Analyzing the factors influencing the demand and supply of solar modules in Japan

Taghizadeh-Hesary, Farhad; Yoshino, Naoyuki; Inagaki, Yugo; Morgan, Peter J. Working Paper Analyzing the factors influencing the demand and supply of solar modules in Japan ADBI Working Paper Series, No. 1084 Provided in Cooperation with: Asian Development Bank Institute (ADBI), Tokyo Suggested Citation: Taghizadeh-Hesary, Farhad; Yoshino, Naoyuki; Inagaki, Yugo; Morgan, Peter J. (2020) : Analyzing the factors influencing the demand and supply of solar modules in Japan, ADBI Working Paper Series, No. 1084, Asian Development Bank Institute (ADBI), Tokyo This Version is available at: https://hdl.handle.net/10419/238441 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/igo/ ADBI Working Paper Series ANALYZING THE FACTORS INFLUENCING THE DEMAND AND SUPPLY OF SOLAR MODULES IN JAPAN Farhad Taghizadeh-Hesary, Naoyuki Yoshino, Yugo Inagaki, and Peter J. Morgan No. 1084 March 2020 Asian Development Bank Institute The Working Paper series is a continuation of the formerly named Discussion Paper series; the numbering of the papers continued without interruption or change. ADBI’s working papers reflect initial ideas on a topic and are posted online for discussion. Some working papers may develop into other forms of publication. Suggested citation: Taghizadeh-Hesary, F., N. Yoshino, Y. Inagaki, and P. J. Morgan. 2020. Analyzing the Factors Influencing the Demand and Supply of Solar Modules in Japan. ADBI Working Paper 1084. Tokyo: Asian Development Bank Institute. Available: https://www.adb.org/publications/analyzing-factors-influencing-demand-supply-solar- modules-japan Please contact the authors for information about this paper. Email: [email protected]; [email protected] Farhad Taghizadeh-Hesary is an associate professor of economics at Tokai University, Japan. Naoyuki Yoshino is dean and chief executive officer of the Asian Development Bank Institute and professor emeritus of Keio University, Japan. Yugo Inagaki is a graduate of Keio University, Japan. Peter J. Morgan is senior consulting economist and vice chair for research at the Asian Development Bank Institute, Japan. The views expressed in this paper are the views of the author and do not necessarily reflect the views or policies of ADBI, ADB, its Board of Directors, or the governments they represent. ADBI does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequences of their use. Terminology used may not necessarily be consistent with ADB official terms. Working papers are subject to formal revision and correction before they are finalized and considered published. Asian Development Bank Institute Kasumigaseki Building, 8th Floor 3-2-5 Kasumigaseki, Chiyoda-ku Tokyo 100-6008, Japan Tel: +81-3-3593-5500 Fax: +81-3-3593-5571 URL: www.adbi.org E-mail: [email protected] © 2020 Asian Development Bank Institute ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. Abstract Ever since the occurrence of the first oil shock, which struck the Japanese economy in 1973, Japan has made efforts to increase its energy efficiency, reduce its energy intensity, decrease its dependence on fossil fuels, and diversify its energy basket by introducing alternative energy sources, such as renewable energy, to raise the energy security level. One key technology that has partially supported this change is solar photovoltaic technology. Japan was one of the leading countries in the solar photovoltaic market, with the third-highest number of solar modules installed in the world. However, its development in the solar module sector in recent years, since the tremendous increase in solar module installation after the feed-in tariff policy, has become sluggish. For Japan to put itself back on track, it must find a new strategy to increase its solar module installation and raise its self-sufficiency ratio in the primary energy supply. This paper analyzes the influence of certain factors on the demand and supply of solar modules by employing the fully modified ordinary least squares method using Japanese data. The paper concludes that finance is a key issue in solar module development and that the country needs to put effort into green financing to maintain its progress in solar photovoltaic technology. Keywords: solar module, photovoltaic, renewable energy, green finance JEL Classification: Q21, Q41, O3 ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. Contents 1. INTRODUCTION ......................................................................................................... 1 2. LITERATURE SURVEY .............................................................................................. 2 2.1 Price Competitiveness of Solar Electricity ....................................................... 3 2.2 GDP and Renewable Energy .......................................................................... 4 2.3 Oil Prices and Renewable Technology............................................................ 4 2.4 Effects of the FIT Policy on Solar Modules ...................................................... 4 2.5 Financing and Solar PV Technology ............................................................... 5 2.6 Influence of Silicon Prices on Solar Modules’ Prices ....................................... 5 2.7 Influence of R&D on Solar Modules’ Prices..................................................... 5 3. THEORETICAL MODEL ............................................................................................. 