are the main place where they dotheir grocery shopping. Figure.4:.The.Main.Shopping.Places.inselected.Central.and.East.European.countries.in2009
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Bulgaria Czech Republic Hungary Poland Romania Russia Slovakia (Urban) (Moskow) Slovenia Ukraine (Kiev) 20% 56% 25% 16% 22% 31% 34% 16% 11% 15% 18% 9% 17% 27% 10% 33% 30% 39% 23% 18% 21% 39% 23% 42% 48% Hypermarket Supermarket Discount store Other 10% 39% 29% 29% 42% 35% 30% 29% 13%
Source: Shopping monitor Central & Eastern Europe 2009 Figures 47 illustrate how the share of the modern retail sector (supermarkets, hypermarkets and discount stores) in total retail grew strongly in Poland, Bulgaria and Romania in recent years, based onwhere the main shopping place was (infigure 4) and asashare ofgrocery sales (figures 57): InPoland, around 70 % ofpeople identify modern retail outlets astheir main shopping place. The share ofthe modern sector intotal grocery sales grew from 33 % to48 % between 2004 and 2009. InRomania, the share ofmodern retail ingrocery sales has grown rapidly over the past five years: from 16 % in2004 to42 % in2009, close tothe share inPoland. There are only data available
25
for urban Romania on main shopping place and these data show a very strong domination ofmodern retail: around 85 %. Bulgaria isstill lagging inthis development. Onaggregate inBulgaria more than half consumers still shop inmore traditional retail outlets. Interms ofgrocery sales, the share ofmodern retail has increased from 14 % in2004 to29 % in2009. 4.1.2. Diversification.and.Growth.ofHypermarkets.and.Discount.Formats Soon after the take-off of modern retail chains in the region, retailers started diversify the supermarket concept and invested indifferent formats, such ashypermarkets, discount stores and convenience stores toincrease market coverage, diversify product choice, lower prices and hence increase their market share. Figure 4shows that the relative importance ofthe different formats varies substantially between the countries. Inparticular, within Central and Eastern Europe, discount stores appear much more important asmain shopping place in, for example, Hungary, Czech Republic and Poland (22 % to31 %) than in, for example, Slovakia and Slovenia (around 10 %). Figures 5-7and Tables 12-13show the evolution ofthe different retail formats (e.g. share, number of outlets, the sales value and the selling space of respectively hypermarkets and discounters) inPoland, Bulgaria and Romania. InPoland the number ofhypermarkets has increased from 229 outlets in2004 to389 outlets in2009, oran increase of69 % (Table 12). Also the market share ofhypermarkets increased inthe past five years: from 16 % in2004 to20 % in2009 (Figure 5). Similar results hold for the discount sector where the number ofoutlets increased from 1233 outlets in 2004 to 2083 outlets in 2009 (Table 13) and the market share increased from 8 % in2004 to13 % in2009 (Figure 34). These numbers indicate that inPoland, format diversification already started in the beginning of the 2000s and that by the mid-2000s, hypermarkets and discount stores played avery important role inthe Polish retail market. InRomania, there was avery rapid growth ofhypermarkets and discount stores inthe past 5years. There were only five hypermarket outlets in2004 (oramarket share of6 %) but by2009, this number had increased to104 outlets (oramarket share of25 %) (Table 12). There was also alarge increase inthe number ofdiscounters: in2004 there were 20outlets with acombined market share ofbarely 1 %; while in2009, there were 278 outlets, with acombined market share of8 % (Table 13and Figure 6). InBulgaria, aswith other indicators, the process ofgrowing importance ofhypermarkets and discount stores has been slower than inPoland and Romania. There were 10hypermarkets in2004, with a combined market share of 3 %. By 2009, the number of outlets increased to 36 and the market share to7 % (Table 12and Figure 7). In2009 there were nohard discounters operating inthe Bulgarian retail market (Table 13and Figure 7). The retailer which most closely resembled this format was Kaufland, ahypermarkets
26
Figure.5:.Share.intotal.grocery.sales.ofdifferent.modern.retail.formats.inPoland.(20042009)
100% 90% 80% Share in total grocery sales (%) 70% 60% 50% 40% 30% 20% 10% 0% 2004 2005 2006 2007 2008 2009 67% 65% 64% 60% 57% 16% 9% 8% 16% 15% 17% 18% 20%
10% 8%
12% 9%
13% 11%
14% 12%
15%
13%
Hypermarkets Supermarkets Discounters Other
52%
Source: Euromonitor (2010) Figure.6:.Share.intotal.grocery.sales.ofdifferent.modern.retail.formats.inRomania.(20042009)

100% 90% 80% Share in total grocery sales (%) 70% 60% 50% 40% 30% 20% 10% 0% 2004 2005 2006 2007 2008 2009 84% 82% 77% 6% 9% 1% 7% 9% 1% 11% 9% 4% 8% 5%
19%
22%
25%
9% 6%
9% 8% Hypermarkets Supermarkets Discounters
68%
62%
Other 58%
Source: Euromonitor (2010)

27
Figure.7:.Share.intotal.grocery.sales.ofdifferent.modern.retail.formats.inBulgaria.(20042009)
100% 90% 80% Share in total grocery sales (%) 70% 60% Hypermarkets 50% 40% 30% 20% 10% 0% 2004 2005 2006 2007 2008 2009 84% 82% 79% 76% Supermarkets Discounters 72% 71% Other 3% 12% 4% 13% 5% 16% 6% 18% 6% 7%
21%
22%
Table.12:.HYPERMARKETS:.Value.Sales,.Outlets.and.Selling.Space.20042009 2004 Poland Value sales (PLN million) Outlets Selling Space (000 sqm) Value sales (BGN million) Outlets Selling Space (000 sqm) Value sales (RON million) Outlets Selling Space (000 sqm) 19644,7 229 15015 158,2 10 35 1317,5 5 53 2005 20509 251 16348 206,1 12 44 1991,5 9 81,5 2006 19536 276 18058 291,7 25 89,9 3464,6 36 198,1 2007 23167,7 317 19358 314,2 28 101,5 7688,4 64 336,8 2008 25803,9 352 20630 376,3 34 119,1 10845,4 89 440,8 2009 30278 389 21940 412,8 36 134,7 13838,8 104 526
Bulgaria
Romania

28
chain with alow-price positioning which isperceived asasoft discounter bymany consumers. The success ofKauflandslow-price strategy inhypermarkets suggested that there islarge potential for the conventional discounters format inthe Bulgarian market and anumber ofinternational discounters chains have confirmed their intention toenter the Bulgarian retailing market, most notably Lidl, Penny Market and Plus (Euromonitor, 2010). Inthe beginning of2010, Penny Market and Plus opened their first stores inSofia and other large cities inBulgaria. Currently, there are 24Plus stores and 39Penny Market stores inBulgaria. Table.13:.DISCOUNTERS:.Value.Sales,.Outlets.and.Selling.Space.20042009 2004 Poland Value sales (PLN million) Outlets Selling Space (000 sqm) Value sales (BGN million) Outlets Selling Space (000 sqm) Value sales (RON million) Outlets Selling Space (000 sqm) 9569,5 1233 747,5 0 0 0 235,6 20 23,7 2005 10622,4 1421 899,9 0 0 0 568 57 58,6 2006 11410 1424 790,4 0 0 0 1142 104 104,4 2007 14720 1660 924 0 0 0 1828,5 151 151,2 2008 17468 1860 1070,9 0 0 0 3139,5 206 203,1 2009 19586 2083 1194,8 0 0 0 4416,6 278 273,5
Bulgaria
Romania
Source: Euromonitor (2010) 4.1.3. Retail.Investments.inSmall.Towns.and.More.Remote.Areas Inthe same way that foreign investors inthe retail sector spread their activities from the relativelysaturated countries (or first wave countries) tothe less-saturated (or second wave and latter also third wave countries), retailers also spread their activities within one country. Initially, retailers mainly invested inthe large cities, where the living standard isin general higher, but when this market became more and more saturated they started toinvest insmaller cities and towns. Because ofthe lower population density insmaller towns and rural areas, there isinherently less competition among modern retail chains inthese areas. Often the company which introduces the first hypermarket inasmall town may effectively lock out competition asthere may only beamarket for one major store (Dries etal. 2004). Information on this process of investments in rural and more remote areas is more limited. InPoland, there isevidence ofasignificant number ofmodern retail outlets located insmaller cities and towns. In2003, more than 40 % ofthe Albert supermarkets (part ofthe Dutch Ahold group) were located in towns with less than 50.000 inhabitants (Table 14). Besides supermarkets, there
29
are also several discounters active insmaller cities and towns. In2004, 52 % (or352 stores) ofthe Biedronka network (part ofthe Portuguese Jeronimo Martins Group) were active inPolish towns with apopulation ofup to50.000 inhabitants and also Lidl (part ofthe German Scharwz Group) had approximately 35 % ofits network inthese small towns inPoland (Wilk, 2006). InRomania, the discounter Penny Market (part ofthe German Rewe Group) isactive inseveral smaller cities and towns outside Bucharest. In2010, 30out ofthe 51Penny Market stores (ormore than 58 %) were located intowns with less than 50.000 inhabitants. Also inBulgaria, the discounter Penny Market has most ofits stores located intowns with less than 50.000 inhabitants and in2010, 22out of39 stores (more than 56 %) were located insuch small towns. Table.14:.The.spread.ofmodern.retail.insmall.cities.and.towns.inPoland.in2003 Town.size (000.inhabitants) < 20 2050 50100 100200 200500 > 500 Share.oftown.group.in: Urban.population 20,8 % (663)* 17,4 % (132) 14,1 % (49) 12,7 % (22) 16,1 % (13) 18,9 % (5) Number.ofAlbert.supermarkets 12,2 % 28,7 % 17,1 % 18,3 % 10,4 % 13,4 %
Note: (*) inparenthesis isthe number oftowns inthe group Source: Wilk (2006) 4.1.4. Continued.Growth.ofForeign.Participation Foreign investors continue toincrease their share inthe retail sector. Currently, the share offoreign investment intotal food sales grew very strongly inall three countries. Itis about 35 % inRomania, 40 % in Poland and 20 % in Bulgaria (Figure 8). Using 2009 data, we estimated that foreign supermarkets have combined more than $3billion offood sales per year inPoland, Bulgaria and Romania. Table 15lists the top companies intotal supermarket sales10. Foreign companies dominate the list. Yet, there isagreat diversity ofmultinational retailers that have invested inthe region. InPoland, Bulgaria and Romania, the main foreign investors insupermarkets were Billa (Austria), Carrefour (France), Van Holding (Serbia), Tesco (UK) and Mega Image (Belgium). Only inBulgaria, the national retail chain, Bolyari AD, manages tohave dominant position insupermarket sales (third largest player in2009). 4.1.5. Increasing.Concentration Over the three stages, we observe a U shaped pattern of concentration over time. Before transition, the retail sector was highly concentrated, while in the transition period privatization
10.- Foreign companies also dominate the total grocery sales in the three countries. In appendix 7.8, one can find the tables that illustrate this. Total grocery sales include besides grocery sales insupermarkets also sales inhypermarkets and discount stores.
30
Figure.8:.Share.offoreign.retailers.intotal.grocery.sales.inPoland,.Bulgaria.and.Romania
45% 40% Share of foreign retail in total grocery sales (%) 35% 30% 25% 20% 15% 10% 5% 0% 2005 2006 2007 2008 2009
Poland Bulgaria Romania
Source: Euromonitor (2010) caused that the sector became de-concentrated and usually very small, independent shops arose. Inthe globalization period, the multinationalization and inflow ofFDI led toare-concentration ofthe sector. However, this time itis inhands ofthe private sector and not state-owned like before transition. Table 15illustrates this growing consolidation inthe supermarket sector. In2005 and 2009, the top five supermarkets in Bulgaria represented respectively 42 % and 59 % of supermarket sales, while inRomania, this was respectively 53 % and 61 %. InPoland, this number was slightly lower, namely 41 % in2005 and 57 % in2009. These figures are inline with the EU15 average atthe end ofthe 1990s when Clarke etal. (2002) reported that the five-firm concentration ratio inthe grocery retailing sector was close to50 %. However, recent figures byEinarsson (2007) show that the grocery market inthe EU15 isnow even more concentrated: the three-firm concentration ratiosfor 2004 are Denmark (91,2 %), Finland (79,6 %), Iceland (81 %), Norway (82 %), Sweden (91,2 %). These are huge numbers compared tothe Eastern European countries, where in2009 the three-firm ratio inthe grocery market is14 % inBulgaria, 19 % inRomania and 21 % inPoland. Therefore one should expect that inthe future this sector will concentrate even more and will reach the same level asin the EU15.
31
Table.15:.Share.ofthe.largest.supermarkets.intotal.supermarket.sales.(%) BULGARIA Billa Bulgaria EOOD (FDI) Van Holding AD (FDI) Bolyari AD VPMarket Bulgaria EOOD (FDI) CBA Bulgaria AD (FDI) Wild GmbH & CoKG, Rudolf (FDI) Magazini Evropa AD Agro-Industrial EOOD Others POLAND Carrefour Polska Spzoo (FDI) Tesco Polska Spzoo (FDI) ITM Polska Spzoo (FDI) Polomarket SpZoo Piotr iPawel Spzoo Eko Holding Spzoo Grupa ELeclerc Spzoo (FDI) Emperia Holding SA Others ROMANIA Billa Romania SRL (FDI) Mega Image SA (FDI) Artima Retail Investment CoSA (FDI) Gimrom Holding SDA Ivet Comprod Srl Spar Merchandising Romania SRL (FDI) Angst ROSA (FDI) Oncos Impex Srl Others Source: Euromonitor (2009) 4.1.6. Shift.Towards.Private.Standards Generally, growing demand for high standards isanatural consequence ofincome growth. Inrecent years it has been reinforced by several additional events. For example, international campaigns against child labour and genetically modified food, NGO activities expressing growing concerns onclimate change and the loss ofbiodiversity and several food safety crises inthe EU, such asthe
32
2005 13,4 18,8 3,4 1,4 2,8 3,5 1 55,7 2005 7,9 2,7 13,1 8,5 5,4 5,9 4,3 52,2 2005 29,7 7,8 8,2 4,7 1,4 2,8 2,7 42,7
2006 15,2 17,4 6,3 6,7 3,1 3,4 1,2 46,7 2006 7,6 10,9 12,4 9,1 4,9 5,9 3,9 45,3 2006 30,6 7,7 9,7 4,8 1,7 0,9 3,1 3,2 38,3
2007 14,3 16,8 7,3 10,4 3,1 3,1 1,5 1,2 42,3 2007 13 13,5 12,1 10,8 5,6 6,8 3,6 5,1 29,5 2007 34,3 10,6 10,6 6,3 2,5 4,1 3,6 3,1 24,9
2008 16 17,4 8,1 10,7 4,2 2,7 1,6 1,1 38,1 2008 12,2 12,7 11,9 11,1 6,3 6,5 3,3 5,3 30,7 2008 27,6 10,6 9 5,1 2,8 3,8 3,2 2,5 35,4
2009 17,7 16,8 10,1 9,9 4 2,4 1,6 1,1 36,4 2009 13,9 13,3 12 10,3 7 6,2 5,4 5,1 26,8 2009 31 14,3 8,4 3,8 3,2 3,1 3 2,5 30,7
food dioxin crisis and the appearance ofBSE, have all contributed toarising demand for high quality, safe, sustainable and traceable products inthe production chains ofmany nations (Xiang etal. 2009). These developments started inthe EU15, but are increasingly important incountries like Poland, Bulgaria and Romania. First, with rising incomes, retailers are making new demands onlocal producers inorder toserve the high-end income consumers (Reardon etal., 2003). Second, the rapid growth of modern retail chains, where multinational companies are the top players inthe market, increased the spread ofhigh standards asthe modern retailing companies have begun to set standards for food quality and safety in the sector wherever they are doing business (Dolan and Humphrey, 2000; Henson etal., 2000). The demand for higher food standards changed the way ofdoing business along the food chain (Kinsey, 2003). Food standards are nolonger only introduced and regulated bythe government, but also byprocessing and retailing companies, who develop their own private corporate standards (Swinnen and Vandemoortele, 2008). Fulponi (2007) illustrates the importance ofprivate corporate standards based onasurvey of16 leading food retailers. About 90 % ofretailers reported that the standards they required for doing business were higher than those set bythe public authorities, and about one-half reported that they were even significantly higher (Figure 9). Based on our interviews, we find that there have been important improvements in the quality offood products such asstricter legislation onchemical use and more stringent cooling and storage conditions. For example, in 2005, less than 50 % of the raw milk delivered to milk processing establishments in Romania was compliant with the current quality requirements (e.g. bacterial count, presence ofantibiotics, ..). Partly, these improvements are introduced under impulse ofEU integration, but we have some indications that especially modern retailers are concerned about food safety and quality standards (see also section 5.1.3.). First, wefind evidence that consumers prefer modern retailer over traditional channels, because the quality ofsome products ishigher inthe modern retailer than inthe traditional shop oron the open market. One ofthe consumers summarized itlike this: Inthe summer, some ofthe traditional shops donot have sufficient cooling facilities, while inthe supermarket itis sometimes socold that Ineed toput onextra clothes. This isthe main reason why Ibuy meat and dairy products inthe supermarket, especially insummer. Second, based on our interviews with producers and their representatives we find that modern retailers include specific quality and safety standards in their contracts with producers and request certification onchemical use, while this isless the case incontracts between producers and traditional wholesalers. Inthis perspective, private labels are increasingly important inthe region. For example, inPoland, Tesco, Carrefour and E.Lelerc had together 5607 private labels in2006, while in2008, this number increased to13.284 labels (Figure 10). Intotal, private label goods accounted for 9 % oftotal food sales in2008. The private label market inPoland recorded growth not only interms ofnumber but also quality ofthe products. Today, besides low and medium price private labels, retailers inPoland offer also premium brands, where the quality and prices are comparable toor higher than those
33
Figure.9:.Retailers.Self.Assessed.Standards.compared.tothose.ofthe.Government
60 50 40 Retailers (%) 30 20 10 0 Food safety Environmental Animal welfare Labour The same as Slightly higher Significantly higher
Source: Fulponi (2007) Figure.10:.Development.ofPrivate.Labels.inthe.Low,.Medium.and.Premium.Price.Brackets.atTesco,. Carrefour.and.E.Leclerc.inPoland,.March.2006.and.March.2008

14000 317 12000 Number of private labels 10000 8000 6000 4000 2000 0 2006 Economy
34
9627
2819
2788
3340
2008 Medium Premium
Source: PMR Publications (Private Label inPoland 2008)
ofleading manufacturers brands, such asCarrefoursBio organic label (PRM Publications, 2008). Currently, the leading private-label players include Tesco (with 700 names infood and chemical sectors), Carrefour (4,300 names, with plans toincrease this byan additional 100 during 2009) and Real, which offers anumber ofitems under its Real Quality label. In Romania, Metro, which sold EUR160mn private label goods in 2009 (or some 10 % of its total sales), reported that itwould like toexpand its portfolio. Metro expects that its private label portfolio which includes brands such asRioba, Fine Foods, H-Line, Sigma and Aro will top EUR300mn by2012. Intotal, the portfolio has over 1.800 products, with the aim ofdoubling this number by2012. Incontrast, the company has noplans toexpand the number ofits stores during the current year. The information obtained from our interviews inthe rural areas isconsistent with these conclusions. Inall interviewed modern retailers the number ofprivate labelled products increased. For example, in the Polish supermarket the number of labelled products rapidly increased, from no products inthe beginning ofthe 2000s toapproximately 50products in2010. Also inthe interviewed discount store inRomania, the number ofprivate labelled products largely increased (by20 % inthe past year, from 500 products in2009 to600 products in 2010). In Bulgaria, which is lagging behind tothe other two countries, the number ofprivate labelled products inone supermarket increased from 20products in2004 to30 products in2010.
35
5. Impacts on Consumers, Producers, Traditional Retail and Rural Society

The previous section already identified several effects ofinvestments bymodern retail companies and discounters. Here weanalyze more specifically the effects onspecific groups insociety.
5.1 Impact on Consumers

