Está en la página 1de 10

Telematics and Informatics 32 (2015) 79–88

Contents lists available at ScienceDirect

Telematics and Informatics
journal homepage:

The effect of ICT development on economic growth and energy
consumption in Japan
Hazuki Ishida ⇑
Faculty of Economics, Doshisha University, Karasuma Higashi-iru, Imadegawa-dori, Kyoto City, Japan

a r t i c l e

i n f o

Article history:
Received 26 January 2014
Accepted 11 April 2014
Available online 23 April 2014
Energy demand

a b s t r a c t
A strongly held belief in Japan is that information and communication technologies (ICT)
contribute to both a reduction in energy use and an increase in economic growth. As this
assertion is presently unproven, the purpose of this analysis is to estimate the long-run
relationship between ICT, energy consumption, and economic growth in Japan. Using an
autoregressive distributed lag (ARDL) bounds testing approach, we estimate two different
multivariate models corresponding to the production function and the energy demand
function, both including ICT investment as an explanatory variable, over the period
1980–2010. The results reveal the presence of a long-run stable relationship, not only
for the production function, but also for the energy demand function. However, in the
production function, the long-run coefficient estimate for ICT investment is statistically
insignificant, unlike the coefficients for labor, stock, and energy. In the case of the
energy demand function, the coefficients for GDP, energy price, and ICT investment are
statistically significant. The results also indicate that the long-run ICT investment elasticity
of energy consumption is 0.155. On this basis, we conclude that while ICT investment
could ceteris paribus contribute to a moderate reduction in energy consumption, but not
to an increase in GDP.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction
From the end of World War II until the early 1970s, the average annual growth rate of energy consumption in Japan was
higher than the nation’s remarkable economic growth rate. However, following the 1973 oil crisis, the Japanese economy
was obliged to make some efforts to decrease its energy intensity (the ratio of energy consumption to GDP), and largely succeeded in slowing the growth rate of energy consumption without an accompanying decrease in GDP. Nonetheless, the fact
remains that energy consumption has continued to grow in the long term, with Japanese energy use rising by nearly 50%
from 1973 to 2010. In other words, the decrease in energy intensity has not revealed the concomitance of a reduction in
energy use and economic growth.
However, there is some hope that information and communication technologies (ICT) have the potential to solve the
dilemma of this ‘‘win–win’’ situation, that is, producing more output from less energy. For example, the Global Information
Infrastructure Commission (GIIC, 2008) emphasizes the role of ICT in protecting the environment without any sacrifice in
economic output. Similarly, the Japanese government also relies strongly on ICT, not only to provide the economy with

⇑ Tel./fax: +81 75 251 3656.
E-mail address:
0736-5853/Ó 2014 Elsevier Ltd. All rights reserved.