6 3.1 Demand Side ................................................................................................... 6 3.2 Supply Side ..................................................................................................... 7 4. EMPIRICAL ANALYSIS ............................................................................................ 10 4.1 Test Procedures ............................................................................................ 10 4.2 Test Results .................................................................................................. 11 4.3 Regression Results ....................................................................................... 12 5. CONCLUDING REMARKS AND POLICY RECOMMENDATION ............................. 14 5.1 Policy Recommendation ................................................................................ 14 REFERENCES ..................................................................................................................... 16 ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 1 1. INTRODUCTION Due to a scarcity of resources, Japan has been dependent on fossil fuels. For this reason, the Japanese economy took a direct hit from the 1970s’ oil shocks. Since crude oil has been a vital energy commodity, fueling the world economy, several studies (see, inter alia, Hamilton 1983; Barsky and Kilian 2004; Taghizadeh-Hesary et al. 2013, 2016) have evaluated the impacts of oil price fluctuations on various macroeconomic indicators. In addition, in March 2011, a devastating earthquake and tsunami struck eastern Japan and damaged the nuclear power plant at Fukushima, resulting in a nuclear shutdown throughout the whole country. Due to the lack of safety and people’s opposition to nuclear power, the government kept the nuclear power plants off and substituted the nuclear power lost with more imported fossil fuels, which reduced the self-sufficiency of energy in the country. Taghizadeh-Hesary, Rasoulinezhad, and Kobayashi (2016), in an empirical analysis, found that, after the Fukushima disaster, because of the greater reliance on oil imports, the sensitivity of most sectors to oil price volatility declined, endangering the energy security in the country. Figure 1: Cumulative Capacity of Solar Photovoltaics in Japan from 1992 to 2017 (MW) Note: Created by the authors using data collected from the Global Market Outlook for Solar Power 2016–2020, 2015–2019, and 2014–2018 and the Snapshot of Global PV 1992–2014. After the Fukushima disaster, Japan’s energy self-sufficiency fell from a high of 20.2% in 2010 to 6.4% in 2014 (METI2016). At the same time, this incident engendered an opportunity for the renewable energy (RE) sector. The share of renewable energy in the generation mix rose from 9% in 2010 to 12% in 2014. Japan’s Ministry of Economy, Trade and Industry (METI) set several objectives for its energy generation mix by 2030. It set targets for fossil fuels’ share to decrease to 56% and nuclear energy to restart and to account for 20% to 22%. The RE source objectives are ambitious, set to reach 22%– 24% by 2030, with hydropower accounting for 8.8%–9.2%, solar 7%, wind 1.6%, biomass 3.7%–4.6%, and geothermal 1.0%–1.1% (JOGMEC 2018). Japan has experienced significant growth in the renewable energy sector, specifically in the solar photovoltaic area. The changes in the total amount of solar photovoltaics installed in Japan are shown in Figure 1. In the past 5 years, the amount of solar photovoltaics installed has grown tenfold. Driving this growth is the change in the global price of solar modules. In the past 5 years, the unit price of solar modules has fallen by half its original price. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 2 In addition, Japan has witnessed tremendous growth in its installations since 2012, the year when it implemented the feed-in tariff (FIT) policy. The FIT policy is a policy that allows electricity generators to sell renewable energy at a fixed price to the grid. The government’s aim was to encourage renewable electricity generation. Its expectation was for the learning curve to help cut the production cost of solar modules through increased production. The FIT policy was successful in terms of introducing the use of solar modules. However, the growth in the production of solar modules in recent years has started to diminish. To be able to increase the self-sufficiency of energy and raise the share of renewable and clean energy in the energy basket, Japan needs to implement new plans to keep the installed solar projects perdurable even without the existence of the FIT. The aim of this paper is to understand further the structure of the solar module market in Japan by assessing the factors that determine the demand and supply sides of the solar module market. The goal is to offer an effective policy recommendation to make the renewable energy sector of the country sustainable. Section 2 provides the results of a literature review in an effort to understand the past research on the drivers influencing the solar module market. We reconfirm the importance of providing a structural demand and supply model for solar modules using economic variables. In section 3, we construct an original model of the solar module supply and demand. Our solar module model takes into account the impact of solar module prices, the price of oil (substitute energy), the GDP (economic activity), the exchange rate, and the FIT as explanatory variables on the demand side. On the supply side, our model has solar module prices, the price of silicon, the GDP, the real interest rate, and the exchange rate as explanatory variables. Section 4 presents the results of the econometric analysis that we conduct on our structural model. Section 5 provides the concluding remarks and the policy recommendations. 