Inthe literature different consumer effects are identified (Minten etal., 2010): the impact ofmodern retail on consumer choice of shops, on prices, on quality and variety of offered food products, ondiet trends, and onthe profile ofconsumers. Wewill consider each ofthese effects here. 5.1.1. Impact.onConsumer.Choice.and.Shopping.Behavior In the previous section we have provided extensive documentation of the growth of both the number of modern retail outlets as the diversity of formats in Bulgaria, Romania and Poland. Wehave documented how the growth has been stronger inRomania and Poland than inBulgaria where there are less modern outlets and where discounters were absent until the beginning of2010. It is clear from the evidence presented there, in particular from the aggregate data on where consumers shop that the arrival ofmodern retailers and discounters has major effects onconsumer behavior, inparticular inthat they are increasingly and significantly moving away from shopping intraditional retail outlets toshopping insupermarkets, hypermarkets and discounters where they are available. Our own information based onour interviews inremote and rural areas islargely consistent with the aggregate data from other sources which wepresented before. InBulgaria, Poland and Romania, wefind that the different stages ofretail development are also reflected inconsumer behavior. Itis useful tocompare the beginning of2000s for Poland and the mid-2000s for Bulgaria and Romania with the current situation since inthe beginning ofthe 2000s, Polish retailers started investing in retail outlets in rural regions. In the second wave countries, Bulgaria and Romania, this evolution took place in mid 2000s. By taking these time periods asabaseline, wecan compare how consumer behaviour has changed since the first years that there was amodern retailer inthe region. Initially, market penetration of supermarket sales was relatively low for most of the products. However, the market share ofmodern retail rapidly increased and by2010 the modern retailer was the main shopping place for almost all rural consumers that weinterviewed. Interesting, wefound that also consumers ofthe traditional shop far from the supermarket buy some products atthe modern retailer. Onaverage, consumers oftraditional shops far from the modern retailer gotwo times per month tothe modern retailer, while consumers oftraditional shops close tothe modern retailer and consumers of the modern retailer went respectively five and nine times per month to the modern retailer to buy groceries. Nevertheless, there were important differences between product categories (Table 15A).
36
Rapidly after the establishment of the modern retailer in the rural region, consumers started to buy staple food products (44 % of the consumers indicates the modern retailer as the main shopping place for this product type), processed food (40 %) and wine (46 %) inthe modern retailer. The importance ofthe modern retailer asthe main shopping place for these products continued toincrease and by2010 almost all consumers buy these products inamodern retailer (staple food products: 78 %; Processed food: 71 %; Wine: 77 %). Initially the importance of the modern retailer as the main shopping place for semi-processed food products, such asdairy products and meat, was smaller than for staple and processed food products (38 % for dairy products and 28 % for meat products). However, also for these products, the importance of modern retailers as the main shopping place rapidly increased and by 2010 already 67 % and 55 % ofthe consumers buy respectively dairy and meat products inamodern retailer. The consumers that donot buy meat and dairy products inamodern retailer buy these inthe general shop orat the market (dairy)/ inaspecialized shop (meat) (Table 16). For fruits and vegetables our findings are different. Inthe first years after the emergence ofmodern retailers only 19 % ofthe consumers purchased fruits and vegetables inamodern retailer, but also in2010 the importance ofthe modern retailer asthe main shopping place for fruits and vegetables was relatively limited compared to staple, processed and semi-processed food products (fruits: 37 % and vegetables: 35 %). Most consumers indicate that they still buy fruits and vegetables atthe market orin the traditional general shop (Table 16). These results are also reflected in the monthly expenditures, which differ substantially between traditional shops and modern retailers depending onthe product category. For fruits and vegetables, the average monthly expenditures per household member in a modern retailer are respectively 3,8and 3,5compared to 6,1and 5,8inthe traditional shop. Incontrast, the average monthly expenditures for staple and processed food products are higher inthe modern retailer compared tothe traditional shop (Table 17). 5.1.2. Impact.onConsumer.Prices Inorder tofind out ifpoor consumers could benefit from the emergence ofmodern retail channels, itwill becrucial todetermine whether modern retailers offer their products athigher oralower prices than traditional retailers. Based onareview ofthe literature (see Box 2for alist ofstudies), Minten and Reardon (2008) conclude that price differences between modern and traditional retailers largely depend onthe product type and the level ofmodernization ofthe procurement system. One can distinguish three stages: (1) Atthe very early stage ofpenetration ofmodern retail formats intransition and developing countries, prices offered in modern retail formats are equal or higher (compared to traditional retailers) for processed food products and higher for fresh food produce. (2) In the a more intermediate (but still early) stage, when modern retail channels start their take-off , processed food ischeaper insupermarkets, but the results for fresh produce are mixed (for some mass produced products itis cheaper inmodern retailers while asfor other itis more expensive than intraditional retailers). This isbecause retailers first introduced changes intheir procurement strategies for processed food products (and rapidly afterwards also semi-processed food products such asmeat and dairy). The changes inprocurement strategies substantially reduce
37
Table.15A:.Modern.retailer.astheir.main.shopping.place.(%.ofthe.interviewees) Beginning.ofthe.2000s.(Poland)/. Mid.2000s.(Bulgaria.&.Romania) Fruits Vegetables Dairy Meat Staple food Processed food Wine Other beverages Source: Calculations based onown survey results Table.16:.Main.shopping.place.in2010.(%.ofthe.interviewees) Modern.retailer Fruits Vegetables Dairy Meat Staple food Processed food Wine Other beverages 37 % 35 % 67 % 55 % 78 % 71 % 77 % 64 % Market 32 % 35 % 5% 2% 1% 2% 1% 1% Traditional. general.shop 29 % 28 % 28 % 34 % 21 27 % 20 % 35 % Specialized. shop 2% 2% 0% 9% 0% 0% 2% 0% 19 % 19 % 38 % 28 % 44 % 40 % 46 % 37 % 2010 37 % 35 % 67 % 55 % 78 % 71 % 77 % 64 %
Source: Calculations based onown survey results Table.17:.Monthly.expenditures.per.household.member.(ineuro) Traditional.Shop* Fruits Vegetables Dairy Meat Staple food Processed food Wine Other beverages 6,1 5,8 5,0 7,8 3,3 3,2 2,3 3,6 Modern.retailer 3,8 3,5 6,8 8,3 4,9 4,3 5,1 4,6
* Traditional shops include the general shop, market, specialized shop and neighbors. Source: Calculations based onown survey results
38
the transaction costs and incase ofhigh competition between retailers, the retailers will pass onthe lower transaction costs totheir customers, which will result inlower consumer prices (Reardon etal., 2010). This isillustrated byFarina etal. (2005), who observed the existence of symbiotic relationships between large processors and retailers. For example, inBrazil and Argentina, they observe retailers procuring large volumes of milk, which substantially reduces transaction costs with the processor and reduces milk prices for consumers. Similar relationships are observed byDries and Reardon (2005) inRussia. (3) In a more advanced stage, food prices in modern retail shops tend to be generally lower (compared totraditional retailers) for almost all food products, including fresh produce, because modern retailers also adopted the changes inprocurement strategies for these products. One ofthe motives tooffer also low prices for fresh products isto gain market share byproviding consumers with one-stop shopping convenience and integrating the conventional supermarket and fresh food market into one store. Inour interviews with consumers inremote areas inthe countries, when weasked for the reasons behind the change in consumer shopping choice, we find that price was the most important reason tochange towards more purchases inthe modern retailer and also the representatives ofthe consumer organizations mention price asthe main reason toshop inmodern retailers (Table 18). Table.18:.Main.reason.toincrease.the.share.oftotal.purchase.bought.inthe.modern.retailer.in2010. (%.ofthe.interviewees) Price Fruits Vegetables Dairy Meat Staple food Processed food Wine Other beverages 41 % 36 % 42 % 39 % 63 % 61 % 39 % 63 % Variety 13 % 9% 23 % 16 % 13 % 16 % 41 % 14 % Quality.and.Safety 10 % 19 % 18 % 22 % 2% 4% 2% 3% Convenience 36 % 36 % 17 % 23 % 21 % 18 % 18 % 20 %
Source: Calculations based onown survey results While wedo not have detailed representative data onthis because ofthe absence ofsufficient outlets for comparison in the same rural towns or remote areas one would expect that with regard toreducing prices the effect ofdiscounters would beparticularly important, since they put particular emphasis onthis element intrying toattract customers. For this reason, their investments may beparticularly beneficial for the poorest groups inthose regions who would benefit most from such low priced food commodities, such as the elderly and unemployed assuming that these groups can access the low prices byshopping inthe discount stores, which isnot always the case (see further inthis section). However, there isno guarantee that consumers ofmodern retailers will benefit from lower prices inthe long run, when the sector becomes more concentrated. Ingeneral, economic theory suggests that high concentration will lead tohigher market power and hence higher seller power vis--vis consumers, which istranslated inhigher consumer prices.
39
However, empirical evidence from the developed world israther inconclusive. Onthe one hand, Hall etal. (1979), Lamm (1981), Marion etal. (1993), and various studies byCotterill (Cotterill 1986; 1999; Cotterill and Harper, 1995) find that there is a positive correlation between retailer concentration and food prices. Onthe other hand, Kaufman and Handy (1989), Newmark (1990) and Binkley and Connor (1998) find anegative orinsignificant correlation between concentration and food prices. Likewise, Binkley et al. (2002) find little compelling evidence that consolidated markets engage innon-competitive pricing behaviour. Cooper (2003) indicates that based onthe findings from the Competition Commissionsreport onthe UK, prices inmodern retailers did not vary with the number oflocal competitors, but were often lower inthe proximity ofcertain lower priced large stores, asmodern retailers often benchmarked the prices ofcertain key products with the prices oftheir main competitors. Inaddition, the presence ofdifferent formats can play arole. For example, Dobson etal. (2001) find that while retail concentration inGermany isquite high, there isstrong competition onthe selling side, with asmain engine the hard discounters such asAldi and Lidl. Finally, there isalso evidence that modern retailers are giving with one hand and taking with the other, meaning that they charge higher prices for some products, while attracting clients with low prices for other products. For example, in the UK modern retailers Tesco and Asda dramatically increased the prices ofsome key items inthe run-upto Christmas, such astoys, light bulbs and batteries. Both companies ran marketing campaigns before Christmas boasting price cuts but many consumers will have been unaware that they were also raising prices ofother products inthe same period.11 Hence, inthe short run, the emergence ofmodern retailers and discounters inparticularly is expected to lead to reduction in consumer prices, although the effect depends on the stage ofmodernization ofthe supply chain and the product type. Inthe long run, the effect ofan increase inconcentration and selling power vis--vis consumers onprices isuncertain asthe existing empirical evidence indeveloped countries onthe impact ofincreased concentration onprice isinconclusive.
Box 2: Case stuDies onthe impaCt ofthe retail seCtor onConsumer priCes Ho (2005) finds that inHong.Kong (afirst wave country) the first modern retailers set higher prices than traditional retailers inthe beginning ofthe 1970s. However atthat time, these retail outlets only served the needs ofaselected segment ofthe local population who were familiar with the Western lifestyles and also demanded high quality products. When atthe end of the 1970s, modern retailers became more important, they started decreasing the price ofrice, the most important staple food inHong Kong, such that their rice was cheaper than in traditional shops. In the 1980s and 1990s, they decreased the prices of processed commodities below prices asked bytraditional retailers, whereas prices offresh produce were still substantially higher inmodern retail outlets compared towet markets. Inthe 1990s-2000s, modern retailers started to decrease the prices of fresh produce in order to gain market share byproviding their customers one stop shopping, which became more attractive with female participation inthe labour force. This strategy seemed tobe very successful and the market share ofsupermarkets inthe sales offresh produce increased from almost 0 % inthe 1980s toapproximately 50 % in2005.
11.- Based onarticle inthe Guardian, Systematic, cynical, aggressive: expert verdict onTesco and Asda prices, 12February 2010.
40
InSouth.Africa (afirst wave country), DHaese and Van Huylenbroeck (2005) surveyed poor rural consumers about their food purchases inrural areas and nearby cities. They show that most households buy most oftheir food (except meat) insupermarkets, where processed food and staple food, that account for the top ten ofconsumer items, are offered atlower prices than intraditional outlets. However also prices for some bulk fresh products, such ascabbage, and semi-processed items, such asdairy, are substantially lower insupermarkets compared totraditional outlets. For example, the main staple food product maize meal is17 % cheaper insupermarkets; rice 32 %; milk 32 %; beans 89 % and cabbage 44 %. InEquador (asecond wave country), Zamora (2005) finds that private label, high quality, washed potatoes in supermarkets are 10 % cheaper than bulk, unwashed potatoes intraditional outlets. InMexico (asecond wave country), Martinez (2006) examines retail data for 2003 and 2005. She compares the 2003 prices of22 leading food (processed and fresh) products insupermarkets, traditional shops and wet markets. The results show that 3products (out of22; or14 %) are the cheapest intraditional shops and 7products (or32 %) are the cheapest inwet markets, while 12 products (or 56 %) are the cheapest in supermarkets. In the same study, she also compares the 2005 prices of17 types offruit and tomatoes and finds that supermarkets offer the lowest prices for 10 (out of17) types offruit. She notes that the 2003/2005 results differ sharply from data from the mid-1990s when supermarkets had higher prices than traditional retailers except for the main processed food products. InThailand.(asecond wave country), the Thailand Development Research Institute (2002) shows that onaverage hypermarkets sell their processed food products 12 % cheaper than in traditional shops, while fresh food products are 10 % more expensive in hypermarkets compared totraditional shops. In Kenya (a third wave country), Neven et al. 2006 compare prices in supermarkets and traditional shops. They find that onaverage supermarkets are 34 % cheaper than traditional shops for processed food items, which represents the majority ofconsumer food expenditures. After controlling for differences inquality, the price for fresh products are ingeneral higher insupermarkets compared totraditional shops, except for some mass produce items, such askale, that supermarkets use toattract consumers. InBotswana and Zambia (both third wave countries), Emongor (2007) presents retail price survey data which indicate that supermarkets offer the lowest prices for processed food products. For example, for maize flour, the price inBotswana and Zambia was respectively 54 % and 22 % lower insupermarkets than intraditional shops. Fresh products were more expensive in supermarkets compared to traditional shops in Zambia, which is consistent with the stage ofdevelopment ofthe retail sector. InVietnam (athird wave country), Moustier etal. (2006) shows that processed food products, such aspork, are ingeneral cheaper insupermarkets compared totraditional shops, whereas fresh products are more expensive. For example, tomatoes are 1020 % more expensive insupermarkets and spinach is3050 % more expensive insupermarkets.
41
In Madagascar (a third wave country), Minten and Reardon (2008) find based on survey evidence that, after controlling for quality differences, prices insupermarkets are significantly higher than traditional shops. Inthe case ofrice, prices are 74 % higher than onthe wholesale markets and about 68 % more expensive than other retail outlets. Tomatoes are 120 % more expensive than onthe wholesale markets and between 72 % and 84 % more expensive than inother retail outlets.
5.1.3. Impact.onProduct.Quality Besides differences in prices between modern retailers and traditional markets, there are also differences inthe quality and variety offered inmodern retailers compared totraditional markets. Similar asin case ofprices, the quality and variety offered inmodern retailers depends onthe level ofdevelopment ofthe retail sector. Initially orin poorer countries, modern retailers tend tofocus onalimited range ofhigh quality packaged and processed food products. For example, inHong Kong, Ho (2006) finds that modern retailers focused on a niche market of expats and upper income consumers that demand high quality products. However, over time they tend toshift towards avariety ofproducts (also semi processed and fresh products) toadd the mass-market totheir marketing strategies. Reardon etal. 2010 indicate that ingeneral the quality offood products sold inmodern retailers is higher than in traditional shops. In some cases this difference in quality is reflected in the price ofthe products and modern retailers focus onaniche market ofupper and middle income consumers that are willing topay for high quality products. For example, inIndonesia, Natawidjaja etal. (2007) show that tomatoes insupermarkets have asuperior quality and price than those sold ontraditional markets. However, insome cases modern retailers offer higher quality products for alower price than traditional shops, e.g. Zamora (2005) inEquador. Our own consumer surveys inthe remote areas ofBulgaria, Romania and Poland confirm this. All interviewed stakeholders mention besides price also quality as an important reason tobuy certain products inthe modern retailer. Especially for meat and dairy products, quality and food safety concerns are animportant reason for consumers tostart purchasing these products inmodern retail outlets (Table 18) (see also section 4.1.6). Also the farmers, that we interviewed, confirm that quality requirements are more important in their contracts with modern retailers compared to traditional wholesalers or the processing industry (Table 19). While only in49 % ofthe cases the traditional wholesaler requires the farmer to keep a diary on chemical use, this is required in 62 % of the cases when the farmer delivers toamodern retailer. Similar, wefind that in38 % ofthe cases the modern retailer requires toprovide information onthe harvest date and specific parcel, while this only required in11 % ofthe contracts between farmers and traditional wholesalers. However, several studies have already argument that the superior quality of products from modern retailers isdebatable and that itis important tomake adistinction between true quality
42
Table.19:.Quality.requirements.requested.bymodern.retailers Modern. Retailer Requirement tokeep adiary onchemical use Restrictions in (legal) chemical use Indication ofthe harvest date and parcel Source: Calculations based onown survey results (based onhealth and safety standards) and esthetic quality (based onstandards for size, shape and colour). Modern retailers have been criticized totake mainly the latter inaccount toreject alarge percentage ofproducts that are produced according tothe health and safety standards, but donot fulfill the modern retailers esthetic requirements. This isapractice which isquestionable from sustainability/ food security point ofview. Based on our interviews, the evidence of such practices is mixed. In Bulgaria, for example, the representative of the producer organization indicated that important benefit contracting with amodern retailer, isthat they buy the entire production, including the products which are classified assecond class ornon-standardized. The first class products are sold inthe fruit and vegetable department ofthe modern retailer, while the second class and unstandardized products are used inthe prepared dishes. Also the managers ofthe modern retail stores that weinterviewed, indicate that they never reject deliveries because of esthetic quality standards.12 However, 24 % of the producers mention that the modern retailer towhom they delivered in2010 rejected atleast once adelivery because esthetic characteristics while their products were fresh and fulfilled all quality standards. This was never the case for deliveries toatraditional wholesaler. 5.1.4. Impact.onProduct.Variety Interms ofvariety ofproducts several studies find that modern retailers offer more variety (e.g. Reardon etal. 2007a; Dries and Reardon, 2005). However, the literature mentions that there are large differences between the different product categories. For example, for staple and processed commodities, modern retailers offer the same brands than the traditional outlets, but modern retailers offer more private brands (see rapid growth ofprivate brands inthe retail sector, which isdiscussed insection 4.1.6), more imported brands and more health-related brands, such asorganic products. This isalso confirmed byour interviews where modern retailers indicate that they offer such labeled products, while those are not offered intraditional shops. Moreover, the modern retailers indicated that private labeled products have largely increased inimportance over the past years and they expect that this trend will continue inthe future. For dairy products, there isalarge difference between the offerings ofmodern retailers compared totraditional retailers. Modern retailers have economies ofscale, much more self-space and better refrigeration facilities than small shops, which allow them tooffer more dairy product diversity. In addition, especially for their dairy activities, modern retailers benefit from their symbiotic
12.- Nevertheless, all managers, except two managers, indicate that they sometimes reject deliveries (from very rare toeach week), but because the products were not fresh.
Traditional. Wholesaler 49 % 57 % 11 %
62 % 64 % 38 %
43
relationship with big dairy processors, such as Nestl and Danone (For example, see Hu et al., 2004 for China,; Farina etal., 2005 for Latin America and Dries and Reardon, 2005 for Russia). Our interviews confirm that this isalso the case inremote areas ofBulgaria, Romania and Poland: in a modern retailer there were on average 120 different dairy products, while only on average 11different dairy products inatraditional shop. Interms ofvariety offresh products, Reardon etal. (2010) observe that there isamore diverse offer offruit and vegetables inmodern retailers compared tofruit and vegetable shops, but amodern retailer would be similar in diversity (or less) compared to a large wet market. However it is important tonote that there isatime dimension inthe diversity ofofferings offruit and vegetables. In general, traditional markets offer a highly diversity of fresh produce, with sharp seasonality. When modern retailers are intheir early development stage, they stock less local diversity than local markets, but onthe other hand, they offer more imported fresh food products (see Reardon etal. 2007b for Mexico). When modern retailers develop further, the fresh products section becomes a battle arm and consumers turn tomodern retailers rather than local markets because ofthe large diversity oflocal and exotic products. InBulgaria, Romania and Poland, the interviewed stakeholders indicate that inall shops (modern retailers and traditional shops) the variety inproducts substantially increased and they relate this toacombination of Increased purchasing power: Inthe period 2000 2010 all three countries experienced alarge increase inGDP per capita, which isreflected inincreased consumer demands. Privatization of the processing industry: In the period after transition, the processing sector was privatized, which led tothe emergence ofseveral new companies and increased competition inone product category, whereas before there was only one processing company responsible for the production ofone product category. Competition: Both traditional shopkeepers and supermarkets mention increased competition in the retail sector as an important reason to increase the variety offered in the shop in order toattract more clients and gain market share. Onaverage, the number ofdifferent products inthe modern retailers increased by53 % compared tothe beginning of2000s inPoland and the mid-2000s inBulgaria and Romania. (Table 20). Also the variety ofproducts intraditional shops increased substantially. Inthe interviewed traditional shops close tothe supermarket there were onaverage 368 products inthe beginning/mid 2000s, while in 2010 this number increased to on average 415 products. Similar, we find that also the number ofproducts intraditional shops far from the supermarket increased (from onaverage 223 products to228 products in2010). This indicates that there might beimportant spillover effects of the changes in the retail sector on the traditional shopkeepers. This is also reflected in the fact that the traditional shops close to the supermarket (and hence those most under influence ofcompetition from the supermarket) experienced the largest increase inthe assortment. However, the differences between traditional shops close and far from the supermarket are not statistically significant (but this could berelated tothe small sample size).
44
Table.20:.Variety.insupply.inmodern.retailers.and.traditional.shops.(number.ofproducts) Beginning.of2000s.(Poland)/. Mid.2000s.(Bulgaria.&Poland) Modern retailer Traditional shops close tothe supermarket Traditional shops far from the supermarkets 3.500 368 223 2010 5.350 415 228 Change.(%) 53 % 13 % 2%
Source: Own calculations based onstakeholders interviews 5.1.5. Impact.onDietary.Trends One ofthe concerns related tothe rapid emergence ofmodern retailers isthat itworsens diet trends bystimulating overconsumption ofprocessed food and abandonment offresh food. In early stages, modern retailers indeed focus on packaged and processed food products, such asflour, bread, rice, vegetable oil and condiments. However, when modern retailers and regions develop, there isgrowth insales insemi-processed products, such asmeat, dairy and fish (see also section 5.1.1). More recently, the market share ofthe fresh produce section rapidly increased. For example, inHong Kong the market share offruit and vegetables increased from approximately 0 % inthe 1980s toapproximately 50 % in2005 (Ho, 2006). InMexico, fresh products represented 12 % offood sales inthe beginning ofthe 2000s, while currently this isapproximately 11 % (Reardon etal. 2007b). However, these global characteristics donot tell usmuch about individual consumption patterns of supermarket customers. There are a few studies that have analyzed the relationship between shopping inmodern retailers and individual diet composition effects. Asfaw (2008) finds for Guatemala that consumers that do their shopping at supermarkets tend to consume processed foods at the expense of staple foods. A 1 % increase in the value of supermarket purchases increases the shares of pastries (including cookies and crackers) and other highly processed foods (sweets, chocolate, sausages, ice creams, etc.) in the total calorie availability by14,4 % and 6,0 %, respectively. The value ofmodern retail purchase does not have animpact onthe share ofvegetables and fruits. But itdoes have anegative impact onthe share ofstaple foods such ascorn and pulses. The results also reveal that the value of modern retail purchase is positively associated with the individual Body Mass Index (BMI), indicating that consumers that dotheir shopping atmodern retailers tend tohave ahigher BMI and consequently are more likely tosuffer from obesity and health concerns associated with it. In Tunesia, Tessier et al. (2007) find that the emergence of modern retailers has a positive (but small) effect ondiets. Modern retailers offer some new food products (such asavocados, asparagus, kiwis, mangoes, salmon, basmati rice, low-fat products, etc.) that cannot befound elsewhere and
45
consumers have shifted from alimited number ofstaple products toamore diverse diet. Hence, they find a slight improvement of diet quality among those people who regularly visit modern retailers. Inthe past 20years, wefind significant changes inthe dietary patterns ofthe inhabitants ofthe three case study countries. However, weshould note upfront that itis unclear towhat extent wecan relate these changes tothe emergence ofmodern retailers asthis evolution coincides with the transition process and EUintegration process. There are substantial differences between countries. In Bulgaria, we find a continuous decrease infood consumption. In1989, individuals consumed onaverage per day 3.622 kcal, while in1995 this was on2.899 kcal and in2007, 2.766 kcal per day (Table 21). Interms offat consumption, wefind adecline inconsumption inthe years after transition and, more recently, anincrease infat consumption (Table 22). InPoland, wefind adecrease infood and fat consumption inthe years after transition, but more recently food and fat consumption isrecovering and almost atthe pretransition level (Table 21and 22). These evolutions are very different from the evolution offood and fat consumption in Romania, where food and fat consumption has continuously increased over time (Table 21and 22). There is also a shift in the types of fat that are consumed (Figures 11A, 11B and 11C). In all three countries, the intake ofvegetables fat (unsaturated, healthy fat) increased atthe expenses ofanimal fat (saturated, unhealthy fat). However, onthe other hand there isan increase inthe consumption ofprocessed food products, such asfor example pre-cooked meals, sweets, confectionery, sauces, condiments, InBulgaria, for example, sales ofconfectionery increased by3 % per year inthe period 20042009, while inPoland and Romania, this was respectively 5 % and 8 % per year (Table 2325). Also for other processed food products sales increased, although more recently sales ofprocessed food seems tostabilize due toadecreasing purchasing power asaresult ofthe global economic and financial crisis. This isespecially the case inBulgaria, the poorest country ofthe three. Most ofthe consumers who Table.21:.Average.consumption.inthe.case.study.countries.(kcal/capita/day) 1989 Bulgaria Poland Romania Source: FAOstat Online Database Table.22:.Average.fat.consumption.inthe.case.study.countries.(grams/capita/day) 1989 Bulgaria Poland Romania Source: FAOstat Online Database
46
1995 2899 3306 3065
2001 2773 3410 3276
2007 2766 3421 3455
3622 3490 2938
1995 122 119 84 92 110 87
2001 93 113 100
2007 96 114 108
Figure.11A:.Share.ofdifferent.food.categories.inthe.total.fat.consumption.inBulgaria.(%)
100% 80% 60% 40% 20% 0% 1990-92 1995-97 2003-05 4 18 19 14 18 5 9 22 4 8 15 16 Cereals - Excluding Beer Animal Fats Meat Milk - Excluding Butter Vegetable Oils
34
37
50
Source: FAO Figure.11B:.Share.ofdifferent.food.categories.inthe.total.fat.consumption.inPoland.(%)