However. However. capital. Ishida / Telematics and Informatics 32 (2015) 79–88 market competitiveness through technical innovation and global entrepreneurship. on this point. In the worst-case scenario. Gilling. Currently. in the monetary approach. for which there are two main ways: nonmonetary and monetary. (1993) investigated the causal relationship between ICT development and economic growth and find evidence of a bidirectional relationship in the US. Nevertheless. as measured in monetary terms. energy). Section 2 reviews the literature on related studies. Zachariadis. Importantly. ICT has also been one of the fastest growing components of total investment. it is preferable to employ monetary measures to represent ICT development. Cronin et al. we discuss the empirical results and Section 6 concludes. 1980. Note especially that even a developed country such as Japan does not have a long series of data on ICT investment or stock when measured in monetary terms. we develop a conventional energy demand function. the use of different countries over dissimilar periods. which could be potentially substitutable for each other (Stern. 2001. teledensity (the number of fixed telephones per hundred inhabitants. ICT and economic growth Studies on the relationship between ICT development and economic growth began in the 1960s. Stern (2000). there is insufficient evidence showing whether ICT investment can actually lead to a win–win lower energy use-economic growth situation. First. 2000. Considering the problems associated with the use of a bivariate model pointed out by Stern (1993). for the most part these studies have been based on static data. buildings. such as Japan. the Japanese government has emphasized the importance of introducing policies and institutions to stimulate additional investment in ICT.1. Developed by Pesaran and Shin (1999). In Section 5. but also to enable transport. ICT investment represents about 25% of total investment in Japan. that is. 1975. Other than the problem of measuring ICT development discussed earlier. 1983). 2012). . In other words. these studies do not provide sufficient information about the actual contribution of ICT development to economic growth. the bivariate approach may also lead to erroneous conclusions given it ignores the substitution effects between ICT and other factors of production (e. most existing studies of the ICT–growth nexus employ bivariate analysis. and industry to reduce their energy consumption. Section 4 reports the results of the respective econometric analyses. Gross. while data on teledensity are usually easy to obtain.. Section 3 describes the methods and data used in this analysis. Saunders et al. the bivariate approach is subject to some criticism as it effectively ignores any bias arising from omitted variables. Therefore.. The remainder of the paper is organized as follows. Studies on the ICT–growth nexus using time-series analysis first began in the 1990s. for which they found a unidirectional causality running from GDP to ICT investment. the ARDL bounds cointegration approach has been used extensively in recent years to examine the existence of long-run relationships among time series variables. (1991). For this reason. 2005). Alternatively. implying the rejection of the ICT-led growth hypothesis.g. 2. having almost doubled over the last 15 years. They therefore say nothing about the long-run equilibrium among the various variables included. previous studies leave much room for improvement in the methods used. cointegration analysis has become a widely used technique for investigating long-run association among the time-series variables. there have been many similar studies in both developed and developing countries. and so on. As a nonmonetary variable. While most of these have confirmed the existence of a causal relationship running from ICT to growth (Madden and Savage. 1998. In the case of the ICT–growth nexus.80 H. and GDP in Japan. Dutta. also including ICT investment as an explanatory variable. we employ a multivariate approach in our analysis. Second. Pesaran et al. For the ICT– energy nexus. For example. Although these results have generally found a positive correlation between ICT and economic growth (Jipp. 2007. The aim of this paper is to explore the nexuses between ICT–GDP and ICT–energy using an autoregressive distributed lag (ARDL) bounds test approach. labor. one disadvantage of using such data is that they do not reflect the investment in software that has become so important in developed countries. (2001). A brief literature review 2. Stern. including mobile phone subscribers and internet broadband subscribers) is usually used to investigate the ICT–growth nexus. 1993. there is even the possibility of a deadweight loss stemming from excessively high ICT investment. Shinjo and Zhang (2004) is the only study that has conducted time-series analysis to investigate the relationship between ICT development. Cronin et al. we specify a neoclassical aggregate production function including energy consumption and ICT investment as explanatory variables. Since the seminal work by Engle and Granger (1987). there are a few cases where there is no evidence of causality from ICT to GDP (Shiu and Lam. Subsequently. the average annual growth rate of ICT investment in Japan is relatively low when compared with the US or many European Union countries. For the ICT-growth nexus. Hardy. if available. and so it is inappropriate to suggest that ICT investment is an absolute good for society. Until now. Chakraborty and Nandi.. It is natural that empirical results sometimes diverge because of the different methodologies employed. with software accounting for about 50% of total ICT investment and equipment the remainder. 2008). when investigating the effect of ICT development on economic growth in a developed country. Chu et al. One important problem is to select a variable to represent the level of ICT development. 2003. there is no precedent for studying the ICT–growth nexus using an ARDL model. 1963. One of the reasons for preferring this approach is that it is more robust and performs better with small sample sizes than many other techniques. only two variables (ICT and GDP) are specified in the estimations. either ICT investment or ICT stock is typically used. and have since been conducted mainly in the US.