2. LITERATURE SURVEY The market structure of the solar module plays an important role in finding a solution to accelerate the development in the solar power infrastructure. One main market factor that has been the key in this field of study is the price of solar modules. Originally, the high price of solar modules acted as a barrier, preventing countries from investing in solar photovoltaic technology. The high price of solar modules reduces the rate of return on investment in solar power projects; therefore, the private sector shows reluctance to invest in these projects (Yoshino, Taghizadeh-Hesary, and Nakahigashi 2019). Therefore, the drivers influencing solar module prices have become a hot topic in the field of renewable energy studies. One of the most common methods of evaluating the tendency of solar module price change is to analyze the learning curve effect of solar modules. The learning curve effect refers to the tendency in the manufacturing industry to optimize the manufacturing process over time by considering the decrease in the price of products as cumulative production rises. The solar energy policies in Japan are based on the estimated future price of solar modules by taking into account the learning curve. Researchers have created several learning curve models to explain better the cost reduction process in solar modules. Numerous learning curve models have emerged in the field of renewable energy. The research by Neij (1997) indicated that learning curves are applicable to solar PV technology. Messener (1997) later created one of the first learning curve models applicable to renewable energy. This model became the basis for many renewable energy cost studies, such as the study by Barreto et al. (2000), which created the commonly known renewable learning curve model called the ERIS (Energy Research and Investment Strategy). However, the learning curve effect is only a tendency in the manufacturing price, which is the sum of numerous factors, such as the ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 3 wage rate and the material price. Nemet (2006) concluded that the actual effect of the learning curve is less than 10% of the overall cost reduction. In another group of studies that considered various determining factors for the price of solar modules, Pillai (2014) created a model explaining the cost reduction of solar modules using independent variables such as the time, polysilicon price, plant size, and firm investment. De la Tour, Glachant, and Ménière (2013) used a similar model to predict the solar module price from 2011 to 2020. They identified an experience curve model using cumulative production, R&D expenditure, the silicon price, and the silver price as independent variables. These studies concentrated on the change in the solar module price from the engineer’s perspective. Several researchers have expanded their scope to include the influence of certain factors on the demand side of the solar module market. Gan and Li (2015) conducted a quantitative study on the long-term global solar photovoltaic market using the learning curve model. They took into account the influence of the silicon price, the influx of lowercost Chinese modules, and the supply–demand gap in the PV market to explain the recent cost reduction and to forecast the future decline in solar module prices. Taghizadeh-Hesary, Yoshino, and Inagaki (2019) used an oligopolistic model and econometric method to determine the economic factors that have an influence on solar module prices by employing a reduced-form model for the top solar module suppliers, namely the People’s Republic of China (PRC), Germany, Japan, the Republic of Korea, and the United States (US). Their empirical results showed that the interest rate (cost of capital) has a positive correlation with solar module prices, while the exchange rate, knowledge stock, and oil price have a negative association with solar module prices on average. Although a large number of studies have deconstructed the dynamics of the solar module price, no research has analyzed the influence of certain factors on the solar module supply and demand separately using a structural model. Earlier research on the solar module market has provided insights into the impact of different factors on solar module prices. To understand better the influence that various economic and non-economic factors have on solar module prices, it is also necessary to examine the changes in the solar module demand and supply. This is the contribution of this paper to the literature: to develop a structural model to identify the determining factors of the demand and supply of solar modules. 2.1 Price Competitiveness of Solar Electricity The price reduction surrounding the installation of solar photovoltaic systems has been tremendous. According to Lazard (2017), the unsubsidized leveled cost of electricity (LCOE) of utility-scale solar photovoltaic systems in November 2017 was $43/Mwh, whereas conventional energy sources, such as coal, IGCC, gas peaking, and nuclear, had an LCOE of $60/Mwh, $96/Mwh, $156/Mwh, and $112/Mwh, respectively. Figures show that solar PV technology has become cheaper than conventional electricity sources in recent years. A report by IRENA (2017), which calculated the global-averaged LCOE in the years 2010 and 2017, showed that the reduction in solar PV module and system costs drove the decline in utility-scale solar PV projects. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 4 2.2 GDP and Renewable Energy The GDP influences the demand side of the solar module industry. It affects solar modules by promoting an increase in renewable energy consumption. Studies have shown that countries with a higher GDP tend to consume more electricity. Kantar and Keskin (2013) found that there is a strong relationship between electricity consumption and economic growth regardless of a country’s income level. As a country progresses economically, the amount of electricity grows. This is because it is possible to understand the GDP as a scale for the volume of economic activity within a country. Therefore, countries with a higher GDP are more likely to consume a larger amount of electricity due to corporate activity. Especially for countries such as Japan, which has scarce energy resources, the demand specifically for renewable electricity skyrockets to meet the increasing domestic demand for electricity without importing exhaustible resources from foreign countries. 2.3 Oil Prices and Renewable Technology There are two ways in which oil prices influence renewable energy. First, oil, a fossil fuel with a high level of greenhouse gas (GHG) emissions, acts as a substitute for renewable energy. Therefore, with a rise in the oil price, a country’s demand for renewable technology will increase to substitute oil with renewable energy. For countries such as Japan, which has scarce energy resources and imports most of its energy in the form of fossil fuels from abroad, the influence of an oil price increase on the demand for renewable energy technology is significant. Another way in which oil prices influence renewable energy is through its influence on renewable R&D. Cheon and Urpelainen (2012) conducted an empirical analysis on how oil prices influence renewable energy development. Their analysis showed that an increase in oil prices increases countries’ incentives to invest more in renewable energy technology. Similarly, Wong, Chia, and Chang (2013 found that oil prices and renewable R&D show a positive correlation. Therefore, oil prices influence both the supply and the demand side of the solar module market as a substitute good for solar modules and as a key driver of renewable R&D. 2.4 Effects of the FIT Policy on Solar Modules The FIT allows electricity generators to sell their renewable electricity to consumers at a fixed price. The purpose of this policy is to reduce the risk of investment in renewable energy technology to promote the incentive to invest in them. The implementation of the FIT in the early stage of renewable energy installation allows for a cost reduction in renewable energy equipment through the learning curve effect, eventually leading to competitiveness of renewable energy technology in the energy market. Japan first implemented this policy in 2012. Following the implementation of the FIT, the generation cost of solar PV dropped by 38% by the fifth year of implementation and the solar module installation grew to double the original amount, which suggests that the FIT has a correlation with solar module infrastructure in the short term (REI 2017). Moreover, the number of corporations involved in solar PV, seeking a profit, has risen since the implementation of the FIT (REI 2017). The impact of the FIT on solar module production mainly occurs through the stimulation of the demand side. The FIT policy raises the demand for solar modules. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 11 4.2 Test Results 4.2.1 Unit Root Test It was necessary to test all the series for the presence of a unit root (stationarity test). We conducted two common unit root tests, namely the augmented Dickey–Fuller (ADF) test and the Phillips–Perron (PP) test in level, to check for the presence of a unit root. Except for the interest rate, we tested all the series in their logarithmic form. Table 1 summarizes the results, which show that, for all the series, we could not reject the null hypothesis of the presence of a unit root in level. This means that all the series have a unit root and are non-stationary and that we needed to conduct the unit root test on their first differences. The results rejected the null hypothesis in the first difference. This means that the series are integrated of order 1 or I(1). As all the series share the same order of integration, we needed to conduct a cointegration test between all the series. Table 1: Unit Root Test (Augmented Dickey–Fuller Test and Phillips–Perron Test) Augmented Dickey–Fuller Test (ADF) Levels First Differences Country Variable t-Statistics p-Value t-Statistics p-Value Japan psolar –0.705 0.407 –3.375* 0.027* Production (Supply) –2.861 0.181 –4.031* 0.035* Demand –0.470 0.890 –2.142* 0.031* Interest rate –0.937 0.307 –2.783** 0.006** M2 –2.578 0.102 –6.962** 0.000** Exchange –1.636 0.459 –6.117** 0.000** Silicon –0.083 0.946 –3.118* 0.029* GDP –1.884 0.338 –5.606** 0.000** poil –1.930 0.316 –3.426* 0.012* Phillips–Perron Test Levels First Differences Country Variable t-Statistics p-Value t-Statistics p-Value Japan psolar –1.562 0.110 –2.478* 0.013* Production (Supply) 3.188 0.999 –8.597** 0.000** Demand –0.443 0.895 –9.384** 0.000** Interest rate –0.705 0.407 –3.034** 0.002** M2 –2.307 0.172 –6.827** 0.000** Exchange –0.927 0.774 –6.547** 0.000** Silicon –1.126 0.701 –6.723** 0.000** GDP –2.382 0.149 –6.168** 0.000** poil –1.224 0.660 –3.625** 0.007** Note: * denotes significance at the 5% level and ** denotes significance at the 1% level. 