100% 80% 60% 26 40% 17 20% 19 0% 1990-92 1995-97 2003-05 30 29 15 13 4 29 4 21 4 21
24
25
Cereals - Excluding Beer Animal Fats Meat Milk - Excluding Butter Vegetable Oils
Source: FAO
47
Figure.11C:.Share.ofdifferent.food.categories.inthe.total.fat.consumption.inRomania.(%)
100% 80% 60% 40% 20% 0% 1990-92 1995-97 2003-05 5 19 21 17 7 12 19 24 24 Cereals - Excluding Beer Animal Fats Meat Milk - Excluding Butter Vegetable Oils 5 7 17
31
30
37
Source: FAO stated that they reduced consumption of (processed) food products were either older people orthose living inrural areas (Euromonitor, 2010).13 Ingeneral, processed food products contain large amounts oftrans fat, which isthe common name for unsaturated fat with trans-isomer fatty acid. The excessive consumption oftrans fat can lead toserious health risks, such asfor example obesity and coronary heart disease. Astudy byStender etal. (2006) onthe occurrence oftrans fat inprocessed food products inseveral European countries showed that processed foods in all three countries contained high levels of trans fat compared toother European countries. Inorder toencourage healthier eating habits, the Romanian minister ofhealth proposed inthe beginning of2010 a fat tax onthe consumption ofunhealthy food products, such assweets, soft drinks and fast food.14 However critics ofthe Romanian tax indicate that the tax will beineffective as it only targets certain food products and for example kebabs one of Romanias favorite foods and pizza will beexempt from the levy. Nevertheless, weneed toemphasize that wehave noevidence that any ofthese evolutions infood consumption patterns are related tothe emergence ofmodern retailers and more profound research isneeded toseparate out this effect from the changes that the transition process and, more recently, economic development asresult ofEU integration, caused.
13.- The decrease insales of (processed) food products can beexplained byadecrease inpurchasing power, including due toadecline inincome from remittances ofrural migrants working abroad. 14.- Also other European countries have introduced measures toencourage consumption ofhealthy products: Denmark and Austria intoduced aban onthe use oftrans fat infood products; Norway has atax onsugar and chocolate; Norway, Sweden and the UKhave aban oncommercials for fast food atcertain times ofthe day (Holt, 2010).
48
Table.23:.Sales.ofprocessed.food.products.inBulgaria.(000.tonnes;.20042009) 2004 Confectionery Sweet and savoury snacks Ready meals Sauces and condiments Processed food (frozen, dried and chilled) Source: Euromonitor (2010) Table.24:.Sales.ofprocessed.food.products.inPoland.(tonnes;.20042009) 2004 Confectionery Sweet and savoury snacks Ready meals Sauces and condiments Processed food (frozen, dried and chilled) Source: Euromonitor (2010) Table.25:.Sales.ofprocessed.food.products.inRomania.(tonnes;.20042009) 2004 Confectionery Sweet and savoury snacks Ready meals Sauces and condiments Processed food (frozen, dried and chilled) Source: Euromonitor (2010) 5.1.6. Type.ofConsumers An important issue is the profile of the consumers shopping at modern retail chains. Different studies have analyzed the socio-economic characteristics ofconsumers shopping inmodern retailers compared to traditional shops (e.g. Pingali, 2007; Gorton et al. 2009; Rodriguez et al., 2002). Ingeneral, they find that consumers ofmodern retailers are living inurban areas, are younger and better educated. The classic hypotheses ofGoldman (1974) isthat indeveloping countries mainly the rich urban consumers, who are able todrive tothe store, buy big units and store the products, will benefit from the emergence ofmodern retail chains.
49
2005 25,5 5,4 2,2 18,9 170,9
2006 26,4 5,5 2,3 19,5 176,6
2007 27,2 5,7 2,5 19,0 170,6
2008 28,0 5,5 2,6 18,8 165,7
2009 28,6 5,5 2,7 18,6 163,3
25,1 5,3 2,0 18,3 169,4
2005 168,1 101,7 53,8 244,2 407,6
2006 175,2 105,8 58,1 255,8 424,5
2007 184,3 107,8 61,3 262,2 437,5
2008 196,7 110,3 64,9 269,4 451,1
2009 203,4 113,0 68,9 276,4 467,1
160,1 97,7 49,4 232,5 390,1
2005 38,7 17,7 2,2 11,6 288,9
2006 41,2 21,5 2,2 12,1 302,9
2007 45,2 25,1 2,3 12,8 322,6
2008 48,0 28,1 2,3 13,4 338,6
2009 49,2 30,6 2,3 13,9 353,4
35,2 16,3 2,1 11,0 269,3
However, Goldmans statement was based on anecdotal evidence. Currently, there is more empirical evidence and the retail sector itself has changed. Recent evidence shows that in the early stages ofsupermarket penetration, itis mainly the upper and middle class that tends toshop insupermarkets. For example, inVietnam, acountry inthe early stage ofsupermarket penetration, Figui and Moustier (2009) find that poor consumers purchase very little from supermarkets due tomaterial constraints (price, transport, etc.). Incontrast, inKenya, Neven etal. (2006) show that the poorest households are not excluded from the potential benefits that supermarkets may offer asthe majority ofNairobispoorest households (60 %) visit supermarkets (compared tomore than 99 % for the richest households). However, their data show that the results depend onthe product category. For example, only less than 1 % of the households with the lowest incomes buy fresh produce from supermarkets, while 40 % ofthe richest households buy fruits and vegetables inthe supermarket. InBulgaria, Romania and Poland, wedid not have reliable information onthe income ofconsumers, but nevertheless we found some evidence of differences between consumers going to modern retailers and consumers visiting traditional shops close to the supermarket. However, we note upfront that these results should beinterpreted with caution aswe did not found the differences tobe statistically significant, which could berelated tothe small sample size. Ingeneral, wefind nosignificant difference inage, gender, education orhousehold size between consumers visiting modern retailers and those visiting the traditional shop close tomodern retailer (Table 26). The only factor that significant differed between consumers ofmodern retailers and consumers oftraditional shops close tothe modern retailer isthe distance between their home and the shop (see also section 5.1.7) Table.26:.Differences.between.consumers.ofmodern.retailers.and.traditional.shops.close.tothe.modern. retailers Consumers. modern.retailers Age (years) Education level (1= Primary;
2=Secondary; 3= Tertiary)
Consumers.traditional.shop. close.tomodern.retailer 49,74 2,54
T-test. (P-value) 0,19 0,73
45,95 2,59
Gender
(0= Female; 1= Male)
0,52 3,10 1,72
0,40 3,12 0,75
0,17 0,95 0,00
Household size Distance tohouse
Source: Own calculations based onstakeholders interviews
50
5.1.7. The.impact.ofRural.Infrastructure The most important factor that significantly differs between consumers of modern retailers and traditional shops isthe distance tothe shop. Ingeneral, consumers ofmodern retailers live onaverage 1,74km from the shop, while consumers oftraditional shops live onaverage onadistance of0,54km from the shop (Table 26). Proximity ofthe shop isalso one ofthe most mentioned reasons tobuy inthe traditional shop, especially for older individuals. Interestingly, consumers interviewed inatraditional shops located far from the modern retailer (between 15-30km) also indicate that they buy atleast some oftheir groceries (mainly processed, staple food and beverages) inamodern retailer. Onaverage they go2times per month toamodern retailer compared to 4 times per month for consumers interviewed in a traditional shop close toamodern retailer and 8times per month for consumers interviewed inamodern retailer. Approximately one third of the consumers in traditional shops far from the modern retailer indicate that they would like tochange their consumer behaviour and buy more inmodern retailers. The most important constraint for changing their behaviour isthat they donot have acar and that public transport isinsufficient.15 Also the consumer organization representatives mentioned limited car ownership and poor public transport as important constraints for rural consumers. Moreover, inRomania the representative indicates that even ifrural consumers have acar, itis difficult togo totown, where the modern retailer islocated, asthe condition ofthe roads isin general very poor. The results ofour analysis insection two indicate that especially the poorer regions with alarger share ofolder inhabitants have, onaverage, fewer cars per hundred inhabitants. InPoland, these are the regions close to Belarus and Ukraine; in Romania, the regions close to Moldova and inBulgaria, the regions close toRomania. These results are anindication that especially older and poorer households are likely tobe excluded from the benefits (lower prices, more variety ...) that modern retailers offer. 5.1.8. Impact.onRural.Services Traditional retail shops offer some services which modern retailers and discounters typically donot offer. One important service for poor people inremote areas isthat traditional shopkeepers often provide consumer credit. Several consumers mentioned that What Ilike about the traditional shop isthat they know meand when Idonthave sufficient money, they allow meto pay after Ireceived mysalary. Traditional shops sometimes also provide other services such asdelivering the products athome, which appears to be especially important for the older people which are less mobile. Given the mobility constraints and the demographic characteristics (ageing) ofthe rural population this may bean important constraint inthe most remote areas.
15.- Incase they buy some oftheir groceries atthe modern retailer, 86 % ofthe consumers ofthe traditional shop far from the modern retailer use their own car tothe modern retailer, while 14 % use public transport. Note that for all consumers, wefind that 62 % ofthe consumers use their car to go to the modern retailer, while 20 % go on foot and the remaining 18 % of the consumers use public transport. 30 % ofthose that use their own car orgo onfoot indicate that they would like touse public transport ifthere was amore frequent connection between their house and the modern retailer.
51
Modern retailers, onthe other hand, offer their consumers inmost cases aloyality card. When aconsumer visits the store, he/she receives anumber ofpoints, depending onthe amount ofmoney that he/she spends during the visit. After accumulating aspecific amount ofpoints, the consumer can exchange these points for gifts or discounts. However, from our interviews we learnt that usually consumers only receive avery limited number ofpoints per leva, lei orzloty, that they spend inthe store. Another aspect that consumers appreciate about modern retailers, isthat they receive brochures, which allow them tocompare prices and anticipate when there are special discounts (e.g. buy seven units and get one for free).
Box 3: impaCt onConsumers16 Opportunities The emergence ofmodern retail inrural regions has major effects onconsumer.behavior as consumers moved increasingly and significantly away from traditional shops and open markets to supermarkets, hypermarkets and discounters. However, there are substantial differences between product categories: consumers buy especially processed and semi processed products inmodern retailers, while they still purchase fruits and vegetables through the traditional channels. Interestingly, wefound that also consumers living relatively far from the modern retailer goon average two times per month todo their grocery shopping inthe modern retailer, indicating that they are not exante excluded from the potential benefits the modern retailer can offer. However, there are also consumers living far from the modern retailer that would like tobuy more inthe modern retailer (see also threats). Prices are lower inthe modern retailer although this depends onthe product type and the level ofmodernization ofthe procurement system. InBulgaria, Romania and inPoland, prices appear tobe animportant factor inconsumer behavior for all product types, indicating that prices are lower inmodern retailers compared totraditional outlets. Aprice reduction will benefit all types ofconsumers, but lower income consumers will proportionally benefit more asthey usually spend alarge proportion oftheir income onfood consumption. All interviewed stakeholders mention quality as an important reason to buy certain products inthe modern retailer. Especially for meat and dairy products, quality and food safety concerns are important reasons for consumers to start purchasing these products inmodern retail outlets, but also for fruits and vegetables, producers indicate that the quality requirements that modern retailers impose are more stringent than those ofother trading partners (e.g. traditional wholesalers). Modern retailers offer a larger variety of products, but all stakeholders mention that also traditional shops increased their product offer. This can partly explained byincreased competition with modern retailers, but also because increased purchasing power and more choice offered byprocessing companies.
16.- Weopted not todistinguish between the effects onmiddle and low class consumers asmost opportunities and threats are similar for both groups. However, when one ofthe issues disproportionally affects one the groups wewill clearly indicate this.
52
Over the past years, consumption patterns changed and in all countries, the intake of vegetables fat (unsaturated, healthy fat) increased at the expenses of animal fat (saturated, unhealthy fat). However, wewant emphasize that wedo not have any evidence towhat extent wecan relate this change tothe emergence ofmodern retailers asthis evolution coincides with the transition process and EUintegration process. Consumers indicate that modern retailers offer less services than traditional shops (see threats), but nevertheless they indicate that they use the services that the modern retailer offers. For example, they use the loyalty card although they mention that the discounts are very small. Inaddition, they also mention the importance of (free) brochures, which allows them tocompare prices and beaware ofspecial discounts. This service may benefit especially lower income consumers. Modern retailers are typically more costly in terms of transport to visit than small traditional shops in the consumers neighborhood (more far away from the consumers home), but modern retail stores also tend tohave afar wider assortment ofproducts than asmall shop, and thus one trip toalarge store would beequivalent tomany trips toavariety of small shops. This is also reflected in the importance of convenience as a reason to buy specific products inthe modern retailer. Threats Inthe short run, prices inmodern retailers appear tobe lower compared totraditional shops, but this isno guarantee that also inthe long run, consumers will benefit from lower prices, when the sector becomes concentrated and modern retailers can exercise more seller power vis--vis their consumers. Nevertheless, itis important tomention that the existing empirical evidence from developed countries on the impact of increased competition onprices isinconclusive. Several respondents indicate that they would like tochange their consumer.behavior and go more to the modern retailer, but that they are constrained in doing so as they do not have acar and public transport facilities are insufficient. Inaddition, they mention the poor state ofthe roads asan important constraint for going tothe modern retailer. Asaresult, especially old and poor (without car) rural households are excluded from the potential benefits (e.g.lower price, more variety, convenience, ) that modern retailers offer. Several consumers indicate that the services that traditional shopkeepers offer (e.g. credit provision, home delivery of groceries) are important to them and they fear that when modern retailers will become more important that these services will disappear. Currently, especially old (home delivery) and poor (credit) households benefit from these services. Inthe past years, the consumption.ofprocessed.food.products strongly increased. Ingeneral, these products contain ahigh proportion oftrans fat, which has adisproportionally large negative impact onhealth. However, wewant emphasize that wedo not have any evidence towhat extent wecan relate this change tothe emergence ofmodern retailers asthis evolution coincides with the transition process and EUintegration process.
53
Recommendations tomaximize positive and mitigate adverse social impacts onconsumers Improvement.ofrural.infrastructure.and.public.transport.facilities will becrucial tomake sure that all rural inhabitants can benefit from the changes inthe retail sector. Inaddition, investment in the rural infrastructure will make rural regions more attractive for modern retail investment. This investment should come from the government, but inthe meanwhile, modern retailers and local governments could engage in private-public partnerships toprovide some services and initiatives that increases access tomodern retailers for the rural consumers that are currently excluded from the potential benefits that modern retailers may offer. For example, they can provide abus service connecting the village tothe shop. Food consumption isan individual choice and hence modern retailers are not responsible for their customers. However, the modern retailer may.promote.the.consumption.ofhealthy. products, for example byincreasing the visibility ofhealthy products orby introducing certain food and safety standards inthe production oftheir private-label products. Inaddition, they could also provide more information on healthy food products and potential risks of an unhealthy diet toincrease consumer awareness. However, not only the modern retailer can provide information, also the government can play animportant role inincreasing consumer awareness onthe risks ofan unhealthy diet, for example, byfinancing TVspots that promote the consumption offruits and vegetables. Currently, producers indicate that modern retailers sometimes reject their deliveries which are produced according to the health and safety standards, but do not fulfill the modern retailers esthetic requirements. Modern retailers should be stimulated to find a good. balance.between.true.quality.and.esthetic.quality.requirements. They could engage tobuy a part of the production that does not fit their esthetic requirements (but for example at a lower price). These products could be sold with a discount or could be used in the preparation ofpre-cooked dishes.
5.2 Impact on Producers