ði ¼ 1. whereas ICT investment in the service sector and most manufacturing sectors increases electricity consumption. 2008.. and Z is ICT investment. Hilty and Ruddy. 2. we develop an aggregate production model where not only labor and the capital stock but also energy are treated as production factors. Methods and data 3. Szklo and Schaeffer. it is widely acknowledged that ICT contributes to ‘‘dematerialization’’ in broad industrial sectors through the substitution and optimization of material or energy consumption (Hilty. 2013. Accordingly. Huber and Mills (1999) estimate that the energy requirements for computer. . Fuchs.2. 2000. (1). no studies using time-series techniques have examined the long-run relationships between ICT and energy. 2010). Roth et al. Ghali and El-Sakka. The magnitude of these direct impacts remains controversial. 4Þ correspond to the elasticity of GDP with respect to labor. 3.. 2004. Bomhof et al. use and disposal of ICT equipment. there is the possibility that ICT development in fact stimulates the demand for energy through the globalization of markets and the distribution of production forms induced by the growth of telecommunications networks (referred to as the ‘‘induction effect’’ by Hilty (2008)). Similarly. ICT and energy consumption Several reasons exist to explain how ICT development can lead to a significant reduction in energy consumption. the impact of ICT on energy use as a whole hinges on whether these negative (direct and indirect) effects offset the associated substitution and optimization effects. 2004. To start with. 2002. Assuming that ICT investment also affects output. we obtain the following equation: ln Y ¼ a0 þ a1 ln L þ a2 ln K þ a3 ln E þ a4 ln Z ð2Þ where the coefficients ai .. Barratt (2006) argues that education and training about environmental management and technology can be achieved through distance teaching via the Internet. 2006. 2007. these are just some of the examples. Wang et al. Yuan et al. energy economists (Stern. office equipment.. E is the energy input. The direct impact relates to the production. the production function is defined as follows: Y ¼ f ðL. 2011). macro-level empirical investigation of the relationship between ICT development and energy consumption is less common. 2010. e. Ishida / Telematics and Informatics 32 (2015) 79–88 81 2. seemingly in spite of its environmental importance. 2004) have increasingly emphasized the crucial role of energy as a production factor. K is the capital stock. K. Toffel and Horvath (2004) emphasize the potential energy savings accruing from reading newspapers and decreasing the need for business travel through the use of wireless information technologies. Moreover. Sadorsky (2012) examines the impact of ICT on electricity consumption in 19 emerging countries. ICT development could affect energy use in the economy via two channels: direct and indirect impacts. increasing ICT investment in the US would increase energy use. For example. 2008. (2002) suggest that office and communication equipment in the US consume less than 3% of delivered electricity nationwide. In addition.. 3. Although the results of these studies imply that the negative effects of ICT investment on energy consumption offset the positive effects (if any). not total energy (including electricity) consumption. Elsewhere. Production model Although conventional neoclassical production models consider labor and capital stock as the only inputs. energy consumption. 2003. mainly because of concerns about the negative side effects of ICT development (Rejeski. (2007) investigate the effects of ICT investment on electricity consumption in South Korean industries and conclude that ICT investment in some manufacturing sectors brings about a reduction in electricity consumption. 2008). Cho et al. The estimation of the indirect impact of ICT on energy consumption is even more complicated. while Forge (2007) points out the unsustainability of ICT development. 2000. Compared with studies of the ICT–growth nexus. (2007) and Sadorsky (2012) investigated the relationship between ICT and electricity consumption. and finds a positive relationship between ICT (as measured by teledensity) and electricity consumption. Taking the loglinear form of Eq. and the Internet range from 8% to 13% of total US electricity consumption.1. and conclude that while Japan could conserve energy by increasing investment in ICT. To start with. After all. Conversely. and ICT investment. Takase and Murota (2004) conducted one of the few studies on this subject. E. Yi and Thomas. 2007. As Hilty (2008) argues. Chiabai et al. the capital stock. However. complexity in the relationship between ICT development and economic activity leads to deep uncertainty about the direct impact of ICT development on energy consumption.g. Oh and Lee. respectively. Soytas and Sari. L is the labor input. many recent studies have incorporated energy as a factor additional to labor and capital stock in the production function (Stern. Cho et al. ZÞ ð1Þ where Y is real GDP. After considering the importance of energy as a factor of production. Ghali and El-Sakka.H. power consumption in data centers. In fact. some researchers are skeptical about the idea that ICT development automatically leads to a substantial reduction in energy usage. They analyze the effects of ICT investment on energy consumption in the US and Japan. While there certainly could be substitution and optimization effects of ICT with energy use. and as Laitner (2002) points out. For instance. we should note that Cho et al.