4.2.2 Cointegration Test To identify the cointegrating vectors, that is, the production, GDP, solar module price, interest rate, exchange rate, price of silicon and price of oil, we conducted a cointegration analysis using the Johansen cointegration test for the set of variables used for each of the demand and supply analyses. Again, we conducted the test in their logarithmic form, except for the interest rate. Table 2 summarizes the results ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 12 of the Johansen cointegration test, indicating that all the series are cointegrated and that there is a long-run association between the production (in logarithm), price of solar modules (in logarithm), interest rate, exchange rate (in logarithm), price of oil (in logarithm), and GDP (in logarithm). When there are co-integrations, the regression analysis should use an error correction model. As all the variables for this research are I(1), we considered the fully modified OLS that Phillips and Hansen (1990) introduced. This model allowed us to provide optimal estimates of the cointegrating regressions. Table 2: Cointegration Test Trace Maximum Eigenvalue Country Hypothesized No. of CE(s) Eigenvalue Trace Statistic Prob. Max. Eigen. Statistic Prob. Japan (D) r=0 0.482 153.209 0.000** 46.097 0.033** r<=1 0.425 107.111 0.001** 38.742 0.044* r<=2 0.303 68.369 0.020* 25.298 0.269 r<=3 0.256 43.070 0.048* 20.715 0.204 r<=4 0.205 22.355 0.128 16.119 0.140 r<=5 0.085 6.235 0.431 6.235 0.431 Japan (S) r=0 0.567 189.952 0.000** 58.739 0.000** r<=1 0.439 126.784 0.000** 33.876 0.007** r<=2 0.321 79.578 0.023* 27.584 0.056 r<=3 0.206 48.738 0.201 21.131 0.212 r<=4 0.097 25.647 0.492 14.264 0.468 r<=5 0.008 7.941 0.446 3.841 0.446 Note: * denotes significance at the 5% level and ** denotes significance at the 1% level. 4.3 Regression Results Table 3 summarizes the results for fully modified ordinary least squares (FMOLS). The adjusted R squared for the regression shows high figures of 0.969 for the demand side model and 0.897 for the supply side model. Table 3: Empirical Results Variable C.E. S.E. t-statistic Variable C.E. S.E. t-statistic Demand (n=72) Supply (n =72) GDP 3.353 1.352 2.478* r –0.504 0.114 –4.404** e 1.458 1.162 1.254 E –0.121 1.0190 –0.119 FIT 1.005 0.250 4.006** M2 0.48719 0.115 4.205** 𝑝𝑝𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 –3.054 0.361 –8.459** 𝑝𝑝𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 2.641 0.014 3.206** 𝑝𝑝𝑠𝑠𝑒𝑒𝑠𝑠 0.274 0.323 0.849 Silicon –1.363 0.202 –6.729** C –81.784 35.593 –2.297* C –145.657 16.847 –8.645** Adjusted R squared: 0.969 Adjusted R squared: 0.897 Note: * denotes significance at the 5% level and ** denotes significance at the 1% level. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 13 4.3.1 Demand Side The coefficients for the GDP (𝑀𝑀𝑑𝑑𝑝𝑝) on the demand side show a positive figure of 3.353 and a statistically significant t-statistic. This is consistent with the understanding that economic growth raises energy consumption. The results indicate that, for every 1% increase in the real GDP, the production of solar modules increases by 4.38%. The coefficient for the exchange rate (𝑒𝑒𝑒𝑒) on the demand side is 1.458; however, the t-statistic is not significant. Therefore, the exchange rate is not a main driver of the solar module demand. The t-statistic for the FIT on the demand side produced a significant result. This suggests that the implementation of the FIT had a significant impact on the increase in solar module installation. The FIT surely increased the incentive for people to install solar modules and maximize their profit through electricity generation. The price of solar modules has a statistically significant impact on the solar module demand. The coefficient is -3.054. This suggests that a 1% increase in the solar module price results in a 3.054% decrease in the solar module demand. This is consistent with the assumption that the solar module industry is elastic. The results also show that the price of oil has a positive impact on the solar module demand. This is consistent with our assumption that oil acts as a substitute for solar module technology in the market. Our results show that the substitution elasticity is 0.274, which means that a 1% increase in the oil price results in a 0.274% increase in solar modules. However, the t-statistic is not significant. Therefore, the price of oil was not a main driver of the solar module demand in the period of our analysis in Japan. 