The introduction of modern retail chains induced major changes in the product procurement systems and modernized procurement systems, emerged. This emergence iscrucial tostudy the implications of modern supply chains for local producers. If procurement systems change, the restructuring ofthe retail sector translates into atransformation ofthe market that farmers face. Procurement system modernization typically includes important elements: the introduction of private standards for quality and safety; concentration of the buyers and associated changes inbargaining power; the use ofdistribution centers for supply sourcing; and ashift from spot markets relations intraditional wholesale markets tosome form ofvertical coordination mechanisms inthe relationship with suppliers. These changes may have avariety ofeffects onproducers: how many producers can supply, under which conditions, what type ofproducers can supply, etc. all ofwhich ultimately affects employment
54
and incomes of farmers in the region. An important issue, besides the overall employment and income effects isthe impact ofthe growth ofmodern retailers onsmall producers. Many reports have emphasized that especially small and low skilled producers will not be able to address the quality standards imposed by modern food chains, and will be disproportionately negatively affected. Wewill discuss this below. 5.2.1. Changes.inthe.Supply.Chain Until twenty years ago, vertical coordination of the supply chains was widespread in transition countries asall stages inthe food production and distribution were coordinated and determined bythe central command system (Rozelle and Swinnen, 2004). Inthe 1990s, privatization and market liberalization led tothe emergence ofprivate wholesalers that replaced the government-controlled retailers. For example, by1999, there were inPoland five nationwide wholesale networks (exclusively owned either by foreign capital or by partnerships of foreign and domestic companies), five purchasing groups, 60 regional wholesale networks (holding a30 % share ofthe wholesale market), and 14.000 local wholesale company warehouses (mainly family businesses). These new private wholesalers collected their products from farms and processing companies and they delivered the products tothe remaining state-owned chains, the new private chains, and even tothe green markets and small shops. However, the growth of modern retail formats, facilitated by substantial FDI, and increased competition in the sector created the need to reduce costs, to deliver consistent volumes ofstandardized products and toincrease product quality and variety. This led tolarge changes for the wholesale sector and the farmers delivering tothe wholesales sector. Changes.inthe.relationship.wholesaler.retailer There were substantial changes for the wholesales sector as retailers started to work with centralized procurement systems that were delivered byspecialized wholesalers. Inorder toreduce the transaction costs, generate economies ofscale, work with fewer wholesalers and have atight control onquality ofthe product, modern retailers prefer towork with asystem where there isone centralized buying office for one product category (e.g. meat) and one orseveral distribution centers over the country tosupply the local stores17. The distribution centers are delivered byawholesaler, which isspecialized inone product category and dedicated tosupplying modern retailers. That means that the wholesaler ismore responsive toquality, safety, and consistency requirements ofmodern retailers than are traditional wholesalers who aggregate products over many producers and qualities and usually have arange ofproducts that they deliver18. Typically achain moves from adistribution center for azone toadistribution center orset oflinked distribution centers for acountry inorder tosource over the country and get the cheapest and best quality products byhaving alarger supplier pool from which tochoose. This evolution towards
17.- There isalso coordination ofthe distribution centers across borders. For example, in2003, Ahold created Ahold Central Europe (ACE) tocoordinate its operations inPoland, Czech Republic, and Slovakia. ACE isbased inthe Czech Republic, and merges backroom functions such asproduct procurement and administration for the Ahold retail chains inCentral Europe (www.ahold.com). 18.- Inmany cases the retail company retail chain acquires orenters inajoint venture with the wholesale firm. Acquisition has the advantage ofcontrol, exclusivity sothat italso captures asupplier base, and making the wholesaler aprofit center (Dries etal. 2004).
55
centralized distribution centers started inthe first wave countries and iscurrently also happening inthe second and third wave countries. Several cases illustrate these developments. For example, in2009, Tesco has 319 stores inPoland which are supplied bythree distribution centers and recently, Tesco has signed adeal toconstruct the largest distribution centre inPoland. This 57.000m distribution centre will serve Tesco stores insouthern Poland and will potentially employ 1.000 people. (www.polishmarket.com April24, 2010). Also inRomania and Bulgaria, distribution centers start emerging. For example, inRomania, the German retailer Lidl isplanning toopen anew logistic centre, which will eventually serve over 100 stores inRomania in2010. InBulgaria, Lidl isplanning toopen adistribution centre inSofia. Penny Market, another discounter which isactive inRomania and Bulgaria, has also adistribution center ineach country. Based onour interviews with retailers inBulgaria, Romania and Poland, wefind that the above described evolution also depends onthe product type. For processed and semi-processed products, wefind that currently the supply ofthe individual stores iscentralized trough distribution centers. Incontrast toprocessed and semi-processed food products, wefind that the procurement offresh products, such asfruits and vegetables, isoften still organized atthe level ofthe store and that the store isresponsible for the purchases offruits and vegetables trough contracts with local suppliers (mainly large legal entities orcooperatives). However insome, cases also the purchases offruits and vegetables are centralized trough a distribution center as this was for example the case for amajor retailer inRomania. Changes.inthe.relationship.farmer.wholesaler/retailer In the beginning of the 1990s, the traditional wholesales channels were not able to deliver the quantity and quality ofsupplies that modern retailers requested. There are several reasons for this. First, farms were not bewilling tosupply their output tothe processor orwholesalers because they feared not being paid once they delivered the product (Gow and Swinnen, 1998; Van Herck etal. 2010). Second, iffarms wanted tosupply, they were sometimes not beable tobecause they could not access basic production factors (feed, fertilizer, seeds, capital, etc.). Third, if farms wanted tosupply, they were often only able tosupply poor quality supplies because (a) they lacked the necessary inputs toimprove the quality and (b) they lacked expertise and know-how for producing high quality. Retailers and wholesalers typically addressed these problems byintroducing a preferred supplier system (Dries etal., 2004; Dries and Reardon, 2005; byFarina etal., 2005).19 Byworking with contracts (explicit orimplicit contracts), wholesalers (and retailers) were able toselect the suppliers that were able toreach acertain level ofquality. This lowers their transaction costs because itlowers the search costs and the number ofsuppliers per unit sold. Intheir contracts with suppliers, wholesalers give incentives (positive and negative) tomeet the demanded requirements and typically these include vertical coordination mechanisms, such as prompt payments and farm assistance programs. Farm assistance has taken many forms including input supply programs, investment assistance programs, trade credit, bank loan guarantee programs, extension and management advisory services, etc. (Dries etal. 2009; Gorton and White, 2007; Swinnen, 2007).
19.- See more onthe purchasing practices ofmodern retailers insection 5.2.3.
56
Over the past ten years, the importance of vertical coordination of the supply chain has been rapidly increasing inthe transition and developing countries. Inthe Czech Republic, Slovakia and Hungary, 80 % ofthe corporate farms, who dominated farm production inthese countries, sold crops oncontract, and 6085 % sold animal products oncontract (World Bank, 2005). The main reason for contracting was inall three countries, the guaranteed sales. However, besides offering a guaranteed supply to farmers, these contracts often also include farm assistance programs, which are also afairly important reason for farmers tocontract. The growing importance offarm assistance programs and its positive impact on agricultural yield and product quality is also indicated by other studies, including in Dries at al. (2009) who interviewed dairy processors inseveral Eastern European countries and inGorton and White (2007) who interviewed agro-food processors infive CIS countries. InBulgaria, Romania and Poland, weonly find limited evidence ofthe existence offarm assistance programs offered bysupermarkets. The programs that are offered tothe producers are transport, assistance insorting and packaging and advance payments (offered bysome retailers). InBulgaria, the producer organization representative mentioned that specialized wholesalers sometimes provide bank loan guarantees toaminority oftheir suppliers. Interestingly, all producer organization representatives indicate that the most important benefit ofcontracting with modern retailers orspecialized wholesalers isthat these partners offer written contracts, while the traditional wholesalers still work with oral contracts. Usually a written contract includes conditions on price, frequency and quantity of delivery and food safety and quality standards that need tobe respected and they are more elaborated than contracts between farmers and traditional wholesalers. For example, while only 45 % ofthe contracts between farmers and traditional wholesalers include conditions on the frequency of deliveries, this is included in94 % ofthe contracts between farmers and modern retailers (Table 27). For producers, the most important benefits ofthese written contracts that they are away toreduce market risk and have guaranteed sales. The Polish producer organization formulated itas follows: Cooperation with supermarkets made the trade more civilized. Table.27:.Contract.terms.(%.ofthe.contracts.that.includes.this) Modern. retailer Quantity Frequency ofdelivery Minimum quality requirements Food safety requirement Requirements related tosize, shape orcolour Price Mode and speed ofpayment Premiums for quality and/orlarge quantities Penalties for not fulfilling the terms ofthe contract Source: Own calculations based onstakeholders interviews
57
Traditional. wholesaler 71 % 45 % 50 % 45 % 27 % 27 % 50 % 9% 23 %
94 % 94 % 91 % 79 % 70 % 45 % 91 % 15 % 42 %
5.2.2. Impact.onProcessing.Companies In the past years, there was a rapid increase in product variety for all food products, including processed food, such asmilk, bread and sugar, which modern retailers and discounters purchase from food processing companies (see section 4.1.4). However, this isno guarantee that local food processors benefited from these changes asit ispossible that modern retailers only increased the sales from large, multinational food processing companies. The number of food processing companies has decreased over time, except for Romania where the number was relatively stable. InBulgaria, the number offood processing enterprises decreased by2,2 % per year and inPoland by3,4 % (Table 28). Much ofthis was due torationalization and scale increases. Firm size and the total number of employees employed in the food processing sector increased. InBulgaria, the number ofemployees infood processing companies increased per year with 2,4 %, while inPoland, the increase was more limited, only 0,4 % per year. InRomania, the number ofemployees was relatively stable in2007 compared to2003. The impact onfirm size interms ofturnover islarger than onemployment. Turnover increased byapproximately 15 % per year ineach ofthe three case study countries. Itis unclear towhat extent changes inmodern retail sector had animpact asthese changes coincide with the recovery from transition, strong economic growth and EUintegration. Most likely the impact israther limited. Table. 28:. Number. of employees. in food. processing. companies,. number. of companies. and. annual. turnover.(inMio.Euro) Bulgaria Number ofpersons employed Number ofenterprises Turnover (Mio Euro) Number ofpersons employed Number ofenterprises Turnover (Mio Euro) Number ofpersons employed Number ofenterprises Turnover (Mio Euro) 2001 93509 6265 1858,2 na na na 208924 9929 4913,0 2003 102444 6375 2122,9 439050 17975 27308,8 200476 10688 5233,1 2005 106962 5937 2811,9 438833 16998 34501,9 203840 10820 7171,3 2007 110208 5300 3910,8 447061 15498 43639,6 207638 9993 10065,9
Poland
Romania
Source: Eurostat Online Database 5.2.3. Impact.onSmall.Producers Akey concern inthe debate onthe welfare implications ofthese changes isthat the emergence and rapid spread ofmodern supply chains incountries will push alarge share offarmers and inparticular the poor, small farmers, out ofthe market asretailers prefer tocontract with larger and wealthier farmers. This can potentially affect the way income isdistributed within the rural economy and can exacerbate existing patterns ofeconomic stratification. The reason that inparticularly small farmers are said tolose out isbecause oftwo reasons. First, there isan important fixed transaction
58
cost component in costs of exchanges between farms and retailers, making it more costly for retailers to deal with many small farmers than with a few larger suppliers. Second, when some amount ofinvestment isneeded inorder tocontract with orsupply tothe company, small farms are often more constrained intheir financial means for making the necessary investments, either because they donot have sufficient own resources orbecause they have problems accessing external funds inimperfect rural financial markets. However, one can also identify several reasons why retailers could want to work with small, poor farmers such that the impact of the emergence and rapid growth of modern supply chain onfarmers, including small and poor farmers, may beless dramatic than predicted and that there may even have been positive effects on farmers welfare. First, the most straightforward reason isthat companies have nochoice with whom tocontract and with whom not. Insome countries, farm output is dominated by larger farming companies, while in other countries small farms dominate. In the case that small farmers dominate and represent the majority of the potential supply base, retailers need tointegrate these small farmers inthe supply chain inorder tohave sufficient supplies. This is, for example, the case inthe dairy sector inPoland and Romania, and in many other sectors in Eastern European countries (World Bank, 2005). Second, case studies from transition countries suggest that company preferences for contracting with large farms are not as obvious as one may think. While retailers may prefer to deal with large farms because of lower transaction costs, contract enforcement may be more problematic, and hence costly, with larger farms. Third, insome cases small farms may have substantial cost advantages. This isparticularly the case inlabour intensive, high maintenance, production activities with relatively small economies ofscale. Fourth, retailers may prefer amix ofsuppliers inorder not tobecome too dependent onafew large suppliers.
Box 4: stuDies onthe impaCt ofmoDern retail anD stanDarDs onsmall proDuCers The empirical evidence onexclusion ofpoor, small farmers and the welfare effects ismixed (Swinnen, 2010). In the early literature, there were several studies that confirmed the prediction that small and poor farmers are systematically excluded from the supply chain, which negatively affected their incomes (Farina and Reardon, 2000). For example, several studies inLatin America and Africa argued that small farmers were being left behind inthe supermarket-driven horticultural marketing and trade (Key and Runsten, 1999; Dolan and Humphrey, 2000; Weatherspoon etal., 2001; Reardon etal., 2003; Weatherspoon and Reardon, 2003; Boselie etal., 2003; Humphrey etal., 2004;). In contrast, recent research suggests a more nuanced picture of the effect of vertical coordination onsmall farmers exclusion. Infact, surveys inEastern Europe and the CIS find noevidence that small farmers have been systematically excluded from supply chains (Dries and Swinnen, 2004; Dries et al., 2009; World Bank, 2005; Maertens and Swinnen, 2009a, Bellemare and Stifel, 2009; Wang et al., 2009; Miryata et al., 2009) or find that inclusion isconfined tothose that are relatively well capitalized with non-land assets such as irrigation (Hernandez et al., 2007; Neven et al., 2009) or those with access through established producer organizations (Balsevich etal., 2005). Moreover, evidence suggests that farmers who deliver tomodern supply chains experience higher, more stable incomes. Minten
59
etal. (2009) and Maertens and Swinnen (2009a) find that poor rural households experienced measurable gains from supplying high standards horticulture commodities toglobal retail chains inMadagascar and Senegal. InChina, Miryata etal. (2009) find significant effects onincome for supplier farmers and besides aneffect inincome, Bellemare and Stifel (2009) also find a significant reduction in income volatility for supplier farmers in Madagascar. In Senegal, where most of the small farmers are not integrated in modern supply chains, the poor (including small farming households) benefit from increased participation through labor markets (Maertens etal. 2010). Recent studies from Eastern Europe and elsewhere (see Box4) conclude that the impact ofmodern retail investments and the associated standards onsmall producers isvery mixed and the situation isquite heterogeneous across countries and sectors. Moreover, even inthose regions where small farmers are excluded from the growth of modern food supply chains, they may benefit from increased employment that iscreated onlarger and more productive farms. Information obtained from our interviews inBulgaria, Romania and Poland confirms that there isalot ofheterogeneity and that the effects are mixed. Ingeneral, retailers indicate that they prefer topurchase fresh products, such asfruits and vegetables, from large legal entities. However, when this isnot possible, they supply from small farmers trough aspecialized wholesaler. The representatives ofthe producer organizations indicated several constraints for small farmers delivering tomodern retailers. First, procurement mechanisms vary between retailers. Some modern retailers are found tousually pay three weeks orone month after the delivery ofthe products20, which can beproblematic for small farmers that donot have afinancial buffer toovercome this period Others pay more promptly, such asfor example Billa Romania and Penny Market. Infact, our producer survey indicates that in17 % ofthe sales toamodern retailer, the modern retailer pays apart ofthe price inadvance and the remaining part atthe time ofdelivery, while the traditional wholesalers never paid inadvance. Second, inorder tobe allowed todeliver tosome large retailers farmers have topay anentrance fee, socalled access charges, which are often too high for small farmers (inBulgaria and Romania). For example, inRomania, the representative ofthe producer organization, indicated that the access charges can beup to10 % ofthe price that the farmer will receive from the modern retailer for his products. Already in2008, the employer organizations and trade unions inRomania, indicated that itvery difficult for small farmers todeliver tomodern retailers because they cannot supply sufficient quantities. They also indicate that the access charges that modern retailers charge are substantially higher for local producers that are only able tooffer small quantities tothe modern retailer compared tothose delivering large quantities which makes itvirtually impossible for small producers todeliver toamodern retailer. Asaresult, the Romanian government imposed aban onsuch practices. However, our interviews with producers and their representatives, indicate that practices such asasking local producers for access charges are still inplace and are still used byat least some retailers toexclude small producers from getting acontract with amodern retailer.
20.- This isthe case for 78 % ofthe sales toamodern retailer.
60
Finally, the producers and their representatives also mention that increasing quality standards (requirement of several certificates on chemical use) and the poor packaging and sorting infrastructure are important constraints for small farmers to deliver to supermarkets (see also section 5.1.3 on the product quality for differences in quality requirements between modern retailers and traditional wholesalers). However, based on (mostly anecdotic) evidence weobserve with respect tothis aspect variation between modern retailers, For example, inBulgaria, aproducer, who delivered in 2009 to Carrefour and Kaufland (most related to the discount format at that time), indicated that Carrefour was demanding higher quality products than Kaufland although both retailers paid the same unit price. Important tonote with respect toquality requirements, isthat wedo not find asignificant difference inthe quality that supermarkets requested compared tothe quality that discounters requested (both demand extra or/and first class products). Nevertheless they indicate that it is not impossible for small farmers to contract with modern retailers, but they emphasize the importance ofcooperation between small farmers such that they are able todeliver sufficient quantities tothe modern retailers. There can bean important role inthis cooperation for the producer organizations asthey already help farmers toconnect tothe market byproviding assistance programs, such extension services and storage facilities, and establishing contacts between farmers and modern retailers. Finally, when discussing the impact ofretail investments onsmall and large farmers itis important tokeep inmind that for most ofthe products which are sold insupermarkets and discount stores that the relationship with farms isindirect, atbest. Itis primarily inthe area offresh fruits and vegetables that there isapotential direct relationship between the supermarket and the farm. For all other products, such asmilk, bread, processed fruits and vegetables, wine, sugar, etc the impact ofthe retail investments onthe farms isindirect atmost. Inthese cases, supermarkets and discounters purchase their supplies from food processing companies, such asdairy companies, sugar processors, etc. Most, ifnot all ofthese companies, are already operating according tostandards imposed byretailers inthe major (urban orforeign) consumer areas and following EUrules. Hence for most ofthese sectors the impact ofretailer and discounter investments onlocal agricultural producers may bequite limited. 5.2.4. Impact.onAgricultural.Employment.and.Incomes Agricultural employment has declined strongly over the past 15 years in all three countries. InPoland the decline inagricultural employment started already inthe mid-1990s, inBulgaria and Romania the decline was strongest inthe past decade (Figure 12). The cause ofthis strong average fall inemployment isdue toacombination offactors, most ofwhich have nodirect relationship with the growth ofmodern retailing. The overall growth ofthe economy has both attracted much surplus labor from agriculture which was typically characterized bylow productivity and low wages inthe 1990s. Inaddition, the improvements ofthe countries overall budgetary situation and anassociated increase inunemployment benefits and pensions over the past decade has induced many older and low productive people employed inagriculture tostop farming. Atthe same time the effects ofEuropean integration have been very positive instimulating higher productivity and higher incomes inthese countries for those remaining infarming (Figure 13and Figure 14).
61
Figure.12:.Agricultural.Employment.(inAWU.Annual.Working.Units)
4000 3500
Agricultural Employment ('000 AWU)
3000 2500 2000 1500 1000 500 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Bulgaria Poland Romania
Source: Eurostat Online Database Figure.13:.Agricultural.Productivity:.Wheat.Yields

4,5
Wheat Yield (ton/ha)
3,5 Bulgaria Poland 3 Romania
2,5
2 1989-1992 1993-1996 1997-2000 2001-2004 2005-2008
Source: FAOstat
62
Figure.14:.Agricultural.Labour.Productivity.(ALP)
8000 7500
Agricultural Output per AWU
7000 6500 6000 5500 5000 4500 4000 3500 3000 Bulgaria Poland Romania
2000-2001 2002-2003 2004-2005 2006-2007 2008-2009 Source: Eurostat Online Database

The strong impact ofEuropean integration relates, first, tothe EUaccession for Poland in2004 and ofRomania and Bulgaria in2007 which (a) increased subsidies for farms inthese countries, (b) improved the institutional and macro-economic framework in which they operated, and (c) imposed aseries ofstringent public standards onfarmers and food production. Second, European integration also induced massive foreign investments and trade integration prior toaccession which made technology, capital, etc. available tothe food chains inthese countries. Atthe same time, atleast part ofthe type ofproduct requirements which are now imposed bythe retail chains were already introduced with the global integration ofthese countries over the past years. Inthis sense the imposition ofthe product requirements bymodern retailers ismore anevolution ofaprocess that has started over the past 15years rather than asudden dramatic change with the arrival of new investments. It is not possible with the available data to measure the impact ofthe growth ofmodern retail and ofdiscounters effect inarigorous econometric way. Itis likely tohave had two opposing effects. Onthe one hand, byenhancing the requirements for producers itis likely tohave enhanced production constraints and therefore reinforced incentives for the less skilled and low productive farmers toeither stop farming or (continue to) produce for subsistence purposes orfor local informal markets. Onthe other hand, itwill have reinforced incentives for others toupgrade the quality oftheir produces and inthis way continue toor enter producing for the formal higher quality, and higher value, markets. One argument that isoften used isthat modern retailer exercise buyer power and pay farmers prices that are very low and insome cases not even cover the costs. However, inthe existing literature onthis matter these isremarkably little compelling evidence about abuse ofbuying power onthe upstream side. Onthe one hand, Lloyd etal. (2009) test buyer power and show that the hypothesis ofperfect competition can berejected in7ofthe 9food products investigated. Onthe other hand,
63
Sexton et al. (2005) mention that it is practically impossible to measure the retailers buying power, asprices paid byretailers totheir suppliers are typically not revealed. Apart from some weak anecdotal evidence, also Dobson etal. (2001) fail tofind clear evidence ofabuse ofmarket power vis--vis farmers. They review a case where a farmers organization in the UK alleged that supermarket chains pay very low prices for farm products, but fail to pass low prices on to consumers. After aninvestigation, however, the Competition Commission (2000) argued that the low producer prices were mainly aresult ofexcess supply, and had been passed onto consumers, orwere compensated byother cost increases. Based on our interviews, we did not find inclusive evidence on this. The representative of the producer organization inRomania accused modern retailers ofexploiting their market power and paying very low prices tothe producers (often even lower than production costs). However, when wecompare prices received byfarmers for deliveries ofthe same quality tomodern retailers and traditional wholesalers, prices offered bymodern retailers were even higher than those paid bythe traditional wholesalers, but the price difference isnot significant.21 InBulgaria, the representative ofthe producer organization indicates that every week the price isrenegotiated and that insome weeks prices cover production costs, while insome weeks they donot, but this isthe same with prices offered by traditional wholesalers. However, we want to emphasize that we have only anecdotic evidence and that detailed data onprices atall stages ofthe supply chain (information which iscurrently not available yet) isabsolutely necessary toprovide insights inthis matter.
Box 5: impaCt onproDuCers Opportunities In their contracts with suppliers, specialized wholesalers (and modern retailers) give incentives to meet the demanded requirements and typically these include vertical coordination mechanisms, such asfarm.assistance.programs. Inthe case study countries, the programs that are offered tothe producers are transport, assistance insorting and packaging, advance payments and bank loan guarantees (toaminority ofthe suppliers inBulgaria). Typically contracts with specialized wholesalers (and modern retailers) are written. contracts, while the traditional wholesalers still work with oral contracts. Awritten contract includes price arrangements, frequency and quantity ofdelivery and food safety and quality standards that need to be respected. For producers, written contracts are important as it isaway toreduce market risk and have guaranteed sales. Threats By charging a access. charges, which are disproportionally higher for producers delivering only small quantities, modern retailers exclude small producers from having individual contracts with modern retailers. Such practices are commonly observed inBulgaria and Romania, although they are prohibited bylaw (inRomania).
21.- Wedid not find asignificant difference between prices paid bysupermarkets and discounters. Only the prices received onthe local market are significantly higher which isthe reason why most producers sell apart oftheir production onthe local market.
64
The representatives of producer organization in Romania accused modern retailers of exploiting their market. power and paying low prices (lower than the production costs) to producers. However, these findings are not confirmed by the price data that we have collected in the producer survey. Hence, more information on price formation trough the supply chain isneed toprovide insights onthis matter. Modern retailers are found to usually pay only one month after the delivery of the products and providing trade.credit (even when agreed beforehand) can beproblematic for small farmers that donot have afinancial buffer toovercome this period. Nevertheless, there is large variation between retailers and in fact, in 17 % of the contracts between modern retailers and farmers wefind that the retailer even pays apart ofthe payment inadvance. Quality standards ofmodern retailers should bebased on true.quality (freshness and safety and health standards) and not esthetic quality (standards related tosize, shape and color). Currently, 24 % ofthe producers report that itoccurred tothem that their delivery was rejected based onsuch esthetic quality concerns despite the fact that they fulfilled all quality requirements stated inthe contract (see also section 5.1). Recommendations tomaximize positive and mitigate adverse social impacts onproducers From our interviews itappeared that asymmetry inbargaining power between modern retailers and their producers may lead to unfair trading practices, such as the existence of a access charge in Romania and Bulgaria. In order to reduce these activities, there should beacommission.that.isoccupied.with.analyzing.potential.unfair.trading.practices.and. increasing. producers. awareness. on their. legal. rights. This commission should analyze the existence and use ofclassic cartels, joint purchasing agreements (buying alliances) and private labels. For such practices acareful balancing ofefficiency enhancing and potentially anticompetitive effects isneeded and acase-by-case analysis based onthe specificities oflocal market conditions isnecessary inorder toestablish the existence ofapossible competitive harm. Competition Commissions are introduced inseveral developing countries (e.g. Brazil, Indonesia and Mexico). Inorder toresolve the tension between modern retailers and their suppliers, they could engage ina code.ofbest.practice. The terms ofsuch acode could include asmain elements: (1) compliance with contracts byretailers and suppliers; (2) equal treatment ofsuppliers; (3) prompt payments; (4) banning unfair trading practices. There are already some countries that introduced such acode. In2002, aprivate sector code was encouraged bythe competition commission inthe UK (and later itwas made mandatory). Also inArgentina there exists a code ofbest practice. Modern retailers demand aconstant delivery ofproducts and because sometimes small producers cannot provide the demanded quantity, modern retailers may prefer towork with one large producer instead ofseveral small ones. Inorder toovercome this problem farmers must form groups (orproducer.organizations) tohelp their members toconnect tothe market byoffering them assistance programs, such asextension services and storage facilities, and
65
by establishing contacts between farmers and modern retailers However, the willingness to cooperate is still weak in most Eastern European countries. First, it will be important that the government creates a clear legal framework in which such activities are possible. For example, inBulgaria, the person incharge ofthe producer organization indicated that the producer organization asalegal entity isnot allowed tobulk the production ofdifferent suppliers, but only toestablish contacts between individual producers and modern retailers. Second, there should beacommission that analyzes the existence ofjoint commercialization agreements and analyzes on a case-by-case analysis whether there exists a potential competitive harm. Anessential problem for (small) farmers isthat they donot have access tocapital tomake investments, which are needed tofulfill the quality requirements that modern retailers impose ontheir suppliers. Insome cases, modern retailers are intermediating between commercial banks and farmers (e.g. offering bank loan guarantees inBulgaria). However, there isascope for innovative private-public actions which increase farmers.access.tocapital. For example, inMexico there isthe financier rural program (Reardon and Berdegu, 2006).This program allows the supplier toget immediate cash instead ofhaving a waiting period for payment and the bank then invoices the payment from the modern retailers. Inaddition, itwill becrucial toincrease.the.price.transparency ateach stage ofthe supply chain. This will allow usabetter comparison ofprices paid bydifferent stakeholders and provide more information onthe price margin that each stakeholder takes. Second, also for farmers such atool isinformative ascurrently they have only limited information onprices.
5.3 Impact on Traditional Retailers