which is denoted as FðYjL. 1996) and the KPSS (Kwiatowski et al.4. As the ARDL bounds test for cointegration is applicable regardless of whether the variables are Ið0Þ or Ið1Þ. To conduct the F-test. against H1 : h1 – h2 – h3 – h4 – 0. inferring that a cointegrating relationship exists between the variables. Energy demand model It is customary to model energy demand as a function of income and the energy price. ARDL bounds cointegration test We employ the ARDL bounds testing approach (Pesaran and Shin. Ishida / Telematics and Informatics 32 (2015) 79–88 3. An F-test with a nonstandard asymptotic distribution is used to test the null hypothesis. The data for total labor hours.2. 1992) test to confirm the stationarity of the variables. Data We employ annual data for Japan over the period 1980 to 2010. Labor and Welfare. 2004). The data for total domestic primary energy consumption in joules are from the Agency for Natural Resources and Energy. We obtain the data for the prices of imported fossil fuels . Similarly for Eq.. K. 1999. the subscript t denotes the time period. 1992. applying ECM version of ARDL model will be a useful way of establishing cointegration (Kremers et al. we assume that ICT investment also influences energy demand. 3.5. (3). (1) is: D ln Y t ¼ k0 þ q1 X k1i D ln Y ti þ i¼1 þ q5 X q2 q3 q4 X X X k2i D ln Lti þ k3i D ln K ti þ k4i D ln Eti i¼0 i¼0 i¼0 k5i D ln Z ti þ l1 ln Y t1 þ l2 ln Lt1 þ l3 ln K t1 þ l4 ln Et1 þ l5 ln Z t1 þ et ð5Þ i¼0 where D is the first-difference operator. then the null hypothesis is rejected. If the calculated F-statistic is lower than the lower-bound critical value. We estimate the models developed above using the unrestricted error correction model (UECM) approach. (6). The data for real GDP data and the capital stock are from the Cabinet Office.82 H. where we indirectly estimate the long-run relationship and directly estimate the short-run equilibrium. are from the Ministry of Internal Affairs and Communications and the Ministry of Health. Bahmani-Oskooee and Nasir. Government of Japan.. 3. E. Pesaran et al. If the calculated F-statistic falls within the critical band between the upper and lower bound. Y. In addition to this. we obtain: ln E ¼ b0 þ b1 ln P þ b2 ln Y þ b3 ln Z ð4Þ where the coefficients bi . P. ZÞ. we specify the price of fossil fuels as the energy price. In such an inconclusive case.. against H1 : l1 – l2 – l3 – l4 – l5 – 0.. coal and natural gas. GDP and ICT investment. The first set of critical values refers to the lower-bound critical values. then the result is inconclusive. ði ¼ 1. Considering that fossil fuels (oil. The UECM representation corresponding to Eq. we apply the Dickey–Fuller generalized least squares (DF-GLS) unit root test (Elliott et al. calculated by multiplying the number of workers and working time per capita. Taking the log-linear form of Eq. 2. ZÞ ð3Þ where P is the energy price.3. then the null hypothesis cannot be rejected. respectively. If the calculated F-statistic exceeds the upper-bound critical value. and define the aggregate energy demand model as follows: E ¼ gðP. H0 : h1 ¼ h2 ¼ h3 ¼ h4 ¼ 0. coal and natural gas) are major sources of the energy supply in Japan. Stationary test Given that the conventional augmented Dickey–Fuller (ADF) test has low power with a short span of data. 3. ZÞ. The UECM representation corresponding to Eq. which is denoted as FðEjY. being a quantity-weighted average of the imported prices (given Japan imports almost all its fossil-fuel needs) of oil. the stationary tests are only to ensure that there are no Ið2Þ variables. meaning there is no evidence of cointegration between the variables. 2001) to examine the long-run equilibrium relationships between the variables. we use the two sets of critical F-values provided by Narayan (2005) as our sample size is only 30. (3) is: D ln Et ¼ c0 þ q1 X c1i D ln Eti þ i¼1 þ h4 ln Z t1 þ et : q2 X q3 X q4 X i¼0 i¼0 i¼0 c2i D ln Y ti þ c3i D ln P ti þ c4i D ln Z ti þ h1 ln Et1 þ h2 ln Y t1 þ h3 ln Pt1 ð6Þ The null hypothesis of no cointegration between the variables in Eq. (5) is H0 : l1 ¼ l2 ¼ l3 ¼ l4 ¼ l5 ¼ 0. and the second set of critical values comprises the upper-bound critical values. 3Þ refer to the elasticity of energy consumption with respect to the energy price. and et is a disturbance term assumed normally distributed white noise.