4.3.2. Supply Side The results show that the coefficient for the exchange rate is negative. This is consistent with the assumption that the exchange rate influences the import price for silicon, which in turn increases the production costs, resulting in a decrease in the production of solar modules. However, the t-statistic is not significant. Therefore, the exchange rate does not have a significant impact on the solar module demand. The results exhibit a price elasticity of supply of 2.641 for solar modules. This suggests that suppliers increase production when the product price rises, as we would expect. The silicon price shows a negative coefficient for the supply side. This coincides with our assumption that the price of silicon wafers makes up a significant portion of solar module production. The coefficient shows that, for every 1% increase in the silicon price, solar module production falls by 1.363%. The real interest rate exhibits negative coefficients for the supply side, which is in accordance with our assumption that the solar module industry is capital intensive. The coefficient suggests that, for every 1% increase in the real interest rate, the solar module production falls by 0.504%. The money supply (M2) presents positive coefficients for the supply side, which agrees with our assumption that the number of green financing channels influences the solar module supply. An expansionary monetary policy and an increase in the amount of M2 reduce the interest rate, which reduces the cost of capital and the production cost of solar modules and their prices. However, in reality, many banks and financiers consider renewable energy projects and renewable technology to be risky projects; hence, they are reluctant to finance them. This means that, although a low interest rate is important for reducing the price of solar modules, solar module suppliers need to be able to access credit. This suggest that the government needs to cover part of the risk. Taghizadeh-Hesary and Yoshino (2019) proposed the establishment of a green credit guarantee scheme (GCGS) to reduce the risk of green finance, as the government will alleviate part of the risk. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 14 5. CONCLUDING REMARKS AND POLICY RECOMMENDATION In this paper, we investigated the factors influencing the demand and supply of solar modules in Japan. On the supply side, we examined the influence of the real GDP, price of solar modules, FIT, real interest rate, price of silicon, and exchange rate. On the demand side, we studied the influence of the real GDP, price of solar modules, FIT, price of oil, money supply, and exchange rate. We constructed two models, a solar module supply model and a solar module demand model. Based on the two models, we conducted fully modified OLS (FMOLS) to estimate the influence of each factor using quarterly data from 1997 to 2015. Our empirical analysis results produced several findings concerning the solar module supply and demand. On the demand side, the results show that the price of solar modules and the GDP have a significant impact. On the supply side, the results indicate that all the variables, the GDP, exchange rate, FIT, price of solar modules, price of silicon, and interest rate, have a significant impact. All the coefficients are consistent with our understanding of each factor. 5.1 Policy Recommendation We confirmed through the analysis results that international and external factors, such as the price of silicon, have a strong influence on the solar module market, and the Japanese Government does not have control over these external factors. For Japan to expand its solar module market, it must shift from the reliance on the FIT policy to enhancing the financing channels for solar module production, which are an important internal factor. The feed-in tariff policy (FIT) manipulates and forces a fixed price on solar electricity to enhance the market demand for solar electricity. As the empirical results in Table 3 show, it has a significant impact on the demand side of the solar module market. We can still consider the FIT policy to be an effective tool for enhancing solar module installation. However, the manipulation of the solar electricity price through the FIT is not the most efficient target in terms of accelerating solar PV development. First, the FIT policy involves careful price setting to maintain company incentives to invest in solar PV and minimize the cost imposed on electricity consumers. A report by REI (2017) emphasized the difficulty of setting an efficient price for the FIT and the insufficiency of the discussions focusing on FIT price setting. Therefore, the continuous use of the FIT policy has the risk of misconfiguration. Moreover, the FIT policy might increase the incentives for green technology investment, but the absence of channels for initial finance might act as a bottleneck, limiting the potential increase in solar PV market entries. Therefore, the more efficient target for the further development of solar PV technology is access to finance. Easing the access to finance and reducing the capital costs will lower the initial financing hurdle, allowing an increase in the number of new entries into the solar PV market. At the same time, it will not impose the cost of capital directly on the consumers. As Taghizadeh-Hesary, Yoshino, and Inagaki (2019) noted, for the solar module industry to accelerate its development, the government must implement new renewable energy funding tactics to reduce the capital cost as well as to enhance the financing opportunities in the industry. We reconfirmed in this research that the interest rate as a capital cost as well as M2 as the money stock have significant impacts on the supply side of the solar module market. As renewable industries are high-tech, the influence that the capital cost has on the technology price is significant. The government’s efforts to provide industries with low-interest finance will accelerate renewable business. Countries are implementing many different policies regarding green finance globally, such as green bonds and green banks. However, Taghizadeh-Hesary