Besides affecting consumers and producers, the increase inmodern retail isalso affecting traditional retailers. On the one hand, one might expect that increased competition from modern retailers decreases revenues of traditional shopkeepers and may even force traditional shopkeepers out ofbusiness. Onthe other hand, modern retailers create new employment opportunities and there might bepositive spillover effects (e.g. increased variety and increased labeling tomimic the supply offered bythe modern retailer). There are a number of studies that indicate a negative correlation between the establishment ofamodern retailer and the number ofsmall retail shops inthe developed world. However, much ofthis literature isuninformative asit does not say anything onthe causal impact ofthe emergence ofmodern retailers onretail employment and earnings (and therefore itis not discussed here). There are afew academic studies that deal with the causality and most ofthese studies focused on impact of the emergence of large (Wal-Mart) retail stores on regional retail employment. Contrary topopular belief, Hicks and Wilburn (2001) find that the opening ofWal-Mart stores inWest-Virginia has apositive impact onretail employment and the number ofretail companies inthe period 19891996. Using nation-wide data and controlling for the endogeneity ofthe timing ofWal-Martsentry inaparticular market, Basker (2005) estimates that inthe year ofWal-Mart
66
entry, retail employment inacounty increases by100 jobs. Half ofthis gain disappears over the next five years as other retail establishments exit and contract, leaving a long-run statistically significant net gain of 50 jobs. In a more recent study, Sobel and Dean (2008) not only focus onemployment insmall businesses inthe retail sector, but also onoverall self-employment and their results suggest that the rapid spread ofWal-Mart stores has had nostatistically significant long-run impact on the overall size and profitability of the small business sector in the United States. They use Schumpeters theory on creative destruction toexplain that there are inventions, such asfor example the emergence ofaWal-Mart store, that result inbusiness failures incertain sectors, but despite these failures yield net gains because of the positive impacts on economic activity in other sectors. For example, a small traditional shop is converted into an art gallery orthe office ofatravel agent. Most studies indeveloping countries focus onthe correlation between the emergence ofmodern retailers and the number oftraditional retailers and mainly find anegative correlation (Gutman, 1997; Faigenbaum etal., 2002; Rodriguez etal. 2002). Ingeneral, they find that small shops are hit the hardest, while specialized shops and street fairs tend tohold out much longer. When differentiated by products sold, those selling dairy products or processed food tend to go out of business the earliest, with fresh produce shops and wet markets following afterwards (Faigenbaum etal., 2002; Rodriguez etal. 2002). Box6presents conclusions from these studies.
Box 6: stuDies onthe impaCt onthe numBer oftraDitional shops InArgentina, Gutman (1997) finds that the number ofshops declined from 209.000 in1984 to145.000 in1993. Also inArgentina, Rodriguez etal. (2002) note that while the number general small shops rapidly decreased, those inaspecialized niche, inparticular bakeries, fresh fish and meat, and fruit and vegetable shops, disappeared less quickly. After several years ofcompetition, the traditional retailers that are usually still inbusiness are those selling niche products. InChile between 1991 and 1995 15.777 small shops disappeared, ofwhich 2122 % ofgeneral shops, 25 % inthe number ofshops with dairy and meat products and adecline ofonly 17 % inthe number ofshops that sell fruit and vegetables (Faiguenbaum etal. 2002). Anexception tothe evolutions isBrazil, where itappears that traditional traders are able tocoexist with modern retailers bydepending ontraditional food preparation habits and ensuring fresher products (Zinkhan etal., 1999; Farina etal., 2005). However, like already indicated for studies in developed countries, these studies do not allow saying anything onthe causality. The only study that conducts animpact evaluation ofthe effect ofsupermarkets ontraditional markets inatransition ordeveloping country isby Suryadarma etal. (2010) and isbased onsurvey evidence from Indonesia. They selected traditional markets close tothe modern retailer asthe treatment group and traditional markets far from the modern retailer as the control group for a difference-in-difference analysis. They find that on average, traders inboth treatment and control markets have experienced adecline intheir business over the past 3years and hence that there isno statistically significant impact ofcompetition from modern retailers onthe profit and revenue oftraditional traders.
67
Qualitative evidence confirms these results as traders relate the decline profits to a decline ofpurchasing power ofthe local population and the competition with street vendors (competition with supermarkets was only the third most important reason). The traders, market managers, and traders representatives all state that the main steps which should beundertaken toensure their survival are the improvement oftraditional market infrastructure, organization ofthe street vendors, and the implementation ofbetter market management practices. Ifthese conditions are met the traders explicitly state their confidence that supermarkets would not drive them out ofbusiness. In order to deal with the possible endogeneity bias due the timing and the location of the establishment ofamodern retailer inaparticular market, agood approach toanalyze the effect ofthe emergence ofamodern retailer onemployment oftraditional retailers would beusing adifferencein-difference-in-difference estimator that compares changes in the number of traditional shops and their earnings over time in regions in which there emerged a modern retailer and regions where there did not emerge amodern retailer, compared tochanges for manufacturing and services employment and earnings. However, such an econometrical approach requires a detailed panel survey in the three countries for which, within the current study, the time nor the money was available. Nevertheless, we want to emphasize the need for such research, which is especially interesting, given the fact that modern retailers are only recently emerging inthe rural areas ofthe countries under consideration. Given the constraints ofthe study, weopted touse asimilar approach asSuryadarma etal. (2010), who interviewed traditional shopkeepers close to the modern retailer (treatment group) and traditional shopkeepers far from the modern retailer (control group). Hence, byapplying this strategy, weassume that traditional shops close tothe modern retailer are indirect competition with the modern retailer, while traditional shops far away (20-30 km) from the modern retailer are assumed toexperience nodirect competition from the modern retailer. Given the poor rural infrastructure (poor roads and public transport facilities), this assumption seems reasonable. InBulgaria, Romania and Poland, weasked traditional shopkeepers far and close tothe supermarket on the evolution of their gross receipts and the reason behind this evolution22. The results are mixed: out ofthe 6interviewed traditional shopkeepers close tothe modern retailer, 3mention that there was adecline inthe gross receipts, 2indicate that gross receipts did not change over time and 1shopkeeper mentions aslight increase inreceipts. Out ofthe five shopkeepers far from the supermarket that were willing totalk tous ontheir gross receipts: 3mention adecline inreceipts and 2indicate that receipts slightly increased23.
22.- Besides traditional small shops also open markets play ankey role inthe food supply chain. Based onour consumer surveys wefind that alarge proportion ofthe rural population buys fruits and vegetables onthe open market and hence open market are still important for rural consumers. However, also for producers, open markets are important and most ofthe producers that weinterviewed sell apart oftheir production (5 % to10 %) onthe open market because ingeneral prices onthe local market are higher than those offered bymodern retailers ortraditional wholesalers. However, 70 % ofthe farmers expect direct sales todecrease over time, while 17 % expect direct sales tocontinue and 13 % expects anincrease indirect sales. These expectations are confirmed bythe representatives ofconsumer and producer organizations. Infact, one ofrepresentatives formulates itas follows: the farmers selling onopen markets will slowly disappear, itare farmers ofthe past generation. However, itis uncertain whether itis the competition with the modern retailer that drives them out ofthe market. Infact, several farmers that diminished direct sales, indicate that the main reason isthat they can only sell small quantities onthe local market and that therefore the unit cost oftransporting their products tothe open market istoo high. None ofthem indicated that itwas due toincreased competition with the modern retailer. 23.- Some shopkeepers indicated the monetary value ofthe increase/decrease intheir gross receipts, however others refused totell this and only indicated whether their gross receipts went upor down. Weopted toonly present this information asthis seemed tous the most reliable information.
68
The fact that both traditional shopkeepers that are situated far away and close toamodern retailer experienced adecline ingross receipts could suggest that there might beanother factor than the emergence of a modern retailer (for example the recent economic and financial crisis) affecting gross receipts oftraditional shopkeepers both far and close tothe modern retailer. However, itis also possible that our assumption that traditional shopkeepers far from the modern retailer are not experiencing direct competition from the modern retailer isfalse. This seems tobe case asfrom interviews with consumers oftraditional shops far from the modern retailer welearnt that those consumers visit the modern retailer onaverage two times per month (see section 5.1.7.), such that at least for some products also traditional shopkeepers far from the modern retailer experience direct competition from the modern retailer. In addition, we find that the most cited reasons for a decline in gross receipts of traditional shopkeepers, both for those close and far from the modern retailer, are increased competition with modern retailers and decreasing purchasing power due tothe recent financial crisis. These findings seem toindicate that the presence ofamodern retailer has anegative impact ongross receipts ofboth traditional shopkeepers close and far from the modern retailer, although wecannot econometrical disentangle the impact ofthe emergence ofamodern retailer and the impact ofthe recent crisis onfarm receipts. However, when weanalyze other indicators, wealso find evidence ofpositive effects ofchanges inthe retail sector ontraditional shopkeepers. For example, insection 4.1.2weshow that since the emergence ofmodern retailers the variety offered intraditional shops has increased, indicating that there may bepositive spillover effects ofmodern retailers ontraditional shops.
Box 7: impaCt ontraDitional retailers Opportunities There isevidence ofpositive.spillover.effects.ofthe emergence ofmodern retailers onthe variety offered bytraditional retailers. Threats The emergence ofmodern retailers have anegative.effect.onthe.gross.receipts (and hence business survival) oftraditional shopkeepers (both those located far and close tothe modern retailer). This isalso reflected inthe overall decline inself-employment inthe retail sector. Recommendations to maximize positive and mitigate adverse social impacts on traditional retailers Increased competition from the modern retailers may negatively affect gross receipts (and hence business survival) oftraditional shops. Inorder tomitigate these effects traditional shopkeepers can focus on providing certain. services. that. modern. retailers. do not. provide, such ashome delivery, credit provision, (see section 2.2.2). Another option isto specialize incertain high.quality.niche.products.oroffering.prepared.dishes that are not (orless) available in the modern retailer. Modern retailers can mitigate the impact they have on traditional
69
shops byproviding advice totraditional shops onoffering new services. For example, Metro Group isplanning tooffer such advisory services totraditional shops when they are setting upcash and carry shops inKazakhstan (Grytsenko, 2011). Nevertheless, there will be shopkeepers who are not able to face the competition with the modern retailer. However, this should not necessary be a negative evolution as the emergence ofamodern retailer, like Sobel and Dean (2008) indicate, can unleash aprocess of creative destruction. This theory, which was first described bySchumpeter, indicates that there are inventions (e.g. the emergence ofamodern retailer) that result inbusiness failures incertain sectors (e.g. traditional shopkeepers), but despite these failures lead tonet gains because ofthe positive impacts oneconomic activity inother sectors. Inorder tofacilitate this transition, it will be important to upgrade. human. capital. in rural. regions. as this. can. beone.ofthe.driving.factors.behind.the.emergence.ofnew.business.activities. The government could provide support measures for job reconversion, but also modern retailers can provide training programs totrain local people for ajob inthe retailer.
5.4 Impact on Employment

The emergence ofmodern retailers affects employment inseveral ways: first, one needs todistinguish between the effect on employment in the retail sector and in the agricultural sector.24 Second, within the retail sector the emergence ofmodern retail channels changed the balance between self (traditional shops) and wage employment (modern retailers). Finally, italso has animpact onthe working conditions. 5.4.1. Impact.onthe.Type.ofRetail.Employment In terms of total retail employment we find that there has been a substitution of self-employed towage employment. Ingeneral, traditional shops only employ 1to3persons, which usually include the owner of the shop, his wife/ her husband and other family members (children, parents,). In the beginning of the 2000s for Poland and the mid-2000s for Bulgaria and Romania, there were on average 3,5 full time workers employed in a traditional shop close to the supermarket and in2010 still onaverage 2,5workers were employed bythese shops (Table 29). Similar findings hold for the traditional shops far from the supermarket, where the number ofemployees decreased from 1,75to 1,5in2010. Onother the hand, inthe modern retail outlets the number ofemployees increased from on average 41 to approximately 59 full time employees in 2010. These findings indicate that modern retailers can play an important role in job creation in the rural regions ofBulgaria, Romania and Poland.
24.- Asalready discussed insection 5.2agricultural employment has declined strongly over the past 15years inall three countries. However, the main causes for this strong decline have nodirect relationship with the growth ofmodern retailing. The overall growth ofthe economy has both attracted much surplus labor from agriculture and increased social payments enabled many older and low productive people employed inagriculture tostop farming.
70
Table.29:.Employment.inmodern.retailers.and.traditional.shops.(number.employees) Beginning.of2000s.(Poland)/. Mid.2000s.(Bulgaria.&Romania) Modern retailer Traditional shops close tothe modern retailer Traditional shops far from the modern retailer Source: Own calculations based onstakeholders interviews Moreover general employment statistics indicate that the emergence ofmodern retailers and the shift from self-employed labour towage employment, has had apositive impact ontotal retail and wholesale employment. InFigure 15, wepresent the evolution ofemployment inthe retail and wholesale sector inBulgaria, Romania and Poland (employees and self-employed individuals). In Bulgaria and Romania, employment inthe retail sector increased byrespectively 25 % and 20 % compared to2000. InPoland, employment increase was more moderated and employment inthe retail sector increased by15 % compared to2000. Inall three countries, there isastabilization oreven aslight decline inemployment inthe past three years, anevolution which wecan probably relate tothe financial crisis. However, there are substantial differences between self-employment and wage employment. While self-employment in all three countries decreased (or remained stable) compared to 2000, wage employment increased in all three countries (Figure 16 on self-employment and Figure 17 on wage employment). These changes are also reflected inthe share ofself-employment. In2000 23 % of individuals employed in the retail and wholesale sector in Bulgaria were self-employed, while in 2010 only 18 % was self-employed. Similar findings hold for Poland and Romania where selfemployment shifted from respectively 28 % and 17 % to21 % and 11 % (Figure 18) (Eurostat Online Database). 5.4.2. Impact.onthe.Quality.ofRetail.Employment When considering the quality ofwork, wefind that both modern retailers and traditional shops have long opening hours (usually from 7-8amuntil 9-10pm) and are open onSaturdays and Sundays. These long opening hours are also reflected inlong working hours, especially for self-employed traditional shopkeepers. Onaverage employees inthe retail sector work 43hours inBulgaria and 42 hours in Poland and Romania, while self-employed shopkeepers work 47 hours in Bulgaria, 48hours inPoland and 44hours inRomania (Eurostat Online Database). The maximum number ofhours that employees are allowed towork each week isdefined bythe countrysemployment legislation. However, inpractice, employees often work longer than officially allowed. For example, in Poland, in a 2008 case study on the working conditions in the retail sector employees ofdifferent retail chains (including discounters) reported violations ofworking time asthe most important concern. These violations include imposing long sequences ofworking days (e.g. seven to nine days of continuous work although an employee is entitled to at least
71
2010 59 2,5 1,5
41 3,5 1,75
Figure.15:.Evolution.ofemployment.inthe.retail.and.wholesale.sector.inBulgaria,.Romania.and.Poland*
150 Employment Index in the Retail and Wholesale Sector (100=2000) 140 130 120 110 100 90 80 70 60 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Bulgaria Poland Romania
* The level ofemployment in2010 only relates tothe first quarter of2010. Data also include repair ofmotor vehicles, motorcycles and personal and household goods. Source: Eurostat Online Database Figure.16:.Evolution.ofself-employment.inthe.retail.and.wholesale.sector.inBulgaria,.Romania.and.Poland*
150 Self Employment Index in the Retail and Wholesale Sector (100=2000) 140 130 120 110 100 90 80 70 60 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Bulgaria Poland Romania
* The level ofemployment in2010 only relates tothe first quarter of2010. Data also include repair ofmotor vehicles, motorcycles and personal and household goods. Source: Eurostat Online Database
72
Figure.17:.Evolution.ofwage.employment.inthe.retail.and.wholesale.sector.inBulgaria,.Romania.and. Poland*
Wage Employment Index in the Retail and Wholesale Sector (100=2000) 150 140 130 120 110 100 90 80 70 60 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Bulgaria Poland Romania
* The level ofemployment in2010 only relates tothe first quarter of2010. Data also include repair ofmotor vehicles, motorcycles and personal and household goods. Source: Eurostat Online Database Figure.18:.Share.ofself-employment.inthe.retail.and.wholesale.sector.inBulgaria,.Romania.and.Poland*
Share Self Employment in the Retail and Wholesale Sector (%) 30% 25% 20% 15% 10% 5% 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Bulgaria Poland Romania
* The level ofemployment in2010 only relates tothe first quarter of2010. Data also include repair ofmotor vehicles, motorcycles and personal and household goods. Source: Eurostat Online Database
73
35consecutive hours ofrest inaweek), working longer shifts than official shifts ofeight hours (weekday) or twelve hours (weekend/holiday) and difficulties in taking breaks during the day. These findings are confirmed bythe State Labour Inspection who revealed the existence ofseveral violations related toworking time indifferent super- and hypermarkets. In2007, they found in31 % of the inspected stores errors in time-taking records (faulty calculations of overtime and night shifts) and 24 % ofthe inspected stores ignored the norm ofa5-day work week (Oponowicz and Chmielecka, 2008). Nevertheless these violations of working time by modern retailers, it is difficult to compare the working conditions toself-employed traditional shopkeepers asoften they also report long working hours (sometimes upto more than 15hours per day, but ofcourse with more freedom totake breaks compared toemployees). Moreover, nine out ofthe twelve shopkeepers weinterviewed reported that they did not have aclosure day and the other three only closed their shop one day (onSunday). There are also important differences inthe work done byemployees inmodern retailers and selfemployed traditional shopkeepers. Traditional shopkeepers usually combine aseries oftasks, from filling the shelves tobookkeeping, for which they need alarge variety ofskills, while employees in the modern retailers have more specific tasks (e.g. cashiers, filling the shelves, keeping the inventories, )25. Currently, most retailers in Poland, Bulgaria and Romania report that they find it increasingly difficult tofind the suitably qualified workers for positions atall levels from low-skilled tosemiskilled warehouse workers tomanagers. Most ofthem try toimprove the human capital oftheir work force byoffering attractive remuneration and organizing training. However, this isthe situation in the capital and the large cities, whereas in small towns and rural areas, the minimum wage26 isthe standard, especially asfar aslow skilled sales agents concerned. Moreover, inrural Romania, Euromonitor (2010) signals practices where modern retailers employ workers who are not legally employed and receive aremuneration below the minimum wage (Euromonitor, 2010). InPoland, astudy byOponowicz and Chmielecka (2008) mentions practices where modern retailers pay the minimum wages and the rest ofthe wage ispaid under the counter. This practice has negative consequences for the employees who not only lose money when they goon sick leave (they receive 80 % ofthe lowest remuneration) but also, inthe long run, when they retire (astheir pension will bebased ontheir official wage). However, weare not able tocompare earnings inamodern retailer with earnings of traditional shopkeepers as these were reluctant to tell us about their earnings. Moreover, itwould beincorrect tocompare earnings ofaself-employed shopkeeper and earnings of, for example, acashier, asthey may have different responsibilities and skills. Itis important tonote that the majority ofthe workforce inmodern retailers are women, which implies that many of the employment effects disproportionately affect either women or men,
25.- Besides differences in the type of tasks that traditional (self-employed) shopkeepers and cashiers need to fulfil there are also differences how individuals can perceive these tasks. One can argue that being acashier isstressful and repetitive. Onthe other hand, being self-employed can also bestressful asone carries all responsibilities onsuccess orfailure ofthe business. Inthis study, weprovide no evidence on stress because of two reasons. First, stress is a highly personal matter. Second, in order to be able to make some intrapersonal comparisons weshould have interviewed individuals that switched between being aself-employed shopkeeper tobeing an employee and the other way around. However, given the time and budget constraints and the refusal of most modern retailers tointerview their staff itwas impossible toidentify asufficiently large sample ofsuch individuals. 26.- There is no officially established minimum wage in retailing in Poland, but according Euromonitor, the average gross wage inretailing reached PLN 2.750 ( 675) per month in2007. In2009, the minimum wage inBulgaria isBGN 240 (120) per month and inRomania itis RON 600 ( 141) per month.
74
depending onwhether they are self-employed orbeing hired asan employee inthe modern retail sector. Therefore there isastrong overlap between this section and section 5.5, which deals with gender effects.
Box 8: impaCt onemployment Opportunities Since the emergence of modern retailers there has been a shift from self-employment towage employment inthe retail sector and there isanet gain inemployment, indicating that modern retailers can play animportant role job.creation inrural regions. Usually, the employees ofmodern retailers have shorter.working.hours than traditional, self-employed shopkeepers, although there are indications that modern retailers violate the regulations concerning working time (see threats). Threats InPoland and Romania, there issome evidence ofviolations.ofthe.employment.legislation. in modern retailers. In Poland, these are mainly violations related to working time, while inRomania there are incidences where black workers were hired that were paid awage lower than the minimum wage. Recommendations tomaximize positive and mitigate adverse social impacts onemployment Modern retailers should comply to the local employment legislation (working time, minimum wage, ...). Currently, there are studies that signal violations of this legislation inRomania and Poland. The government plays animportant role indetecting these practices. However, also the modern retailer should take his responsibility. Itis for example possible that the general management isnot aware ofmalpractices inindividual stores ofthe chain (e.g. incase franchising). Inthat case the modern retailer can set upan internal.audit.todetect. malpractices.inthe.individual.stores.and.take.appropriate.measures.toavoid.future.violations. ofthe.employment.legislation. Besides setting upan internal audit system, modern retailers should encourage (orat least not obstruct) actions that increase.awareness.among the employees with respect to labour. legislation. In order to help employees, which are mainly women, combine. work. and. family. life, modern retailers could offer their staff the possibility towork part-time, but inaddition they could introduce shift.patterns, which are communicated along time upfront inorder toallow employees toarrange possible family obligations. Inaddition, modern retailers could offer. child.care.facilities (atcost price). The latter will only beprofitable inlarger plants.
75
5.5 Gender Effects