1 4. Empirical results Table 1 provides the results of the DF-GLS and KPSS tests.183⁄ 0. Ishida / Telematics and Informatics 32 (2015) 79–88 Table 1 Results of the DF-GLS and KPSS tests. L. the stationarity tests indicate that ICT stock is Ið2Þ.457(1)⁄⁄ 2. Table 2 Bounds F-test for cointegration for the production model.445 Lag lengths (in parenthesis) are determined by AIC. no trend 1.187 0. not Ið2Þ. (Z—Y. the energy price.565(0)⁄⁄ 4.and short-run coefficients of both the production function and the energy demand function in the long run based on the associated ARDL model and ECM framework. we use ICT investment for the following reasons. The results for the KPSS test show that all of the variables are nonstationary in levels except for labor (with constant and no trend) and the capital stock (with constant and no trend). In contrast.806(0)⁄⁄ 4. In particular. E.654⁄ 0. As shown. from the Institute of Energy Economics. However.868(1) 0.390 0.83 H. The use of ICT investment thus enables us to better compare our results with those of other researchers. Table 3 provides the results of the ARDL bounds test for the four-variable model (corresponding to the energy demand model). Z) Z) Z) Z) E) F-statistics Cointegration 4.967(1) 1.612⁄ 0. Additionally.697(0)⁄ 3.360(1) 1.352 0. and calculate the real price using the GDP deflator.126 0.184⁄ 0. Japan. ⁄ ⁄⁄ .168⁄ 0. Next.434(1) 1.179⁄ 0.072 0.127 0. However. with the first-differences.154. the DF-GLS test results indicate that all variables are nonstationary in levels. The data for ICT investment in real terms are from the Ministry of Internal Affairs and Communications. . no trend Constant and trend Constant. E. the results of the two tests confirm that all variables in question are either Ið0Þ or Ið1Þ.176⁄ 0. I (1) = 5.958 8. We estimate the ARDL model using the 1 If we consider the consistency of the variables used in the neoclassical production model.119 5.899(0)⁄⁄ 0. L. K.331(2) 1.160⁄ 0.164(0)⁄ 2. First. the bounds test indicates that there is no cointegration when real GDP. (L—Y. K.620(1) 1.271(1) 0. K. L.883(0)⁄⁄ 3. namely. we check the coefficient of the ECM term to determine the long-run relationship among the variables (see Table 4). E. (E—Y. The results confirm that cointegration is only present when energy consumption is the dependent variable. Meanwhile.166(0) 1. Variables DF-GLS ln E ln Y ln P ln L ln K ln Z Dln E Dln Y Dln P Dln L Dln K Dln Z KPSS Constant and trend Constant. These results indicate cointegration among the variables. (K—Y.271(3)⁄ 3. The null hypothesis of the DF-GLS test states that the series in question has a unit root. Second. 1992.088 0. we proceed to estimate the long. The results in Table 2 indicate that the calculated F-value for the equation with real GDP as the dependent variable falls between the lower and upper bounds of critical values at the 5% level of significance.376(0)⁄⁄ 2.402(1) 3.540.547 4. but stationary in first-differences. Critical value bounds are obtained from Narayan (2005). whereas the null hypothesis of the KPSS test is that the variable is stationary.783 Inconclusive No Yes Yes No 5% Critical value bounds: I (0) = 4.392 0. the KPSS test reveals that all of the variables are stationary.455⁄ 0.122 0.173(1)⁄⁄ 3.. following Kremers et al.089 0. the calculated F-values for the remaining two equations specifying the capital stock and energy consumption as dependent variables are also higher than the upper-bound critical value at the 5% significance level.730(0)⁄⁄ 0. Since the result of the cointegration test is inconclusive. existing studies (especially concerning Japan) display a strong tendency to use ICT investment in analyzing the relationship between ICT development and economic growth and/or energy consumption. and ICT investment are the dependent variables.579(3)⁄ 5.462(1) 3.217 3.657⁄ 0.427 0. the results show that there is no cointegration when labor and ICT investment serve as dependent variables. Model F F F F F (Y—L.318 0. Indicate significance at 5% and 1% level respectively. Table 2 details the results of the ARDL bounds test for the five-variable model (corresponding to the production model). K.286 0.881(2) 1. it might be more appropriate to use the stock of ICT rather than ICT investment. signifying a long-run energy demand function.211(0) 1.

301[0.089 0. (Z—Y.029 0.0.419[0. Variable Coefficient Standard error T-statistics Dependent variable: ln E Long-run results ln Y ln P ln Z 1.004 2.717[0.706[0.001] 0.980. Variable Coefficient Standard error T-statistics Dependent variable: ln Y Long-run results ln L ln K ln E ln Z 1.453[0. I (1) = 5.096[0.558[0.989 371.047] 3.309 0.715] 2.870 3. Model F F F F (E—Y.002] 0.034 0. Bracket represents probability values.1.771] Short-run results Dln L Dln K Dln E Dln Z ECT (1) 0.845 0.004] 2.067 0.1.369[0.002] R-bar-square F-statistic DW RSS Serial correlation (v-square) RESET (v-square) Normality (v-square) Heteroscedasticity (v-square) 0.84 H.190] 0.057 0.002] 0.018 0.135 0.597 0.061[0.112[0.799] 1. Z) Z) Z) P) F-statistics Cointegration 8. Ishida / Telematics and Informatics 32 (2015) 79–88 Table 3 Bounds F-test for cointegration for the energy demand model.096 0.415[0.2[0.244[0.345 0.088] 3.0) model.002 0.081[0.043 3.043[0.155 0. Table 5 Estimated long-run and short-run coefficients of the energy demand function based on ARDL (1.091] The optimal order of lags in the model are selected based on AIC.369 0. Bracket represents probability values.426] 2. P.489[0.000] 1.233] The optimal order of lags in the model are selected based on AIC.053] 3.006 0.138 0.196 0.0) model.016 0.106 3.005] 2.027 0.096 4.442 0. P.098 0.540[0.336 0.0.776] 1.592[0. (P—Y. E.111] 1.004] 3.998 1751.287[0.349 0.000] 2.000] R-bar-square F-statistic DW RSS Serial correlation (v-square) RESET (v-square) Normality (v-square) Heteroscedasticity (v-square) 0.294[0.075 0.006 0.766] 4.716 1.500 Yes No No No 5% Critical value bounds: I (0) = 4.671 1.000] 2.683.449[0. . E.204[0.075 3.434 0.097 0.061 0.862[0.956[0.052] Short-run results Dln Y Dln P Dln Z ECT (1) 0.013 0. Table 4 Estimated long-run and short-run coefficients of the production function based on ARDL (1. (Y—E. Critical value bounds are obtained from Narayan (2005).781[0.7[0.1.214 0.134] 2.