and Yoshino (2019) stated that the current variety of green financing ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 15 channels and the amount of money invested in green projects are inadequate. They proposed the establishment of a green credit guarantee scheme (GCGS) to reduce the risk of green finance, as the government will alleviate part of the risk. The recent movement in Japan may become a gateway to realize this type of green financing. The Japanese Government has recently started putting effort into the green financing sector, leading to the creation of the Green Finance Network Japan (GFNJ) in November 2018. The Green Finance Network Japan is a taskforce consisting of financial institutions, institutional investors, development banks, think tanks, academics, NGOs, and international organizations to provide a platform to connect Japanese and international stakeholders and encourage the development of the green financing market. Although this network is still in its experimental phase, its broad network, which is not exclusive to financial institutions, may encourage all types of organizations to find their green project investment opportunity. This organization has the potential to become a pathway to the diversification of green financing channels and the expansion of the green financing market. The government should make an effort to create the infrastructure for a global green financing network that is not exclusive to financial institutions, which will consequently reduce the cost of finding a lender, and implement a well-structured policy to create more opportunities for green financing. The construction of this green financing platform altogether will lead to an increase in solar PV installation. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 16 REFERENCES Barreto, L., and S. Kypreos. 2004. “Endogenizing R&D and Market Experience in the ‘Bottom-Up’ Energy-Systems ERIS Model.” Technovation 24 (18): 615–629. Barreto, L., S. Kepreos, S. Capros, and S. Messener. 2000. “ERIS: A Model Prototype with Endogenized Technological Change.” International Journal of Global Energy Issues14 (1/2/3/4). Barsky, R., and L. Kilian. 2004. “Oil and Macroeconomy since the 1970s.” Journal of Economics Perspectives 18: 115–34. DOI: 10.1257/0895330042632708. Bruegel. 2017. “Real Effective Exchange Rates for 178 Countries: A New Database.” http://bruegel.org/publications/datasets/real-effective-exchange-rates-for-178- countries-a-new-database/. Brunnschweiler, C.N. 2009. “Finance for Renewable Energy: An Empirical Analysis of Developing and Transition Economies. Environment and Development Economics, 15 (3): 241–274. http://doc.rero.ch/record/303054/files/S1355770X1000001X.pdf. Cheon, A., and J. Urpelainen. 2012. “Oil Prices and Energy Technology Innovation: An Empirical Analysis.” Global Environmental Change22 (2): 407–417. De La Tour, Arnaud, Matthieu Glachant, and Yann Ménière. 2013. “Predicting the Costs of Photovoltaic Solar Modules in 2020 Using Experience Curve Models.” Energy 62: 341–348. Gan, Peck Yean, and Zhi Dong Li. 2015. “Quantitative Study on Long Term Global Solar Photovoltaic Market.” Renewable and Sustainable Energy Reviews 46: 88–99. Hamilton, J. 1983. “Oil and the Macroeconomy since World War II.” Journal of Political Economy 91 (2): 228–48. DOI: 10.1086/261140. Hayamizu, S., T. Furubayashi, and T. Nakata. 2014. “Quantification of Technological Learning by R&D and its Application for Renewable Energy Technologies.” Transactions of the JSME80 (811): 巻. IEA-PVPS. 1999. “Trends in Photovoltaic Applications in Selected IEA Countries between 1992 and 1998.”: 2018,1,19. http://www.iea-pvps.org/fileadmin/dam/ public/report/statistics/tr_1998.pdf. ———. 2000. “Trends in Photovoltaic Applications in Selected IEA Countries between 1992 and 1999.”: 2018,1,19. http://www.iea-pvps.org/fileadmin/dam/public/ report/statistics/tr_1999.pdf. ———. 2001. “Trends in Photovoltaic Applications in Selected IEA Countries between 1992 and 2000.”: 2018,1,19. http://www.iea-pvps.org/fileadmin/dam/public/ report/statistics/tr_2000.pdf. ———. 2002. “Trends in Photovoltaic Applications in Selected IEA Countries between 1992 and 2001.”: 2018,1,19. http://www.iea-pvps.org/fileadmin/dam/public/ report/statistics/tr_2001.pdf. ———. 2003. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2002.”: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2002.pdf. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 17 ———. 2004. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2003.”: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2003.pdf. ———. 2005. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2004.”: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2004.pdf. ———. 2006. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2005.日: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2005.pdf. ———. 2007. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2006.: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2006.pdf. ———. 2008. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2007.: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2007.pdf. ———. 2009. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2008.: 2018,1, 19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2008.pdf. ———. 2010. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2009.”: 2018, 1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2009_neu.pdf. ———. 2011. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2010.”: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2010_low.pdf. ———. 