There are only few studies who have analyzed the gender effects ofinvestments inthe modern food sector and the associated increase infood standards and restructuring ofthe food chains. Most ofthese studies have not focused onthe retail chain per se, and most are case studies from Africa and Latin America. Inthose cases, they have mostly focused onthe fact that contracted family farms are usually managed by men while employment on larger farms and employment in the processing ofhigh quality products atthe food processing company islargely for women. Hence some ofthe restructuring ofthe chains may have important gender effects inthis way, ine.g. rural Africa (see e.g. Maertens and Swinnen 2009b). This type ofgender effects seem tobe less relevant for investments ofdiscounters inEastern Europe. Overall, gender effects are less important in Eastern Europe compared to developing countries (Lopez-Claros and Zahidi, 2005). Nevertheless, wefeel that that isimportant todiscuss the gender effects related to the development of modern retail formats which are more subtle than in the developing world. Based on our small consumer survey we identified some gender effects in terms of shopping behavior. Itappeared that men are more likely toshop inmodern retailers, but this effect was not statistically significant. Inaddition, weanalyzed whether there were differences inthe per capita monthly food expenditures ofmen and women shopping inthe modern retailer, but also inthis case wedid not find statistically significant results. Moreover, wewant toemphasize that inorder tostudy gender effects onconsumption inamore accurate way, one needs tohave information from (panel) household surveys that collect more detailed information onhousehold consumption and the intrahousehold decision-making process. However, within the current study, time and money constraints did not allow usto collect such data. Inaddition, wefind that buying several products in one retail outlet substantially reduced the time that female consumers ofmodern retailers spend onshopping, leaving more time available for other activities. In 2010, they spend on average 3,0 hours per week on buying groceries, while before this was 2,7 hours per week. Male consumers of modern retailers spend in 2010 on average 2,1 hours on buying groceries, while before this was 2,3 hours. Moreover, we find that female consumers more often indicate convenience asthe most important reason tochange their shopping behavior and buy at least one of the product categories in the modern retailer. Based onour consumer survey, wefind that 17 % ofthe female consumers indicated convenience asthe most important reason tobuy atleast one ofthe eight product categories inthe modern retailer, while only 12 % ofthe male consumers indicated convenience asareason tochange their behaviour. However, none ofthese differences between male and female consumers isstatistically significantly. The main gender effect from discounter and modern retail investments appears to be the shift from employment from small shops tolarger retail outlets, and the associated working conditions. Interestingly, we find that mainly women benefit from this shift from self-employment to wage employment as mainly women are employed as employees in the retail sector. Based on our company interviews, wefind that 65 % to70 % ofthe staff inamodern retail store isfemale. Table 30shows that there isalarge difference between the share ofwomen inself-employment compared towage employment inthe retail sector.
76
For example in Romania, in the first quarter of 2010, approximately 56 % of the employees inthe retail sector were women, while only 28 % ofthe self-employed shopkeepers were women. Similar results hold for Bulgaria and Poland, where respectively 57 % and 58 % ofthe employees inthe retail sector were women and only 37 % and 40 % ofthe self-employed shopkeepers were women. Table.30:.Percentage.offemales.employed.inthe.retail.sector.inself-employment.and.asemployee. inthe.first.quarter.of2010.(%) Self.Employed Bulgaria Poland Romania Source: Eurostat Online Database This implies that many ofthe effects which wedescribed inthe previous section may disproportionately affect women asthey are more likely tobe employed asan employee byamodern retailer. The net welfare effect might be positive or negative depending on what the women were doing before. Ingeneral, most ofthe women employed aslower staff (e.g. cashiers) are young and low skilled and before sending their application tothe modern retail store they were unemployed orstill atschool. For these women employment in a modern retailer might have a positive welfare effect in spite ofthe long working hours and low wages. For those previously employed asself-employed shopkeepers, the welfare effects are less obvious and wage employment may imply a step down, unless there are substantial benefits in working conditions or higher wages associated with wage employment. However, like already indicated in section 5.3, it is difficult to compare working conditions and wages in self-employment and wage employment. Inour interviews, weasked modern retailers which measures they take tohelp their employees tocombine their work and family life, for example, providing their employees the possibility towork part time oroffering child care facilities. Our interviews show that some ofthe modern retailers offer the possibility towork part time although they all indicate that they prefer full time employees. Another concern related to gender inequality in modern retailer is that women get no (or less) opportunities than men toprogress into store supervision ormanagement. Asthe management ofthe modern retailers that weinterviewed was reluctant toanswer this question and refused usto interview their staff, we have only some ad hoc evidence on this matter. This indicates that the problem israther limited. First, several managers ofmodern retail stores that weinterviewed were women, which isafirst indication that gender discrimination israther limited. Second, inaPolish case study onthe working conditions offemale workers insupermarkets/ hypermarkets, Oponowicz and Chmielecka (2008) also find noevidence ofgender discrimination for jobs instore supervision or management. Nevertheless, the women in the study mention that it is not uncommon that during the recruitment process questions about marital status, number and age of children and planned pregnancy are asked, but the authors have noinformation towhat extent employers use this information when selecting acandidate.
77
Employees 37 % 40 % 28 % 57 % 58 % 56 %
The shift from self-employment towage employment also implies that men are disproportionally negatively affected asthey are more involved inself-employment and less in wage employment. However, like already indicated insection 5.4, adecline inthe number oftraditional shopkeepers does not necessary have anegative impact onwelfare asdepends onwhether and inwhich type ofemployment, traditional shopkeepers that stopped their activities, end up. However, within the scope ofthis study (time and money constraints), itwas impossible totrace back shopkeepers that stopped their activities and survey them onthe reason why they stopped their activities and the work that they are currently doing. Moreover, inorder toassess how their welfare changed due tothe emergence ofmodern retailers, weneed detailed data from the period that they were still active asatraditional shopkeeper (before the emergence ofthe modern retailer).
Box 9: GenDer effeCts Opportunities Women (but also men) that dobuy their groceries inmodern retailers indicate convenience asan important reason todo so. This isalso reflected inareduction ofthe number ofhours that they spend ondoing groceries. There are disproportionally.more.women.employed.asan.employee.inamodern retailer, indicating that mainly women benefit from the shift from self-employment to wage employment. However, the welfare effect depends onwhat these women were doing before being employed inthe modern retailer. Ingeneral, most ofthe women that are employed aslower staff (e.g. cashiers) are young and low skilled and they were unemployed orstill atschool when sending their application tothe modern retailer. For these women employment inamodern retailer might have apositive welfare effect inspite ofthe long working hours and low wages. Threats The modern retailers that we interviewed were reluctant to answer any questions related to gender. discrimination. Nevertheless, a case study in Poland indicated that it is not uncommon that during the recruitment process questions about marital status, number and age ofchildren and planned pregnancy are asked, but the authors have noinformation towhat extent employers use this information when selecting acandidate. Recommendations tomaximize positive and mitigate adverse gender impacts Although we have no evidence on gender discrimination by modern retailers, it is important that there is an employee (for example a member of the labour. union) which women can approach with complaints and questions about their legal rights.
78
5.6 Impact on Social Cohesion

Some studies have pointed atthe negative impact ofmodern retail investment and the associated decline in the number of small local shops on the social infrastructure in the area (Goetz and Rupasingha (2006). They argue that small, local shops and their owners play animportant role insocial relationships, norms and trust inacommunity. Moreover, those small shopkeepers are found tobe part ofthe local leadership class. They also understand the interpersonal dynamics ofthe community members and their various networks. Inaddition tothe impact onthe survival ofthe traditional shops, modern retailers affect the supporting industry within communities that serves these shops. This industry includes firms inthe legal, accounting, transportation, logistics, financial, publishing and advertising sectors that work closely with the retailers. The arrival ofamodern retailer, who pursues all these activities atacentralized level, makes that the employees inthese sectors leave the community topursue opportunities elsewhere. Inthe process, the social capital they embody is destroyed, and their entrepreneurial skills and other forms of locationspecific human capital are forever lost tothe community. However, other studies (Carden etal., 2009) have pointed out that there isalso apositive effect of modern retail investments on social capital as usually a modern retailer is located in new commercial center, where there are social interactions on a larger scale possible. Moreover, ingeneral prices inmodern retailers are lower and these stores offer alarger variety ofproducts, hereby effective reducing the time and money that has tobe devoted tobasic consumption. This may have apositive impact onsocial capital inacommunity. For example, assume you like fishing with friends, anactivity which increases social capital. Onthe other hand, you also have tobuy groceries and cook, activities which have noimpact onsocial capital. The emergence ofamodern retailer may increase social capital intwo ways. First, incase that the modern retailer sells fishing material and the price of fishing material decreases, there is a direct impact on social capital asyou are now able tobuy more fishing equipment. Second, even when the modern retailer does not sell fishing equipment, but only food, you will spend less money onbasic food consumption (because oflower food prices inthe modern retailer), such that you have more money available tospend onfishing equipment. Third, shopping inthe modern retailer reduces the time that you spend on buying food, which means that it increases the time that you can potentially spend onfishing. In summary, in the literature both empirically and theoretically the impact of modern retail investments on social capital is still unclear. Goetz and Ruspasingha (2006) find a negative impact ofmodern retailers onsocial capital asthey find that social capital stocks were lower both in communities in which new Wal-Mart stores were built and in communities that already had aWal-Mart store atthe beginning ofthe 1990s decade. However, using the same econometrical model with only slightly different specifications and data, Carden etal. (2009) indicate that they were unable to find a systematic positive relationship between the presence of a Wal-Mart and social capital. We have no evidence on the net effect of investments on social capital in the countries under consideration ingeneral and noclear evidence either way came out from our interviews.
79
Box 10: impaCt onsoCial Cohesion Opportunities Lower prices inmodern retailers and more convenient shopping (only one shop) could lead toan increase inthe money and time available for social capital producing products/ activities. Threats There could beadecline insocial cohesion due tothe disappearance oftraditional shops assmall, local shops and their owners play animportant role insocial relationships, norms and trust that were build upin acommunity Recommendations tomaximize positive and mitigate adverse impact onsocial cohesion Modern retail investments and the associated decline in small local shops may have a negative impact on social interactions within a region. However, modern retailers are usually located inanew commercial center which inmost cases accommodates facilities.where. individuals.can.gather.tohave.adrink.orameal. Ifthese facilities are not provided bythird parties, modern retailers can themselves set upabar orarestaurant. Inaddition, modern retailers can provide for example benches inthe entrance hall and outside the store, where older (and younger) individuals can sit down without being obliged tobuy acoffee orameal.
80
6. References
Asfaw, A.(2008), Does Supermarket Purchase Affect the Dietary Practices ofHouseholds? Some Empirical Evidence from Guatemala, Development Policy Review, Vol. 26, pp. 227243. Basker, E.(2005), Job Creation orDestruction? Labor-Market Effects ifWal-Mart Expansion, Review ofEconomics and Statistics, Vol. 87, pp. 174183. Bellemare, M. and D. Stifel (2009), The Welfare Effects of Contract Farming: Evidence from Madagascar, Working Paper. Binkley, J.K. and J.M. Connor (1998), Grocery Market Pricing and the New Competitive Environment, Journal ofRetailing, Vol. 74, pp. 273294. Binkley, J.,Canning, P.,Dooley, R.and J.Eales (2002), Consolidated markets, brand competition, and orange juice prices, Agriculture Information Bulletin No. 74706, USDA-ERS. Boselie, D., Hensen, S. and D. Weatherspoon (2003), Supermarket Procurement Practices in Developing Countries: Redefining the Roles of the Public and Private Sector, American Journal ofAgricultural Economics, Vol. 85(5), pp. 11551161. Bushnakova, M.(2003), Bulgaria Retail Food Sector 2003, GAIN Report No. BU3015, USDA Foreign Agriculture Service, Washington, DC. Burt, S.(2006), Retailing inEastern Europe Emerging countries within the European Union, InZentes, J.(ed.), Handbuch Handel, Gabler, pp. 139159. Carden, A.,Courtemanche, C.and J.Meiners (2009), Does Wal-Mart Reduce Social Capital?, Public Choice, Vol. 138, pp. 109136. Clarke, R., Davies, S. Dobson, P., and M. Waterson (2002), Buyer Power and Competition inEuropean Food Retailing, Edward Elgar Publications, 220p. Competition Commission (2000), Supermarkets: Areport onthe supply ofgroceries from multiple stores inthe United Kingdom,. Available online atwww.competition-commission.org.uk Cooper, D.(2003), Findings from the Competition CommissionsInquiry into Supermarkets, Journal ofAgricultural Economics, Vol. 54(1), pp. 127143. Cotterill, R.W. (1986), Market power in the Retail Food Industry: Evidence from Vermont, Review ofEconomics and Statistics, Vol. 68, pp. 379386. Cotterill, R.W. (1999), Continuing Concentration in Food Industries Globally: Strategic Challenges toan Unstable Status Quo, Food Marketing Policy Center Research Report 49, University ofConnecticut, Department ofAgricultural and Resource Economics. Cotterill, R. W. and C.D. Harper (1995), Market Power and the Demsetz Quality Critique: AnEvaluation for Food Retailing, Res. Report No. 29. Storrs, CN: Food Marketing Policy Center, Univ. ofConnecticut.
81
Dawson, J. and J. Henley (2010), Internationalisation of Hypermarket Retailing in Poland, Journal ofEast-West Business, Vol. 5(4), pp. 3752. DHaese, M., and G. Van Huylenbroeck (2005) The Rise of Supermarkets and Changing Expenditure Patterns of Poor Rural Households: Case Study in the Transkei Area, South Africa, Food Policy, Vol. 30, pp. 97113. Dobson, P.W., Clarke, R., Davies, S. andM. Waterson (2001), Buyer power and its impact oncompetition inthe food retail distribution sector ofthe European Union, Journal ofIndustry, Competition and Trade, Vol. 1(3), pp. 247281. Dolan, C.and J.Humphrey (2000), Governance and Trade inFresh Vegetables: The Impact ofUK Supermarkets onthe African Horticulture, Journal ofDevelopment Studies, Vol. 37, pp. 147176. Dries, L.and J.F.M.Swinnen (2004), Foreign Direct Investment, Vertical Integration and Local Suppliers: Evidence from the Polish Dairy Sector, World Development, Vol. 32, pp. 15251544. Dries, L.,Reardon, T.and J.F.M.Swinnen (2004), The Rapid Rise ofSupermarkets inCentral and Eastern Europe: Implications for the Agrifood and Rural Development, Development Policy Review, Vol. 22(5), pp. 525556. Dries, L.and T.Reardon (2005), Central and Eastern Europe: Impact ofFood Retail Investments on the Food Chain, London: FAO Investment Center EBRD Cooperation Program, Report Series no. 6, February. Dries, L.,Germenji, E.,Noev, N.and J.F.M.Swinnen (2009), Farmers, Vertical Coordination, and the Restructuring of Dairy Supply Chains in Central and Eastern Europe, World Development, Vol. 37, pp. 17421758. Einarsson, A. (2007), The Retail Sector in the Nordic Countries: A Comparative Analysis, Bifrst Journal ofSocial Science, Vol. 1, pp. 3142. Emongor, R.A.(2007), The Impact ofSouth African Supermarkets onAgricultural and Industrial Development in the Southern African Development Community, Ph.D. Dissertation, University ofPretoria. Euromonitor (2010), Various Reports. Eurostat Online Database, 2010. Faiguenbaum, S.,Berdegu, J.A.and T.Reardon (2002), The Rapid Rise ofSupermarkets inChile: Effects onDairy, Vegetable and Beef Chains, Development Policy Review, Vol. 20, pp. 459471. FAO stat, 2010. Farina, E.and T.Reardon (2000), Agrifood Grades and Standards inthe Extended Mercosur: Their Role inthe Changing Agrifood System, American Journal ofAgricultural Economics, Vol. 82, pp. 11701176. Farina, E.,Gutman, G.,Lavarello, P.,Nunes, R.and T.Reardon (2005), Private and Public Milk Standards inArgentina and Brazil, Food Policy, Vol. 30, pp. 302315.
82
Figui, M.and P.Moustier (2009), Market Appeal inan Emerging Economy: Supermarkets and Poor Consumers inVietnam, Food Policy, Vol. 34, pp. 210217. Fulponi, L. (2007), The Globalization of Private Standards and the Agri-food System, In Swinnen, J.F.M. (ed.), Global Supply Chains, Standards, and the Poor, Wallingford, Oxfordshire: CABI Publishing. Goetz, S. and A. Rupasingha (2006), Wal-Mart and Social Capital, American Journal ofAgricultural Economics, Vol. 88(5), pp. 13041310. Goldman, A.(1974), Outreach ofConsumers and the Modernization ofUrban Food Retailing inDeveloping Countries, Journal ofMarketing, Vol. 38(4), pp. 816. Gorton, M., Esanu, C., Deaconescu, D., Buckwell, A. and S. Davidova (1998), Impediments toEfficiency inthe Agro-Food Chain: Romania, Report presented toOECD, Wye College. Gorton, M.and J.White (2007), Transformation and Contracting inthe Supply Chains ofthe Former Soviet Union: Evidence from Armenia, Georgia, Moldova, Ukraine and Russia, in Swinnen, J.F.M. (ed.), Global Supply Chains, Standards and the Poor, Wallingford, Oxfordshire: CABI Publishing. Gorton, M., Sauer, J. and P. Supatpongkul (2009), Investigating Thai Shopping Behavior: Wet Markets, Supermarkets and the Big Middle, Paper at the International Association ofAgricultural Economists Conference, Bejing, China, August 1622, 2009 Gow, H. and J.F.M. Swinnen (1998), Up- and Downstream Restructuring, Foreign Direct Investments, and Hold-up Problems in Agricultural Transition, European Review ofAgricultural Economics, Vol. 25(3), pp. 331350. Gutman, G.(1997), Transformaciones recientes enla distribuciondealimentos enArgentina, Buenos Aires, SecretaradeAgricultura, Ganadera, Pesca yAlimentacion (SAGPyA) and Instituto Interamericano deCooperacionAgrcola (IICA). Grytsenko, S.(2011), EBRD finances retail operators inKazakhstan: 80million loan toexpand Metro Groupsoperations, News message EBRD website, 19May 2011. Hall, L., Schmitz, A. and J. Cothern (1979), Beef wholesale-retail marketing margins and concentration, Econometrica, Vol. 46, pp. 195300. Henson, S.,Loader, R.,Swinbank, A.,Bredahl, M.and N.Lux (2000), Impact ofSanitary and Phytosanitary Measures onDeveloping Countries, Working Paper, The University ofReading. Hicks, M.J.and K.Wilburn (2001), The Locational Impact ofWal-Mart Entrance: APanel Study ofthe Retail Trade Sector inWest Virginia, Review ofRegional Studies Vol. 31(3), pp. 305313. Ho, S.-C. (2005), Evolution versus Tradition in Marketing Systems: The Hong Kong FoodRetailing Experience, Journal ofPublic Policy and Marketing, Vol. 24, pp. 9099. Hu, D.,Reardon, T.,Rozelle, S.,Timmer, P.and H.Wang (2004), The Emergence ofSupermarkets with Chinese Characteristics: Challenges and Opportunities for ChinasAgricultural Development, Development Policy Review, Vol. 22, pp. 557586.
83
Humphrey, J., McCulloch, N. and M. Ota (2004), The Impact of European Market Changes onEmployment inthe Kenyan Horticulture Sector, Journal ofInternational Development, Vol. 16, pp. 6380. Karasiewicz, G.and J.Nowak (2010), Looking back atthe 20years ofretailing change inPoland, The International Review ofRetail, Distribution and Consumer Research, Vol. 20(1), pp. 103117. Kaufman, P.R. and C.R. Handy (1989), Supermarket prices and price differences: City, firm, and store-level determinants, Technical Bulletin No. 1776. Washington DC: Commodity Economics Division, Economic Research Service, USDepartment ofAgriculture. Key, N.and D.Runsten (1999), Contract farming, Smallholders, and Rural Development inLatin America: the Organization ofAgroprocessing Firms and the Scale ofOutgrower Production, World Development, Vol. 27, pp. 381401. King, R.(1988), Retail Food Distribution inaSocialist State: the Polish Experience, InKaynak,E., Transnational Retailing, Berlin, pp. 239248. Kinsey, J.(2003), Emerging Trends inthe New Food Economy: Consumers, Firms and Science, OECD paper, Paris. Krasney, T.(1992), Retail inCzechoslovakia, International Journal ofRetail and Distribution Management, Vol. 20(6), pp. 3033. Lamm, R.M. (1981), Prices and concentration in the food retailing industry, The Journal ofIndustrial Economics, Vol. 30(1), pp. 6778. Lloyd, T., McCorriston, S., Morgan, W., Rayner, A. and H. Weldegebriel (2009), Buyer Power inUK Food Retailing: AFirst-Pass Test, Journal ofAgricultural & Food Industrial Organization, Vol. 7(1): Article 5. Lopez-Claros, A.and S.Zahidi (2005), WomensEmpowerment: Measuring the Global Gender Gap, World Economic Forum, Geneva. Maertens, M. and J.F.M. Swinnen (2009a), Trade, Standards, and Poverty: Evidence from Senegal, World Development, Vol. 37, pp. 161178. Maertens, M.and J.F.M.Swinnen (2009b), Gender and Modern Supply Chains inDeveloping Countries, Licos Discussion Paper 231/2009. Maertens, M., Colen, L. and J.F.M. Swinnen (2010), Globalization and Poverty in Senegal: AWorst Case Scenario?, European Review ofAgricultural Economics, Forthcoming. Marion, B.W.,Heimforth, K.and W.Bailey (1993), Strategic groups, competition, and retail food prices In: R.W.Cotterill (ed.) Competitive strategy analysis inthe food system. Boulder, CO: Westview Press, pp. 179199. Martinez, A.(2006), Elmercado tradicional defrutas yjitomate enla Ciudad deMexico: elcaso de los tianguis y los mercados p ublicos, Report to USAID; MSU, Rimisp, and Colegio deMexico.
84
Michalak, W.Z. (2001), Retail in Poland: An Assessment of Changing Market and Foreign Investment Conditions, Journal ofRegional Science, Vol. 24(3), pp. 485504. Minten, B.and T.Reardon (2008), Food Prices, Quality and Qualityspricing inSupermarkets versus Traditional Markets inDeveloping Countries, Review ofAgricultural Economics, Vol. 30(3), pp. 480490. Minten, B., Randrianarison, L. and J.F.M. Swinnen (2009), Global Retail Chains and Poor Farmers: Evidence from Madagascar, World Development, Vol. 37, pp. 17281741. Minten, B., Reardon, T. and R. Sutradhar (2010), Food Prices and Modern Retail: The Case ofDelhi, World Development, forthcoming. Miyata, S.,Minot, N.and D.Hu (2009), Impact ofContract Farming: Linking Small Farmers, Packers and Supermarkets inChina, World Development, Vol. 37(11), pp. 17281741. Moustier, P.,Dao, T.A.,Hoang, A.B.,Vu, B.T.,Figuie, M.and T.G.T.Phan (2006), Supermarkets and the Poor in Vietnam, Malica (Markets and Ag Linkages for Cities in Asia) and M4P (Making MarketsWork Better for the Poor), Hanoi. Natawidjaja, R., Reardon, T., Shetty, S. Noor, T.I., Perdana, T., Rasmikayati, E., et al. (2007). Horticultural producers and supermarket development in Indonesia. UNPAD/MSU/World Bank. World Bank report no. 38543. The World Bank, Washington D.C. Neven, D.,Reardon, T.,Chege, J.and H.Wang (2006), Supermarkets and Consumers inAfrica: The Case ofNairobi, Kenya, International Food Agricultural Marketing, Vol. 18, pp. 10323. Newmark, C.M.(1990), Anew test ofthe price-concentration relationship ingrocery retailing, Economic Letters, Vol. 33, pp. 369373. Nowak, J. (1991), Marketing of crop products in Poland, Journal of International Food and Agribusiness Marketing, Vol. 3(2), pp. 6984. OECD (1994), Creating Rural Indicators for Shaping Territorial Policy, OECD, Paris. OECD (2000a), Review ofAgricultural Policies: Bulgaria, OECD, Paris. OECD (2000b), Review ofAgricultural Policies: Romania, OECD, Paris. OECD (2005), Regions atthe glance, OECD, Paris. Oponowicz, K. and A. Chmielecka (2008), Research Report: Situation of Female Employees in Super-and Hypermarkets, Report within the project Improvement of work conditions of women employed in chains of large area retail institutions1 in Poland, Online available athttp://www.eurofound.europa.eu/ewco/2010/02/PL1002029I.htm. Pingali, P. (2007), Westernization of Asian Diets and the Transformation of Food Systems: Implications for Research and Policy, Food Policy, Vol. 32, pp. 281298. PMR Publications (2008), Private Label inPoland 2008, PMR Publications.
85
Reardon, T.,Timmer, C.P.,Barrett, C.B.and J.Berdegu (2003), The Rise ofSupermarkets inAfrica, Asia, and Latin America, American Journal ofAgricultural Economics, Vol. 85, pp. 11401146. Reardon, T.and J.A.Berdegu (2006), The Retail-Led Transformation ofAgrifood Systems and its Implications for Development Policies, Working Paper for Rimisp-Latin American Center for Rural Development. Reardon, T., Henson, S. and J.A. Berdegu (2007a), Proactive fast-tracking Diffusion ofSupermarkets inDeveloping Countries: Implications for Market Institutions and Trade, Journal ofEconomic Geography, Vol. 7, pp. 399431. Reardon, T., Berdegu, J. A., Echanove, F., Cook, R., Tucker, N., Martinez, A., et al. (2007b), Supermarkets and Horticultural Development in Mexico: Synthesis of findings and recommendations toUSAID and GOM, Report submitted byMSU toUSAID/Mexico and USDA, Washington D.C. Reardon, T. Henson, S. and A. Gulati (2010), Links Between Supermarkets and Food Prices, Diet Diversity and Food Safety inDeveloping Countries, InHawkes, C.,Blouin, C.,Henson, S., Drager, N. and L. Dub (eds.) Trade, Food, Diet and Health: Perspectives and Policy Options, Blackwell Publishing, West Sussex, UK. Rodriguez, E.,Berges, M.Casellas, K.,DiPoala, R.,Lupin, B.,Garrido, L.and N.Gentile (2002), Consumer Behaviour and Supermarkets inArgentina, Development Policy Review, Vol. 20, pp. 429439. Rozelle, S.and J.F.M.Swinnen (2004), Success and Failure ofReform Insights from the Transition ofAgriculture, Journal ofEconomic Literature, Vol. 42, pp. 404456. Seitz, H., 1992, Retailing in the East: An Overview, International Journal of Retail and Distribution Management (Special Issue onEastern Europe), Vol. 20(6), pp. 410. Sexton, R.J.,Zhang, M.and J.A.Chalfant (2005), Grocery Retailer Behavior inPerishable Fresh Produce Procurement, Journal ofAgricultural & Food Industrial Organization (special issue). Shopping Monitor Central and Eastern Europe 2009. Sobel, R.S.and A.M.Dean (2008), Has Wal-Mart Buried Mom and Pop? The Impact ofWalMart onSelf-Employment and Small Establishments inthe United States, Economic Inquiry, Vol. 46(4), pp. 676695. Stender, S.; Dyerberg, J. and A. Astrup (2006), Consumer protection trough a legislative ban onindustrially produced trans fatty acids infoods inDenmark, Scandinavian Journal ofFood and Nutrition, Vol. 50(4), pp. 155160. Suryadarma, D.,Poesoro, A.,Akhmadi, Budiyati, S., Rosfadhila, M. and A. Suryahadi (2010), Traditional Food Traders in Developing Countries and Competition from Supermarkets: Evidence from Indonesia, Food Policy, Vol. 35, pp. 7986. Swinnen, J.F.M.(ed.) (2007), Global Supply Chains, Standards and the Poor, CABI publishing, Oxford.
86
Swinnen, J.F.M.and T.Vandemoortele (2008), The Political Economy ofNutrition and Health Standards inFood Markets, Review ofAgricultural Economics, Vol. 30, pp. 460- 468. Tessier, S,Traissac, P.,Maire, B.,Bricas, N.,Eymard-Duvernay, S.,ElAti, J.and F.Delpeuch (2008), Regular Usres of Supermarkets in Greater Tunis have a Slightly Improved Diet Quality, Journal ofNutrition, Vol. 138, pp. 768774. Thailand Development Research Institute (2002), The Retail Business inThailand: Impact ofthe Large Scale Multinational Corporation Retailers, Thailand Development Research Institute, Bangkok. Van Herck, K., Noev, N. and J.F.M. Swinnen (2010), Institutions, Exchange and Growth: Evidence from Bulgarian Agriculture onthe Impact ofHold-ups and Contract Innovations, Paper presented at114th Seminar ofthe EAAE on Structural Change inAgriculture, April 1516, 2010, Berlin, Germany. Wang, H, Dong, X., Rozelle, S., Huang, J. and T. Reardon (2009), Producing and Procuring Horticultural Crops with Chinese Characteristics: The Case of Northern China, World Development, Vol. 37(11), pp. 17911801. Warning M.and N.Key (2002), The Social Performance and Distributional Impact ofContract Farming: AnEquilibrium Analysis ofthe Arachide deBouche Program inSenegal, World Development, Vol. 30(2), pp. 255263. Weatherspoon, D.D.,Cacho, J.and R.Christy (2001), Linking Globalization, Economic Growth and Poverty: Impacts ofAgribusiness Strategies onSub-Saharan Africa, American Journal ofAgricultural Economics, Vol. 83, pp. 722 -729. Weatherspoon, D.D.and T.Reardon (2003), The Rise ofSupermarkets inAfrica: Implications for Agrifood Systems and the Rural Poor, Development Policy Review, Vol. 21, pp. 333Wilk, W.(2006), Western European Retail Chains inthe Czech Republic, Poland and Slovakia: Similarities and Differences, Miscellanea Geographica, Vol. 12, pp. 205214. World Bank (1991), Romania: The Challenge ofTransition, Washington, DC. World Bank (2005), When the Market Comes toYou OrNot. The Dynamics ofVertical Coordination in Agro-Food Chains in Europe and Central Asia, Report, The World Bank, Washington D.C. Xiang, T.,Huang, J.,Kancs, d. and J.F.M.Swinnen (2009), Standards Driven Rural Development: AGeneral Equilibrium Model with Market Imperfections, Contributed paper atthe International Association ofAgricultural Economists Conference, Bejing, China, August 1622, 2009. Zamora, M.(2005), Supermarkets, Dairy, and Potatoes, inEcuador, Report tothe Regoverning Markets Program. Zinkhan, G.,Fontenelle, S.M.and A.L.Balazs (1999), The Structure ofSo Paulo Street Markets: Evolving Patterns ofRetail Institutions, Journal ofConsumer Affairs, Vol. 33, pp. 326.7
87
7. Appendix
7.1 Regional Change in Population
Figure:.Population.Change.inBulgaria.(%;.20022008)