Plot of CUSUM and CUSUMSQ for the production function. Ishida / Telematics and Informatics 32 (2015) 79–88 85 Fig.and short-run estimation of the energy demand model. the short-run elasticity of ICT investment on energy demand is also statistically significant at the 5% level. As shown. More importantly. Moreover. The long-run elasticity of capital stock on real GDP is also positive and statistically significant at nearly the 5% level. Akaike information criterion (AIC) to select the optimum lag order. 1. The estimated coefficient of 0:442 suggests that the convergence to equilibrium following a shock to real GDP takes a little over two years.H. In line with theory. the long-run elasticity of ICT investment on real GDP is statistically insignificant. the lagged error correction term in the energy demand model is negative and statistically significant at the 1% level. Table 5 provides the results for the long. The estimated value of the coefficient is 0:155. the long-run elasticities of labor and energy consumption on real GDP are positive and statistically significant at the 1% level. these results indicate that ICT investment does not have a positive effect on economic growth in either the long or short run. Table 4 details the results for the production model. Together. Lastly.052). Both are significant at the 1% level. and the long-run price elasticity of energy demand is negative. The coefficient of 0:369 suggests that the convergence to equilibrium following a shock in energy demand takes place within three years. As expected. 2004). supporting cointegration among the variables (Bahmani-Oskooee and Nasir. The shortrun elasticity of ICT investment on real GDP as estimated by the error correction representation of the ARDL model is also statistically insignificant at the 10% level. even at the 10% level. However. the long-run impact of ICT investment on energy demand is negative and almost statistically significant (p-value = 0. indicating that a 1% increase in ICT investment results in a 0:155% decrease in energy demand. . the long-run income elasticity of energy demand is positive. the lagged error correction term in the production model is negative and statistically significant.

even without a corresponding increase in energy consumption. Needless to say. 1 and Fig. nonnormality. both of the estimated ARDL models (the production model and the energy demand model) pass the tests for serial correlation. Discussion One of the focuses of this paper is the impact of ICT development on energy demand in Japan. Plot of CUSUM and CUSUMSQ for the energy demand function. Another of our focuses relates to the latter. respectively. In other words. As shown in Table 4 and Table 5. energy intensity can be lowered through either ‘‘the same output with less energy’’ or ‘‘more output with the same energy’’. we test for the stability of the coefficients in the estimated models by using the cumulative sum (CUSUM) and the cumulative sum of squares (CUSUMSQ) stability tests. Ishida / Telematics and Informatics 32 (2015) 79–88 Fig. we note that our findings do not preclude the possibility of ‘‘more output with the same energy’’. but also by ICT investment. The empirical results confirm that aggregate energy demand in the long run is determined not only by the energy price and real GDP. These results . the estimated parameters for both models appear stable throughout our chosen sample period. ‘‘more output with the same energy’’ can be achieved if the input of labor and/or capital stock increase. 2 depict the plots for the stability tests for the production model and the energy demand model. 2. Fig. As shown. Although we find the impact of ICT investment on real GDP is not statistically significant. The negative elasticity of ICT development in the energy demand function implies that a ceteris paribus increase in ICT development will lower energy intensity (as measured by the ratio of energy consumption to GDP). and heteroscedasticity. the results of our empirical analysis based on the production model indicate that a ceteris paribus increase in ICT development will not affect economic growth.86 H. functional form misspecification. 5. Strictly speaking. In addition.