2012. “Trends in Photovoltaic Applications Survey Report of Selected IEA Countries between 1992 and 2011.”: 2018,1,19. http://www.iea-pvps.org/ fileadmin/dam/public/report/statistics/tr_2011.pdf. ———. 2013. “Trends in Photovoltaic Applications 2013.”: 2018,1,19. http://www.ieapvps.org/fileadmin/dam/public/report/statistics/FINAL_TRENDS_v1.02.pdf. ———. 2014. “PVPS Report—Snapshot of Global PV 1992–2013.”: 2018, 1, 19. http://www.iea-pvps.org/fileadmin/dam/public/report/statistics/PVPS_report_- _A_Snapshot_of_Global_PV_-_1992-2013_-_final_3.pdf. ———. 2015. “Trends in Photovoltaic Applications 2015.”: 2018,1,19. http://www.ieapvps.org/fileadmin/dam/public/report/national/IEA-PVPS_-_Trends_2015_- _MedRes.pdf. ———. 2016. “Trends in Photovoltaic Applications 2016.”: 2018, 1,19. http://www.ieapvps.org/fileadmin/dam/public/report/national/Trends_2016_-_mr.pdf. IRENA. 2017. “Renewable Power Generation Costs in 2017.”: 2018,1,19. https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2018/Jan/ IRENA_2017_Power_Costs_2018.pdf. JOGMEC. 2018. Activities of JOGMEC—Geothermal Resources Development Department. Japan Oil, Gas and Metals National Corporation (JOGMEC): Tokyo, October 2018. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 18 Kantar, Keskin. 2013. “The Relationships between Electricity Consumption and GDP in Asian Countries, Using Hierarchical Structure Methods.” Physics A: Statistical Mechanics and its Applications, 392 (22): 5678–5684. Kilian, L. 2009. “Not All Oil Price Shocks Are Alike: Disentangling Demand and Supply Shocks in the Crude Oil Market.” American Economic Review 99: 1053–63. DOI: 10.1257/aer.99.3.1053. Lazard (2017). Lazard's Levelized Cost of Storage Analysis". https://www.lazard.com/ media/450337/lazard-levelized-cost-of-energy-version-110.pdf (retrieved 15 Nov 2019). Messener, S. 1997. “Endogenized Technological Learning In an Energy Systems Model.” Journal of Evolutionary Economics, 7 (3): 291–313. Miketa, A., and L. Schrattenholzer. 2004. “Experiments with a Methodology to Model the Role of R&D Expenditures in Energy Technology Learning Processes: First Results.” Energy Policy, 32(15): 1679–1692. Ministry of Economy, Trade and Industry (METI). 2016. Japan’s Energy—Twenty Questions to Understand the Current Energy Situation. Ministry of Economy, Trade and Industry of Japan: Tokyo. Monnin, P. 2015. “The Impact of Interest Rates on Electricity Production Costs.” 2018, 1, 31. https://www.cepweb.org/wp-content/uploads/CEP_DN_Interest_Rates _Energy_Prices.pdf. (retrieved: 5 Nov 2019). Neij, L., 1997. “Use of experience curves to analyse the prospects for diffusion and adoption of renewable energy technology.” Energy Policy, 25 (13): 1099–1107. Nemet, G.F. 2005. “Beyond the Learning Curve: Factors Influencing Cost Reductions in Photovoltaics.” Energy Policy 34 (17): 3218–3232. Phillips, P.C., and B.E. Hansen. 1990. “Statistical Inference in Instrumental Variables Regression with I(1) Processes.” Review of Economic Studies, 57 (1): 99–125. Pillai, U. 2014. “Cost Reduction in Crystalline Silicon Solar Modules.” https://www.aaai.org/ocs/index.php/FSS/FSS14/paper/viewFile/9131/9122 (accessed 18 Feb 2020). Sachs, J., W.T. Woo, N. Yoshino, and F. Taghizadeh-Hesary. 2019. “Importance of Green Finance for Achieving Sustainable Development Goals and Energy Security.” In Handbook of Green Finance: Energy Security and Sustainable Development, edited by J. Sachs, W.T. Woo, N. Yoshino, and F. Taghizadeh- Hesary. Tokyo: Springer. Solar Power Europe. 2015. “Global Market Outlook for Solar Power 2015–2019.”: 2018,1,19. http://www.solarpowereurope.org/fileadmin/user_upload/ documents/Publications/Global_Market_Outlook_2015_-2019_lr_v23.pdf. ———. 2016. “Global Market Outlook for Solar Power 2016–2020.”: 2018,1,19. http://www.solareb2b.it/wp-content/uploads/2016/06/SPE_GMO2016 _full_version.pdf. Taghizadeh-Hesary, F., E. Rasoulinezhad, and Y. Kobayashi. 2016. “Oil Price Fluctuations and Oil Consuming Sectors: An Empirical Analysis of Japan.” Economics and Policy of Energy and the Environment 2 (19): 33–5. DOI: 10.3280/EFE2016-002003. ADBI Working Paper 1084 F. Taghizadeh-Hesary et al. 19 Taghizadeh-Hesary, F., and N. Yoshino. 2019. “The Way to Induce Private Participation in Green Finance and Investment.” Finance Research Letter. 31: 98–103. DOI: 10.1016/j.frl.2019.04.016. Taghizadeh-Hesary, F., N. Yoshino, M.M.H. Abadi, and R. Farboudmanesh. 2016. “The Response of Macro Variables of Emerging and Developed Oil Importers to Oil Price Movements.” Journal of the Asia Pacific Economy 21 (1): 91–102. DOI: 10.1080/13547860.2015.1057955. Taghizadeh-Hesary, F., N. Yoshino, G. Abdoli, and A. Farzinvash. 2013. “An Estimation of the Impact of Oil Shocks on Crude Oil Exporting Economies and their Trade Partners.” Frontiers of Economics in China 8: 571–91. DOI: 10.3868/s060-002-013-0029-3. Taghizadeh-Hesary, F., N. Yoshino, and Y. Inagaki. 2019. “Empirical Analysis of Factors Influencing the Price of Solar Modules.” International Journal of Energy Sector Management 13 (1): 77–97. DOI:10.1108/IJESM-05-2018-0005. Wong, S.L., W.M. Chia, and Y. Chang. 2013. “Energy Consumption and Energy R&D in OECD: Perspectives from Oil Prices and Economic Growth.” Energy Policy 第 62: 巻. Yoshino, N., F. Taghizadeh-Hesary, and M. Nakahigashi. 2019. “Modelling the Social Funding and Spill-over Tax for Addressing the Green Energy Financing Gap.” Economic Modelling 77: 34–41. DOI: 10.1016/j.econmod.2018.11.018.