88
Figure:.Population.Change.inRomania.(%;.20022008)

89
Figure:.Population.Change.inPoland.(%;.20022008)

90
Table:.Regional.change.inpopulation.inBulgaria.(20022008;.%) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Change. inpopulation 3,18 9,25 12,06 10,34 9,24 8,48 7,42 5,53 4,24 6,74 5,20 6,05 7,02 2,62 0,34 4,78 3,36 -5,74 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Change. inpopulation 3,16 0,56 4,35 7,61 3,56 0,80 5,28 5,16 3,08 6,39 7,17 3,44 1,35 4,89 4,74 7,64 4,10
91
Table:.Regional.change.inpopulation.inPoland.(20022008;.%) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki
Piotrkowski
Change. inPopulation 0,33 2,35 4,87 0,70 1,17 1,70 2,75 1,30 0,93 1,02 1,05 1,36 4,38 5,74 1,53 0,18 3,43 2,25 0,22 2,33 1,99 1,93 1,09 1,43 3,01 3,21 3,98 2,68 0,28
NUTS. code PL31 PL311 PL312 PL314 PL315 PL32 PL323 PL324 PL325 PL326 PL33 PL331 PL332 PL34 PL343 PL344 PL345 PL41 PL411 PL414 PL415 PL416 PL417 PL418 PL42 PL422 PL423 PL424 PL425
Region
Lubelskie
Change. inpopulation 1,61 2,06 2,65 0,06 2,16 0,32 1,78 0,66 1,11 0,34 1,83 1,37 2,56 1,39 0,22 2,79 1,36 1,09 0,63 0,30 3,18 0,15 1,32 8,18 0,36 0,42 0,89 1,91 2,47
Bialski
Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski Przemyski Rzeszowski Tarnobrzeski Swietokrzyskie Kielecki Sandomierskojedrzejowski Podlaskie Bialostocki Lomzynski Suwalski Wielkopolskie Pilski Koninski Miasto Poznan
Sieradzki
Skierniewicki Mazowieckie Ciechanowskoplocki Ostroleckosiedlecki
Miasto Warszawa
Radomski Warszawskiwschodni Warszawskizachodni Malopolskie
Miasto Krakw
Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski
Kaliski
Leszczynski Poznanski
Zachodniopomorskie
Koszalinski Stargardzki
Gliwicki
Katowicki Sosnowiecki Tyski
Miasto Szczecin
Szczecinski
To be continued
92
Table:.Regional.change.inpopulation.inPoland.(20022008;.%).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region
Lubuskie Gorzowski
Change. inPopulation 0,05 0,45 0,35 1,07 1,23 1,96 0,94 3,08 2,70 2,75 2,27 3,06
NUTS. code PL61 PL613 PL614 PL615 PL62 PL621 PL622 PL623 PL63 PL631 PL633 PL634 PL635
Region
Kujawsko-Pomorskie Bydgosko-Torunski Grudziadzki Wloclawski WarminskoMazurskie
Change. inpopulation 0,17 0,44 0,05 0,85 0,16 0,61 0,22 0,15 1,50 0,32 1,61 7,84 1,45
Zielonogrski
Dolnoslaskie Miasto Wroclaw
Jeleniogrski
LegnickoGlogowski Walbrzyski Wroclawski Opolskie Nyski Opolski
Elblaski
Olsztynski Elcki Pomorskie Slupski Trojmiejski Gdanski Starogardzki
93
Table:.Regional.change.inRomania.(2007;.inhabitants.per.km) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor Bistrita-Nasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Population. density 1,40 1,39 1,63 0,49 0,19 1,71 2,43 3,06 1,04 3,09 0,12 1,20 1,58 1,16 0,14 0,57 1,06 1,94 1,75 1,64 0,01 2,04 1,61 3,45 2,87 0,82 2,19 2,68 1,04 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad CarasSeverin Hunedoara Timis Population. density 2,71 1,45 2,12 2,02 3,40 1,87 2,20 6,89 1,27 0,37 7,57 3,30 2,98 1,96 4,46 4,74 2,54 1,60 1,32 3,10 5,24 1,70
94
7.2 Regional Population Density

Figure:.Population.density.inBulgaria.(in2007;.number.ofinhabitants.per.km)

95
Figure:.Population.density.inRomania.(in2007;.number.ofinhabitants.per.km)

96
Figure:.Population.density.inPoland.(in2007;.number.ofinhabitants.per.km)

97
Figure:.Urban-Rural.Classification.inBulgaria

98
Figure:.Urban-Rural.Classification.inRomania

99
Figure:.Urban-Rural.Classification.inPoland

100
Table:.Regional.population.density.inBulgaria.(2007;.inhabitants.per.km) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Population. density 69,0 49,1 37,5 44,7 56,4 64,4 37,9 62,5 60,0 66,2 90,7 51,9 46,3 68,5 120,0 43,2 58,1 52,2 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Population. density 57,0 54,1 58,8 42,8 69,3 104,2 918,8 36,5 51,1 58,2 49,2 69,3 118,2 47,5 66,2 40,4 48,9
101
Table:.Regional.population.density.inPoland.(2007;.inhabitants.per.km) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki
Piotrkowski Sieradzki
Population. density 121,9 140,9 2594,7 170,1 100,5 80,4 91,9 145,4 80,5 62,0 3292,3 108,3 145,2 169,2 215,5 2312,7 163,8 138,8 231,8 176,4 378,6 175,4 275,5 472,0 292,2 573,4 2042,6 402,6 405,1
Region
Lubelskie Bialski Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski Przemyski Rzeszowski Tarnobrzeski Swietokrzyskie Kielecki Sandomierskojedrzejowski Podlaskie
Population. density 86,5 51,8 70,4 169,3 87,8 117,5 86,9 92,3 170,6 137,6 109,3 155,1 74,8 59,3 98,2 47,0 44,5 113,3 62,9 101,6 2156,3 115,3 90,0 111,7 73,9 57,0 55,0 1359,0 58,9
Skierniewicki
Mazowieckie Ciechanowskoplocki Ostroleckosiedlecki Miasto Warszawa Radomski Warszawskiwschodni Warszawskizachodni Malopolskie
Bialostocki
Lomzynski Suwalski Wielkopolskie Pilski Koninski
Miasto Krakw
Krakowski Nowosadecki
Oswiecimski
Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski Gliwicki Katowicki Sosnowiecki Tyski
Miasto Poznan
Kaliski Leszczynski Poznanski Zachodniopomorskie Koszalinski
Stargardzki
Miasto Szczecin
Szczecinski
To be continued
102
Table:.Regional.population.density.inPoland.(2007;.inhabitants.per.km).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region
Lubuskie Gorzowski
Population. density 72,1 62,4 79,6 144,5 2166,0 104,4 129,3 163,6 82,9 110,7 96,3 122,7
Region
KujawskoPomorskie Bydgosko-Torunski Grudziadzki Wloclawski WarminskoMazurskie Elblaski Olsztynski Elcki Pomorskie Slupski Trojmiejski Gdanski Starogardzki
Population. density 115,0 259,9 77,8 94,4 59,0 70,8 59,3 44,7 120,3 58,5 1807,1 110,2 92,5
Zielonogrski
Jeleniogrski
LegnickoGlogowski Walbrzyski Wroclawski Opolskie
Nyski Opolski
103
Table:.Regional.population.density.inRomania.(2007;.inhabitants.per.km) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor Bistrita-Nasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Population. density 93,70 81,0 80,2 60,0 105,2 82,1 84,7 63,9 74,7 60,8 112,0 60,7 49,4 87,5 78,8 103,1 111,3 94,0 153,0 98,0 83,9 87,4 90,8 82,3 81,5 108,6 141,9 49,4 83,2 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad Caras-Severin Hunedoara Timis Population. density 98,7 95,7 65,7 135,1 83,7 67,1 177,5 73,2 1271,9 10504,8 187,3 80,0 98,8 68,6 62,6 89,4 72,9 61,0 60,1 38,9 67,4 78,4
104
7.3. Regional Age Structure

Figure:.Share.ofthe.population.older.than.65years.inBulgaria.(%,.2009)

105
Figure:.Share.ofthe.population.older.than.65years.inRomania.(%,.2009)

106
Figure:.Share.ofthe.population.older.than.65years.inPoland.(%,.2009)

107
Table:.Share.ofthe.population.older.than.65years.inBulgaria.(2007;.%) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Share. ofthe. population 17,4 21,2 23,9 22,7 19,6 20,5 21,3 18,4 18,7 21,6 18,2 15,8 17,7 15,9 15,2 16,5 15,9 17,3 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Share. ofthe. population 17,0 15,5 15,8 20,0 18,3 16,3 14,8 20,1 15,0 21,5 21,1 17,3 17,5 19,3 16,3 16,2 15,6
108
Table:.Share.ofthe.population.older.than.65years.inPoland.(2007;.%) NUTS. code Region Poland Ldzkie Miasto Ldz Ldzki

Piotrkowski Sieradzki Skierniewicki Mazowieckie Ciechanowskoplocki
Share. ofthe. population 13,5 15,0 17,0 14,5 13,5 14,3 14,7 14,5 13,3 13,8 17,2 13,5 12,4 13,1 13,5 15,2 12,9 12,1 13,8 13,3 14,0 14,8 13,7 12,7 14,8 14,4 15,1 14,3 11,8
Region
Lubelskie Bialski Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski
Share. ofthe. population 14,3 14,0 14,9 13,6 14,9 13,0 13,1 13,4 13,1 12,6 14,8 14,3 15,7 14,8 14,1 16,3 13,7 11,9 10,9 11,9 14,7 12,4 11,1 10,0 12,3 12,0 11,6 14,8 10,5
PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C
Przemyski
Rzeszowski Tarnobrzeski Swietokrzyskie Kielecki Sandomierskojedrzejowski Podlaskie Bialostocki Lomzynski Suwalski Wielkopolskie Pilski Koninski Miasto Poznan Kaliski Leszczynski
Ostroleckosiedlecki
Miasto Warszawa
Radomski
Warszawskiwschodni Warszawskizachodni Malopolskie
Miasto Krakw
Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski Gliwicki Katowicki Sosnowiecki Tyski
Poznanski
Zachodniopomorskie Koszalinski Stargardzki Miasto Szczecin Szczecinski
To be continued
109
Table:.Share.ofthe.population.older.than.65years.inPoland.(2007;.%).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region
Lubuskie Gorzowski
Share. ofthe. population 11,8 11,6 11,9 13,5 15,7 13,2 11,3 14,5 11,7 14,2 13,7 14,5
Region
Kujawsko-Pomorskie Bydgosko-Torunski
Share. ofthe. population 12,6 13,2 12,0 12,4 11,8 11,7 11,8 11,9 12,2 11,1 15,6 9,8 10,8
Zielonogrski Dolnoslaskie
Miasto Wroclaw
Grudziadzki
Wloclawski WarminskoMazurskie Elblaski Olsztynski Elcki Pomorskie Slupski Trojmiejski
Jeleniogrski
Gdanski
Starogardzki
110
Table:.Share.ofthe.population.older.than.65years.inRomania.(2007;.%) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor BistritaNasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Share.ofthe. population 14,9 13,9 14,3 13,0 15,4 12,5 12,1 15,4 14,0 15,2 12,7 13,8 14,2 15,3 12,8 14,5 14,0 16,6 12,9 15,5 14,4 15,1 14,8 16,7 18,2 12,3 13,5 13,5 16,3 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad Caras-Severin Hunedoara Timis Share.ofthe. population 16,7 14,6 17,2 15,0 19,0 16,4 16,0 21,6 14,1 14,2 13,9 16,3 16,8 13,7 16,4 17,0 17,0 14,4 15,1 14,9 14,7 13,3
111
7.4. Regional GDP per capita

Figure:.GDP.per.capita.inBulgaria.(in2007)

112
Figure:.GDP.per.capita.inRomania.(in2007)
Source: Eurostat.Online Database

113
Figure:.GDP.per.capita.inPoland.(in2007)

114
Table:.Regional.GDP.per.capita.inBulgaria.(2007) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte GDP.. per.capita 3800 2600 2200 2200 3100 2400 2900 2700 2500 3300 3000 2300 2100 3200 4300 2300 2400 2500 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali GDP.. per.capita 3100 3500 2000 2000 3600 6200 8500 3700 2700 2800 2600 2700 3000 2300 2800 2700 2200
115
Table:.Regional.GDP.per.capita.inPoland.(2007) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki Piotrkowski Sieradzki Skierniewicki Mazowieckie Ciechanowsko-plocki Ostrolecko-siedlecki Miasto Warszawa Radomski Warszawski-wschodni Warszawski-zachodni Malopolskie Miasto Krakw Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski Gliwicki Katowicki Sosnowiecki Tyski GDP.. per.. capita 8200 7500 9900 6900 6900 6000 6000 13100 8500 6100 24900 5900 6400 9400 7000 12900 5200 4600 6200 5000 8700 6900 7900 7200 6500 8800 11700 8700 11000 NUTS. code PL31 PL311 PL312 PL314 PL315 PL32 PL323 PL324 PL325 PL326 PL33 PL331 PL332 PL34 PL343 PL344 PL345 PL41 PL411 PL414 PL415 PL416 PL417 PL418 PL42 PL422 PL423 PL424 PL425 Region Lubelskie Bialski Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski Przemyski Rzeszowski Tarnobrzeski Swietokrzyskie Kielecki Sandomierskojedrzejowski Podlaskie Bialostocki Lomzynski Suwalski Wielkopolskie Pilski Koninski Miasto Poznan Kaliski Leszczynski Poznanski Zachodniopomorskie Koszalinski Stargardzki Miasto Szczecin Szczecinski GDP. per. capita 5500 4800 4900 6900 4800 5500 5000 4700 6000 5900 6300 6700 5600 6100 7100 5100 5600 8500 6500 6200 16500 6200 6800 9100 7300 6600 5200 10500 7100
To be continued
116
Table:.Regional.GDP.per.capita.inPoland.(2007).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region Lubuskie Gorzowski Zielonogrski Dolnoslaskie Miasto Wroclaw Jeleniogrski Legnicko-Glogowski Walbrzyski Wroclawski Opolskie Nyski Opolski GDP.. per.. capita 7200 7200 7300 8900 12100 6300 13800 6500 6800 6800 5400 7700 NUTS. code PL61 PL613 PL614 PL615 PL62 PL621 PL622 PL623 PL63 PL631 PL633 PL634 PL635 Region Kujawsko-Pomorskie Bydgosko-Torunski Grudziadzki Wloclawski Warminsko-Mazurskie Elblaski Olsztynski Elcki Pomorskie Slupski Trojmiejski Gdanski Starogardzki GDP. per. capita 7100 9200 5600 6100 6100 5700 6700 5200 8000 6400 11700 5600 6500
117
Table:.Regional.GDP.per.capita.inRomania.(2007) NUTS. code RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Region Romania Nord-Vest Bihor Bistrita-Nasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea GDP.. per. capita 5800 5600 5800 4700 7800 4100 4400 4800 5900 6400 7100 4700 4800 4900 6400 3700 4100 3100 4400 3500 3800 2500 4700 4300 3800 6800 4200 3900 3500 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad Caras-Severin Hunedoara Timis GDP.. per. capita 4800 6300 3000 4600 2800 3500 6000 3500 12800 13200 10300 4500 4500 6100 3800 3500 5000 6700 6600 4900 5600 8400
118
7.5. Regional Unemployment Rate

Figure:.Unemployment.inBulgaria.(in2008)

119
Figure:.Unemployment.inRomania.(in2008)

120
Figure:.Unemployment.inPoland.(in2008)

121
Table:.Regional.unemployment.rate.inBulgaria.(2008;.%) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Unemployment. rate 5,6 7,1 12,9 8,5 9,0 4,7 na 8,5 7,8 na 8,3 14,9 11,6 8,6 4,3 9,7 16,7 10,8 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Unemployment. rate 5,8 3,4 12,5 6,4 4,4 2,9 2,5 na na na 8,3 5,1 3,9 6,4 5,3 10,9 na
122
Table:.Regional.unemployment.rate.inPoland.(2008;.%) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki Piotrkowski Sieradzki Skierniewicki Mazowieckie Ciechanowskoplocki Ostroleckosiedlecki Miasto Warszawa Radomski Warszawskiwschodni Warszawskizachodni Malopolskie Miasto Krakw Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski Gliwicki Katowicki Sosnowiecki Tyski Unemployment. rate 6,7 6,5 8,2 7,7 6,2 4,6 6,0 9,5 6,5 4,6 10,0 4,3 4,4 6,2 5,4 6,3 7,3 6,5 5,2 6,6 6,6 4,4 6,6 8,7 6,6 6,7 7,7 4,4 6,7 NUTS. code PL31 PL311 PL312 PL314 PL315 PL32 PL323 PL324 PL325 PL326 PL33 PL331 PL332 PL34 PL343 PL344 PL345 PL41 PL411 PL414 PL415 PL416 PL417 PL418 PL42 PL422 PL423 PL424 PL425 Region Lubelskie Bialski
Chelmsko-zamojski Lubelski Pulawski
Unemployment. rate 8,8 7,9 7,4 9,8 10,0 8,2 8,5 8,5 6,1 9,8 8,8 9,8 7,6 6,4 5,9 6,6 7,3 6,1 7,1 8,4 3,3 7,4 4,8 4,1 9,5 11,8 11,6 6,1 8,2
Podkarpackie
Krosnienski Przemyski Rzeszowski Tarnobrzeski
Swietokrzyskie
Kielecki Sandomierskojedrzejowski Podlaskie
Bialostocki Lomzynski
Suwalski Wielkopolskie Pilski Koninski
Miasto Poznan
Kaliski Leszczynski Poznanski
Zachodniopomorskie Koszalinski
Stargardzki Miasto Szczecin Szczecinski
To be continued
123
Table:.Regional.unemployment.rate.inPoland.(2008;.%).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region
Lubuskie Gorzowski
Unemployment. rate 6,5 6,3 6,6 9,1 5,8 11,5 9,5 11,9 7,5 6,5 8,7 5,1
Region
Kujawsko-Pomorskie
Unemployment. rate 9,1 5,9 10,2 11,5 7,4 6,9 7,2 8,8 5,5 8,9 3,3 3,7 7,2
Bydgosko-Torunski
Grudziadzki Wloclawski WarminskoMazurskie Elblaski
Zielonogrski Dolnoslaskie
Miasto Wroclaw
Jeleniogrski
LegnickoGlogowski Walbrzyski
Olsztynski
Elcki Pomorskie Slupski Trojmiejski
Wroclawski
Opolskie Nyski Opolski
Gdanski
Starogardzki
124
Table:.Regional.unemployment.rate.inRomania.(2008;.%) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor Bistrita-Nasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Unemployment. rate 5,8 3,8 na 3,9 2,5 7,0 na 6,0 8,5 4,1 14,3 7,6 7,5 7,7 7,7 4,5 2,7 3,2 4,1 8,1 3,4 7,7 7,2 7,6 7,0 11,0 6,3 6,8 na NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad CarasSeverin Hunedoara Timis Unemployment. rate 6,8 3,2 11,0 7,0 na 7,4 10,2 6,0 3,4 3,1 na 6,5 5,0 14,9 6,2 2,4 10,8 5,7 8,0 7,4 7,4 na
125
7.6. Regional Car Ownership Data