Dev. for ICT investment may not be too serious. Rev. but not ICT investment.. we find the long-run coefficient for ICT investment to be almost statistically significant at the 5% level. Forge. If we avoid the increase in energy consumption. Energy use and output growth in Canada: a multivariate cointegration analysis. C. M.. Chu. 17. 2006. 813–836.T.J.. Adopting the policy instruments to promote heavily ICT development might create a deadweight loss onto the Japanese economy. E. Cointegration and error correction: representation.. though the impact is small. A.. Energy Econ. D. 1987. 1991. Rubbelke. 529–535. Econometrica 55. for example. 2005. References Bahmani-Oskooee. 10. as before. S. Meeting lifelong learning needs by distance teaching – clean technology.. GIIC.K. 1996. the pursuit of economic growth would necessitate (more or less) an increase in the labor force. the results of our analysis imply that long-term economic growth would not be achieved without a continuous increase in either labor or energy consumption. Granger. Elliott. 2003. Strategies 52. 8. 2008. The owner (driver) of the cab could then produce the same amount of service with half the amount of fuel. telecommunications and growth in selected Asian countries: an economic analysis.M. F. The Japanese government should then not overestimate the potential of ICT development to realize sustainable growth with lower energy consumption. Dev. 2008. Chiabai. Although this could contribute to a decrease in the rate of unemployment.. 26. if we hope to achieve economic growth without an increase in energy consumption.. M. Powering down: Remedies fo unsustainable ICT... McGill University. 71–95. Colleran. Ishida / Telematics and Informatics 32 (2015) 79–88 87 imply that the role of ICT development on economic growth should not be overestimated.J. 906–915. P.giic. Telecommunications infrastructure and economic growth: an analysis of causality. Cleaner Product. The price and income elasticity of energy demand are statistically significant with the correct theoretically expected signs. Conversely.. Engle. Manage.. 2001. Given that the average Japanese worker works longer than average among the OECD countries.H. Nasir. and a four-variable model corresponding to the energy demand model. C. 3–21. Maurer. S. However.. Environ. http://www. and energy consumption. M... Lee. Stock. Policy 15. A. K. R.L. Oxley. 4730–4738. 2007. Telecommunications and growth: The contribution of telecommunications infrastructure investment to aggregate and sectoral productivity. This suggests that a technology used to improve energy productivity will not necessarily facilitate growth in labor productivity. R.. Again.J.. 14. 6. Privatization. Lewitzky. it is clear that a doubling of the working hours for the driver must accompany the latter. 1975.I.. however. Nandi.B. . Commun. E. El-Sakka. The implications of new information and communication technologies for sustainability. Commun. 251–276. ICT and causality in the New Zealand economy. defined as the percentage change in real GDP for a given percentage change in energy intensity. Overall. 677–690. Econ. This role for ICT may then be analogous to. J. Parker. both of which incorporate ICT investment as one of their explanatory variables. we conclude that ICT investment directly contributes to a moderate reduction in energy consumption.. If the magnitude of the rebound effect is small. B. mileage) with the same input of fuel. Telecommun. Telecommunications and economic development: inter-country comparisons of the catalytic effect of telephone services on development.. our findings imply that energy could substitute for labor and/or capital (at least to some extent).. Herbert. Telecommun. the direct impact and the induction effect may not be sufficiently serious to offset the anticipated energy savings from the decrease in energy intensity associated with stimulating ICT development. we develop two different multivariate models. Energy Policy 35.. C. 2004. The short-run impact of ICT investment on energy demand is also statistically significant. 77–96. L. 2009. comprising a fivevariable model corresponding to the production model. our result that ICT development does not directly affect economic growth also suggests that the rebound effect. the capital stock. Telecommunications and economic activity: An analysis of Granger causality. or twice as much service (that is. Considering the capital stock originates from past output.H. Cronin. Foresight 9. thereby indicating that ICT investment growth is negatively related to energy consumption. Inf. Carlaw.K. estimation and testing. Donkers. R. T. The results of the ARDL bounds testing approach to cointegration reveal stable long-run production and energy demand functions. N. an increase in labor and/or capital stock must accompany developments in ICT..J... G.F. 2004. The GIIC Tokyo Declaration. Policy 17. Gilling. E. To do this.. P. In the production function. Bomhof. Basque Centre for Climate Change. in: Proceedings of the 2005 International Conference on Simulation and Modelling. Syst. The impact of ICT investment and energy price on industrial electricity demand: dynamic growth model approach. the improvement of fuel efficiency in vehicles used for taxi services.. Chakraborty. Systematic analysis of rebound effects for ’greening by ICT’ initiatives. L. 483–488. Master’s thesis.S. F. ICT applications in the research for environmental sustainability. 31–47. Kim. Ghali. In the energy demand function.. Conclusions This study estimates the impact of ICT investment on economic growth and energy consumption in Japan using annual time-series data over the period 1980 to 2010. 2010.. Cho.. E. Gold.and short-run impact of ICT investment on real GDP is not statistically significant. Efficient tests for an autoregressive unit root. J. T. 225–238.H. J. M.W. K. Y. Fuchs. A. Rothenberg. but not for ICT investment.. This result is consistent with the findings in Takase and Murota (2004) concerning the relationship between energy consumption and ICT investment in Japan. Econometrica 64. van Hoorik. with the production function explained by labor.. 2007. the Japanese government should take deliberate steps to implement a growth-oriented policy. both the long. Strategies 76. 291–309.... Cronin. ARDL approach to test the productivity bias hypothesis. Suppose that the fuel efficiency of a taxicab doubles. Thus. it could also lead to an increase in working hours per capita. 1993.. Barratt. but not to an increase in GDP. Colleran. Working Paper Sries 18. Sustain. Dutta.B.A.

Lee. Energy Econ. 2961– 2970. Energy Policy 39. 2007. Energy Econ. Explaining the (non-) causality between energy and economic growth in the U. Causal relationship between energy consumption and GDP revisited: the case of Korea 1970–1999. Sadorsky.. Environ. Technol. The power of cointegration tests. Toffel. Oxf. S. M. Econometrics and Economic Theory in the 20th Century. Smith. H. 2002. Wang. Eocon. 5. J. Environ.. W... 38. 16. 1998. Ericsson.. CEE telecommunications investment and economic growth. J. Y. Commun. 2012. Ind.. 130–136. 705–718. P. Energy Econ.. Dolado. Telecommun. Shin. Szklo. Rejeski. Zhou. Information Technology and Sustainability.. J.. Energy Econ. D. Cambridge. E-commerce. Zhao. Wang. 2004. 2004. Soytas. L. Madden.W.I. R. R.. 33–37. Sustain. 325–348. Phillips. 2012. Roth.. 2008.. 2007.I. Ishida / Telematics and Informatics 32 (2015) 79–88 Gross. P. Energy Econ. Inf. J. Appl.B. The saving and investment nexus for China: evidence from cointegration tests. Books on Demand. Z. Econ. Y. 54.K. Sustainable development and ICT interpreted in a natural science context: the resulting research questions for the social sciences. Pesaran. 2008.. 1233–1253... Telecommunications and Economic Development.. Energy Econ.... vol. C. X..J. Energy consumption and economic growth in China: a multivariate causality test.. Dig more coal – the PCs are coming. Hu. X. 26. 34. Wealth of nations and telephone density. Lu. Ind. Hilty.. Thomas. Savage. 2004.P.. C.. Inf. Environmental implications of wireless technologies: News delivery and business meetings. P...... A. K.. U. B. 29.F.M. Narayan. Energy Policy 48. 1. The impact of IT investment on energy: Japan and US comparison in 2010. L.S. R. 2000. Appl. Murota. Information communication technology and electricity consumption in emerging economies. Kwiatowski.. 22. S. 1–3. A. J. M. Shin. Y. 2003. . 278–286.R. P. Shin. 2006. Energy Policy 32. energy consumption and CO2 emission in oil refineries.... Policy 4. Shinjo. Yi... 1993.. energy consumption: energy hog. 15.. Regional Studies 42. Little. 137–150. Ruddy.. 1980.. ICT capital investment and productivity growth: Granger causality in Japanese and the USA industries.M. L. T. Kremers. G. 1963. F.. A.J. Ecol. A. Laitner.. in: Energy Consumption Baseline. Energy consumption and GDP: causality relationship in G-7 countries and emerging markets. 1979–1990. Mills. 1291–1301. Sari. 1983.). Energy Econ. 70–72.. Huber. D. A multivariate cointegration analysis of the role of energy in the US macroeconomy. Econometr.. Stern. Takase. 30. A. Y. 2002. productivity tool. 7–22. E-business and sustainable development.. Horvath. N. 2001. 173–195. 489– 499. 1999. The Johns Hopkins University Press. Information technology and U. The role of the telephone in economic development. Energy consumption by office and telecommunications equipment in commercial buildings.J. 159–178.....J. or both? J. Schmidt. Zhu. 289–326. In: Strom..88 H. J.. Zachariadis. Causal relationship between telecommunications and economic growth in China and its regions.. 2008.H. 2005. T.J. J. Energy and economic growth in the USA: a multivariate approach. 37.. 267–283. P... 30. G.M.C. Soc. 4399– 4406. Telecommun. Exploring the relationship between energy use and economic growth with bivariate models: new evidence from G-7 countries. 13.W. Ecol. 2002. Stern. K. Kleinman. D. Energy 32. M. Int. Jipp.. the internet. 54. Hardy. Wellenius. J. 1992. Y.R. 1992. Bounds testing approaches to the analysis of level relationships. Policy 10. K. 51–59. J. Y. Cambridge University Press. Bull. Schaeffer. Saunders. 2010. Statist. 2011. Econometr. Pesaran.. and the environment. Arthur D. Kang. 25. 3077–3094. Hilty. (mimeo). 6. Testing the null hypothesis of stationarity against the alternative of a unit root: how sure are we that economic time series have a unit root. 262–274. Zhang. Sci. Dev.H. Econ. J. An autoregressive distributed lag modeling approach to cointegration analysis. J.P. Lam. 1999. J. Goldstein. – a multivariate sectoral analysis. 13–24... Shiu. Forbes 31. Energy consumption and economic growth: evidence from China at both aggregated and disaggregated levels. K. Oh. Yuan. 6. (Ed. D. 2004. R. Fuel specification..S. 199–201. Warford.J. J. M. 1075–1092. P.