Figure:.Number.ofcars.per.hundred.inhabitants.inBulgaria.(in2008)
*Data only available atthe NUTS2 level Source: Eurostat Online Database
126
Figure:.Number.ofcars.per.hundred.inhabitants.inRomania.(in2008)
127
Figure:.Number.ofcars.per.hundred.inhabitants.inPoland.(in2008)
128
Table.:.Car.ownership.inBulgaria.(in2008;.number.ofcars.per.hundred.inhabitants) NUTS.code BG31 BG32 BG33 BG34 BG41 BG42 Source: Eurostat Online Database Table.:.Car.ownership.inPoland.(in2008;.number.ofcars.per.hundred.inhabitants) NUTS.code PL11 PL12 PL21 PL22 PL31 PL32 PL33 PL34 PL41 PL42 PL43 PL51 PL52 PL61 PL62 PL63 Source: Eurostat Online Database Region Poland Ldzkie Mazowieckie Malopolskie Slaskie Lubelskie Podkarpackie Swietokrzyskie Podlaskie Wielkopolskie Zachodniopomorskie Lubuskie Dolnoslaskie Opolskie Kujawsko-Pomorskie Warminsko-Mazurskie Pomorskie Car.ownership 42 41 49 41 41 38 37 39 37 49 38 44 42 45 41 37 43 Region Bulgaria Severozapaden Severen tsentralen Severoiztochen Yugoiztochen Yugozapaden Yuzhen tsentralen Car.ownership 31 26 26 30 28 40 27
129
Table.:.Car.ownership.inRomania.(in2008;.number.ofcars.per.hundred.inhabitants) NUTS.code RO11 RO12 RO21 RO22 RO31 RO32 RO41 RO42 Source: Eurostat Online Database Region Romania Nord-Vest Centru Nord-Est Sud-Est Sud Muntenia Bucuresti Ilfov Sud-Vest Oltenia Vest Car.ownership 19 18 18 11 15 15 47 15 19
130
7.7. Regional Importance of the Agricultural Sector

Figure:.Share.ofagriculture.inGVA.inBulgaria.(in2007;.%)

131
Figure:.Share.ofagriculture.inGVA.inRomania.(in2007;.%)

132
Figure:.Share.ofagriculture.inGVA.inPoland.(in2007;.%)

133
Figure:.Share.ofagriculture.inemployment.inBulgaria.(in2007;.%)

134
Figure:.Share.ofagriculture.inemployment.inRomania.(in2007;.%)

135
Figure:.Share.ofagriculture.inemployment.inPoland.(in2007;.%)

136
Table:.Regional.share.ofthe.agricultural.sector.inGVA.inBulgaria.(2007;.%) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Agricultural. GVA 6,2 12,3 19,1 17,0 10,1 11,1 9,7 10,6 7,8 4,6 7,9 19,3 24,3 7,7 2,7 18,0 13,4 14,5 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Agricultural. GVA 7,5 5,8 12,1 15,8 6,0 2,3 0,3 6,6 13,5 6,8 15,8 10,5 6,4 13,2 11,2 13,8 25,1
137
Table:.Regional.share.ofthe.agricultural.sector.inGVA.inPoland.(2007;.%) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki Piotrkowski Sieradzki Skierniewicki Mazowieckie Ciechanowskoplocki Ostroleckosiedlecki Miasto Warszawa Radomski Warszawskiwschodni Warszawskizachodni Malopolskie Miasto Krakw Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki Bytomski Gliwicki Katowicki Sosnowiecki Tyski Agricultural. GVA 4,3 6,5 0,1 5,7 8,8 14,3 14,7 3,8 11,6 19,0 0,0 11,2 6,8 4,7 3,0 0,1 6,5 6,0 2,8 6,0 1,1 3,7 1,4 1,2 1,5 0,7 0,1 1,2 0,8 NUTS. code PL31 PL311 PL312 PL314 PL315 PL32 PL323 PL324 PL325 PL326 PL33 PL331 PL332 PL34 PL343 PL344 PL345 PL41 PL411 PL414 PL415 PL416 PL417 PL418 PL42 PL422 PL423 PL424 PL425 Region
Lubelskie Bialski Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski Przemyski Rzeszowski Tarnobrzeski Swietokrzyskie
Agricultural. GVA 7,9 13,9 11,2 3,6 8,6 3,5 3,9 5,7 2,7 2,8 6,3 3,3 12,1 10,7 5,3 17,0 14,5 6,5 12,6 11,5 0,1 9,4 10,2 6,1 4,2 5,7 9,4 0,1 5,0
Kielecki
Sandomierskojedrzejowski Podlaskie Bialostocki Lomzynski Suwalski Wielkopolskie Pilski Koninski Miasto Poznan
Kaliski
Leszczynski Poznanski Zachodniopomorskie Koszalinski
Stargardzki
Miasto Szczecin Szczecinski
To be continued
138
Table:.Regional.share.ofthe.agricultural.sector.inGVA.inPoland.(2007;.%).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region
Lubuskie Gorzowski
Agricultural. GVA 4,4 4,9 4,0 2,3 0,1 3,8 1,7 2,9 6,0 5,3 8,6 3,8
Region
KujawskoPomorskie Bydgosko-Torunski Grudziadzki Wloclawski WarminskoMazurskie Elblaski Olsztynski Elcki Pomorskie Slupski Trojmiejski Gdanski Starogardzki
Agricultural. GVA 5,9 1,5 10,7 9,3 7,8 7,6 6,8 11,0 2,9 6,5 0,1 5,0 5,3
Zielonogrski
Jeleniogrski
139
Table:.Regional.share.ofthe.agricultural.sector.inGVA.inRomania.(2007;.%) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor BistritaNasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Agricultural. GVA 6,5 8,7 9,4 11,6 5,0 10,7 12,9 10,4 8,4 8,4 5,2 18,3 12,7 9,7 5,7 10,4 6,6 17,4 5,5 9,7 16,9 12,7 8,0 10,6 10,0 4,3 8,5 8,5 13,9 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad Caras-Severin Hunedoara Timis Agricultural. GVA 8,0 6,2 12,0 15,7 11,0 10,2 3,5 9,2 0,3 0,0 2,7 7,3 6,2 5,9 8,0 9,7 8,2 7,6 7,4 14,4 5,3 6,9
140
Table:.Regional.share.ofthe.agricultural.sector.inemployment.inBulgaria.(2007;.%) NUTS. code BG31 BG311 BG312 BG313 BG314 BG315 BG32 BG321 BG322 BG323 BG324 BG325 BG33 BG331 BG332 BG333 BG334 Region Bulgaria Severozapaden Vidin Montana Vratsa Pleven Lovech Severen tsentralen Veliko Tarnovo Gabrovo Ruse Razgrad Silistra Severoiztochen Varna Dobrich Shumen Targovishte Agricultural. employment 19,7 27,3 22,4 35,8 23,2 24,9 28,0 27,8 26,6 20,2 20,2 29,6 50,9 20,9 11,1 24,8 26,3 38,8 NUTS. code BG34 BG341 BG342 BG343 BG344 BG41 BG411 BG412 BG413 BG414 BG415 BG42 BG421 BG422 BG423 BG424 BG425 Region Yugoiztochen Burgas Sliven Yambol Stara Zagora Yugozapaden Sofia (stolitsa) Sofia Blagoevgrad Pernik Kyustendil Yuzhen tsentralen Plovdiv Haskovo Pazardzhik Smolyan Kardzhali Agricultural. employment 24,7 21,2 26,2 52,1 13,1 9,8 2,3 27,5 24,5 33,2 18,8 23,6 20,0 25,8 28,1 22,0 32,2
141
Table:.Regional.share.ofthe.agricultural.sector.inemployment.inPoland.(2007;.%) NUTS. code PL11 PL113 PL114 PL115 PL116 PL117 PL12 PL121 PL122 PL127 PL128 PL129 PL12A PL21 PL213 PL214 PL215 PL216 PL217 PL22 PL224 PL225 PL227 PL228 PL229 PL22A PL22B PL22C Region Poland Ldzkie Miasto Ldz Ldzki
Piotrkowski
Agricultural. employment 14,7 17,8 0,7 14,3 23,4 31,0 34,3 11,1 23,1 31,8 0,4 25,6 15,4 12,9 16,1 0,8 29,0 28,1 12,5 25,6 3,2 7,8 6,4 3,4 2,7 1,5 0,2 2,9 3,0
Region
Lubelskie
Agricultural. employment 33,6 37,3 42,0 20,0 41,2 25,4 23,9 31,6 24,5 23,8 31,3 18,2 49,4 27,1 14,5 38,6 33,8 15,0 17,4 23,7 0,6 23,3 20,8 11,5 7,1 9,5 13,0 0,6 9,0
Bialski
Chelmsko-zamojski Lubelski Pulawski Podkarpackie Krosnienski Przemyski Rzeszowski Tarnobrzeski Swietokrzyskie Kielecki Sandomierskojedrzejowski Podlaskie Bialostocki Lomzynski Suwalski Wielkopolskie Pilski Koninski Miasto Poznan
Sieradzki
Skierniewicki Mazowieckie Ciechanowskoplocki Ostroleckosiedlecki Miasto Warszawa Radomski Warszawskiwschodni Warszawskizachodni Malopolskie
Miasto Krakw
Krakowski Nowosadecki Oswiecimski Tarnowski Slaskie Czestochowski Bielski Rybnicki
Kaliski
Leszczynski Poznanski Zachodniopomorskie Koszalinski Stargardzki
Bytomski Gliwicki
Katowicki Sosnowiecki Tyski
Miasto Szczecin
Szczecinski
To be continued
142
Table:.Regional.share.ofthe.agricultural.sector.inemployment.inPoland.(2007;.%).continuation NUTS. code PL43 PL431 PL432 PL51 PL514 PL515 PL516 PL517 PL518 PL52 PL521 PL522 Region Lubuskie
Gorzowski
Agricultural. employment 9,7 9,7 9,7 6,3 0,5 7,9 6,4 6,7 14,6 14,8 19,0 12,5
Region
KujawskoPomorskie Bydgosko-Torunski Grudziadzki Wloclawski WarminskoMazurskie
Agricultural. employment 17,1 5,1 26,9 25,0 13,0 14,0 10,5 17,5 8,6 12,4 0,9 16,0 14,7
Zielonogrski
Jeleniogrski
Elblaski
Olsztynski
Elcki
Pomorskie Slupski Trojmiejski Gdanski Starogardzki
143
Table:.Regional.share.ofthe.agricultural.sector.inemployment.inRomania.(2007;.%) NUTS. code Region Romania RO11 RO111 RO112 RO113 RO114 RO115 RO116 RO12 RO121 RO122 RO123 RO124 RO125 RO126 RO21 RO211 RO212 RO213 RO214 RO215 RO216 RO22 RO221 RO222 RO223 RO224 RO225 RO226 Nord-Vest Bihor Bistrita-Nasaud Cluj Maramures Satu Mare Salaj Centru Alba Brasov Covasna Harghita Mures Sibiu Nord-Est Bacau Botosani Iasi Neamt Suceava Vaslui Sud-Est Braila Buzau Constanta Galati Tulcea Vrancea Agricultural. Employment 30,3 27,3 25,5 37,5 17,9 25,4 40,0 31,5 17,0 27,6 12,1 23,1 16,0 17,7 10,9 48,7 44,5 41,9 45,3 55,7 45,0 61,8 32,4 40,6 37,6 29,1 17,1 42,8 38,1 NUTS. code RO31 RO311 RO312 RO313 RO314 RO315 RO316 RO317 RO32 RO321 RO322 RO41 RO411 RO412 RO413 RO414 RO415 RO42 RO421 RO422 RO423 RO424 Region Sud Muntenia Arges Calarasi Dmbovita Giurgiu Ialomita Prahova Teleorman Bucuresti Ilfov Bucuresti Ilfov Sud-Vest Oltenia Dolj Gorj Mehedinti Olt Vlcea Vest Arad CarasSeverin Hunedoara Timis Agricultural. Employment 39,0 24,1 61,3 23,6 51,3 62,3 24,5 55,9 1,1 0,4 7,6 42,2 44,3 34,2 46,9 50,4 33,6 18,6 20,1 21,6 13,9 19,5
144
7.8. Statistical Analysis: Principal Component Analysis

In order to visualize the differences between rural regions whithin one country, we perform aPrincipal Component Analysis (PCA) for each country. The central idea ofPCA isto reduce the dimensionality ofadata set consisting ofalarge number ofhighly correlated variables, while retaining asmuch aspossible ofthe variation present inthe dataset. This isachieved bytransforming the original data into anew set ofvariables, the principal compents (PC), which are uncorrelated and determined such that the first few retain most ofthe variation present inall original variables. The results ofour analysis show that for the three countries, the eight indicators that wepresented inthis report can besummarized intwo principal components: PC1 is strongly positively correlated with GDP per capita and negatively correlated with the importance of agriculture in the overall economy (share of agricultural labour and agricultural GVA). PC2 is strongly positively correlated with the share of persons older than 65 years and negatively correlated with the change in population (meaning regions where there is high outmigration have ahigh PC2 value). The figures below present the differences between rural regions inrespectively Bulgaria, Poland and Romania based onPC1 and PC2. In Bulgaria, we find that the mountanious regions close to the border with Romania, such asVidin (BG311) and Montana (BG312), are the worst off. GDP per capita islow inthese regions, while the agricultural sector isstill very important, there islarge migration out the region and the share ofold inhabitants ishigh. Incontrast, the regions close tothe Black Sea, Varna (BG331) and Burgas (BG341) are richer, have relatively unimportant agricultural sector, low migration from region (oreven in-migration) and alow share share ofolder inhabitants. Inaddition, there are also regions, such asSilistra (BG325) and Kardzhali (BG425), which are poor and have animportant agricultural sector but have low migration out the region and alow share ofolder inhabitants. InPoland, the regions, such asLomzynski (PL344), Bialski (PL311) and Chelmsko-Zamojski (PL312) that are close tothe border with Belarus and Ukraine, are the poorest regions and they also have the largest share ofolder inhabitants and migration out the region. Similar results hold for some regions inthe middle ofPoland, such asfor example Sieradzki (PL116), Skierniewicki (PL117) and Sandomiersko-jedrzejowski (PL332). The regions that are the best off are the regions close to the border with the Czech Republic, such as Opolski (PL522) and legnicko-Glogowski (PL516), and Bydgosko-Torunski (PL613) and, the rural regions in the predominantly urban voivodeship Slaskie (regions Czestochowski (PL224) and Bielski (PL225)). InRomania, the regions close tothe border with Bulgaria are the worst off (regions Giugiu (RO314), Calarasi (RO312) and Olt (RO414)). In contrast, regions located in central Romania (regions Sibiu (RO126) and Brasov (RO122)) and close tothe border ofHungary (Timis (RO424)) are the best off. Inaddition, also Constanta (RO223), the region close tothe Black Sea, isaregion which isrelatively well off.
145
Figure.:.Classification.ofthe.rural.regions.inBulgaria.based.onPCA
High Share of Older Inhabitants High Migration
BG311
BG321
BG322 BG414 BG415 BG412
BG312 BG314
BG315 BG313
BG344 BG323 BG422 BG343 BG424 BG334 BG332 BG325 BG324 BG423 BG342 BG333 BG413 BG341 BG421
Low Share of Older Inhabitants Low Migration
BG331
BG425
Low GDP per Capita High Importance Agricultural Sector
High GDP per Capita Low Importance Agricultural Sector
Source: Own Calculations based onEurostat Online Database Figure:.Classification.ofthe.rural.regions.inPoland.based.onPCA

Low Share of Older Inhabitants High Migration
PL344 PL332 PL312 PL311 PL345
PL315 PL117 PL116 PL324 PL517 PL224 PL522 PL228
High Share of Older Inhabitants Low Migration
PL323 PL331 PL343 PL521 PL122 PL623 PL615 PL314 PL128 PL326 PL515 PL325 PL621 PL614 PL216 PL121 PL423 PL217 PL622 PL422 PL214 PL416 PL215 PL432 PL414 PL431 PL635 PL631 PL518 PL411 PL417 PL425 PL129
PL22B
PL225
PL227 PL613 PL22C PL516
PL634 PL418

146
Figure.:.Classification.ofthe.rural.regions.inRomania.based.onPCA
Low Share of Older Inhabitants High Migration
RO314 RO414 RO222 RO413 RO221 RO315 RO116 RO212 RO216 RO226
RO411 RO415 RO423 RO316 RO121 RO214 RO422 RO225 RO313RO211 RO421 RO311 RO125 RO412 RO111 RO224 RO124 RO113
RO312
High Share of Older Inhabitants Low Migration
RO215
RO115 RO123
RO112 RO114 RO213
RO122 RO126 RO424 RO223
147
7.9. Foreign Direct Investment in the Grocery Sales

Table:.Share.ofthe.largest.retailers.intotal.grocery.sales.inPoland.(%) 2005 Jeronimo Martins (FDI) Carrefour Polska Spzoo (FDI) Tesco Polska Spzoo (FDI) Real Spzoo (FDI) Auchan Spzoo (FDI) ZKiP Lewiatan 94Holding SA Lidl Polska Spzoo (FDI) Eurocash SA (FDI) Others Source: Euromonitor (2010) Table:.Share.ofthe.largest.retailers.intotal.grocery.sales.inBulgaria.(%) 2005 Billa Bulgaria EOOD (FDI) Bolyari AD Van Holding AD (FDI) CBA Bulgaria AD (FDI) Kaufland Bulgaria EOOD (FDI) VPMarket Bulgaria EOOD (FDI) Metro Cash & Carry Bulgaria EOOD (FDI) Nova Familia 2007 EOOD (FDI) Others Source: Euromonitor (2010) Table:.Share.ofthe.largest.retailers.intotal.grocery.sales.inRomania.(%) 2005 Carrefour Romania (FDI) Hiproma SAReal Hypermarket Romania SRL (FDI) Kaufland Romania SCS (FDI) Cora Romania (FDI) Plus Discount Romania SCS (FDI) Rewe Romania SRL (FDI) Billa Romania SRL (FDI) SCMGV DistriHiper SARom Food SRL Others Source: Euromonitor (2010)
148
2006 5,2 3,6 4,6 2,7 3,9 3,1 1,7 1,7 73,5
2007 6,5 4,7 5,4 3,8 3,8 3,2 2 1,7 68,9
2008 8,4 5,1 5,4 3,7 4,2 3,4 2,3 1,8 65,7
2009 8,9 6,4 6 4,5 4,4 3,4 2,6 2,1 61,7
4,3 3,3 3,9 2,3 3,6 2,9 1,4 1,5 76,8
2006 4,1 2,2 2,7 2,8 1 1 1,8 84,4
2007 4,4 3,1 3,1 2,9 1,5 1,9 1,7 0,6 80,8
2008 5,2 3,6 3,7 3,2 2 2,3 1,7 0,7 77,6
2009 5,7 4,2 3,7 3,2 2,2 2,2 1,7 0,6 76,5
3,5 1,3 2,5 1,7 0,2 0,7 90,1
2006 5 0,9 2,1 2,7 0,9 2,1 2,7 83,6
2007 6,1 3,1 5,4 3,4 1,6 2,1 2,7 75,6
2008 6,8 4,9 5,2 3,7 2,9 2,2 2,5 71,8
2009 8,4 5,4 5 4,1 3,4 2,9 2,8 2,5 68
5,3 0 0,2 2 0,3 1,3 2,8 88,1
Please address comments and inquiries to:

Investment Centre Division Food and Agriculture Organization of the United Nations (FAO) Viale delle Terme di Caracalla - 00153 Rome - ITALY E-mail: Investment-Centre@fao.org Web site: www.fao.org/tc/tci FAO Web site: www.fao.org/investment Report 11/003 EBRD-POL/BUL/ROM

Social Impact of Discount Food Retail

Cargado por

Información del documento

Descripción original:

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Social Impact of Discount Food Retail

Cargado por

Copyright:

Formatos disponibles

FAO INVESTMENT CENTRE / EBRD COOPERATION PROGRAMME

REPORT SERIES N. 16 JUNE 2011

Poland, Bulgaria and Romania

FAO INVESTMENT CENTRE/EBRD COOPERATION PROGRAMME

REPORT SERIES N.16 JUNE 2011

Poland, Bulgaria and Romania

ACRONYMS AND ABBREVIATIONS

Demographic and Socio-Economic Analysis of Regions

The Transformation of the Retail Industry & Recent Developments

Impact on Consumers, Producers, Traditional Retail and Rural Society

2. Demographic and Socio-Economic Analysis of Regions in Poland, Bulgaria and Romania

2.1 Demographic Analysis

Source: Eurostat Online Database

2.2 Socio-Economic Analysis

*Constant 2007 prices Source: Eurostat Online Database

2.3 Statistical Analysis

Population density Population > 65years 0.520

0.748 0.104 0.046

Cars per 100 inhabitants Agricultural GVA Agricultural employment

Source: Own Calculations based onEurostat Online Database

Population density Population > 65years 0.551 0.313

GDP per capita

Unemployment 0.021 0.424 0.391

1.000 0.171 0.153

Cars per 100 inhabitants Agricultural GVA Agricultural employment

Source: Own Calculations based onEurostat Online Database

Change inpopulation 1.000 0.146

Population density Population > 65years 0.327

0.563 0.434 0.132

Cars per 100 inhabitants Agricultural GVA Agricultural employment

Source: Own Calculations based onEurostat Online Database

3. The Transformation of the Retail Industry

3.1 Stages in the Transformation of the Retail Industry

Transition Low Domestic Low Low Cities

Globalization High Foreign High High Everywhere

High Domestic Low Low

Vertically integrated units Farm and downstream combined

State trading companies

Sale of produce at peasant markets

Source: Dries etal. (2004) Figure.3:.Share.ofthe.modern.retail.sector.and.reform.progress.inEastern.Europe

4. Recent Developments and Effects

Hypermarkets Supermarkets Discounters Other

Source: Euromonitor (2010) Figure.6:.Share.intotal.grocery.sales.ofdifferent.modern.retail.formats.inRomania.(20042009)

9% 8% Hypermarkets Supermarkets Discounters

Source: Euromonitor (2010)

Source: Euromonitor (2010)

Source: Euromonitor (2010)

Poland Bulgaria Romania

Source: Fulponi (2007) Figure.10:.Development.ofPrivate.Labels.inthe.Low,.Medium.and.Premium.Price.Brackets.atTesco,. Carrefour.and.E.Leclerc.inPoland,.March.2006.and.March.2008

2008 Medium Premium

Source: PMR Publications (Private Label inPoland 2008)

5. Impacts on Consumers, Producers, Traditional Retail and Rural Society

5.1 Impact on Consumers

1995 2899 3306 3065

2001 2773 3410 3276

2007 2766 3421 3455

3622 3490 2938

1995 122 119 84 92 110 87

2001 93 113 100

2007 96 114 108

Source: FAO Figure.11B:.Share.ofdifferent.food.categories.inthe.total.fat.consumption.inPoland.(%)

2005 25,5 5,4 2,2 18,9 170,9

2006 26,4 5,5 2,3 19,5 176,6