انرژی های تجدید پذیر و بیکاری: تجزیه و تحلیل تعادل عمومی

Using a three-sector general equilibrium model, the impact of renewable electricity support policies on the rate of equilibrium unemployment is analyzed. In a simple two-factor version of the model, the paper shows analytically that renewable electricity support policies lead to an increase in the rate of unemployment. A numerical analysis is conducted with an expanded three-factor model. In this version, most scenarios analyzed also lead to an increase in equilibrium unemployment. However, the paper identifies conditions in which renewable energy support policies can decrease the rate of equilibrium unemployment. In particular, when the elasticity of substitution between capital and labor is low, when capital is not mobile internationally, and when the labor intensity of renewable generation is high relative to conventional generation, renewable electricity support policies may reduce the rate of equilibrium unemployment. The model is parameterized to represent the US economy, such that the magnitudes of quantities can be observed. Although there is some variation in the results depending on parameters, the findings suggest in general that reducing electricity sector emissions by 10% through renewable electricity support policies is likely to increase the equilibrium unemployment rate by about 0.1–0.3 percentage points. Highlights • I modeled renewable electricity support policies in general equilibrium. • The model includes a labor market that does not clear. • Renewable electricity policies generally increase unemployment by a small amount. • I identified scenarios where renewable electricity policies reduce unemployment.

Policy makers concerned about high rates of unemployment as well as the environmental impacts of energy production naturally gravitate towards policies that might be able to address both problems at once. Consequently, support has surged for policies that promote renewable energy generation, partly because of their impact on ‘green job’ creation, and employment overall. For example, a report by the US Mayors states that “[Renewble energy investments] carry macroeconomic benefits as well - they create jobs, increase productivity, and generate income that creates further jobs” (Global Insight, 2008). Similarly, US President Obama's New Energy Plan for America is designed to “help create five million new jobs by strategically investing $150 billion over the next ten years to catalyze private efforts to build a clean energy future.”1 The economics literature contains several contributions that explore the impacts of renewable energy policies on employment and the labor market. Many of these existing studies use a fixed coefficient input–output framework. For example, Hillebrand et al. (2006) conduct an analysis of the German renewable support policy using a modified input–output model, and conclude that the policy is likely to generate a positive level of net employment in the near term, but a negative level in the medium- to long-term. Lehr et al. (2008) use a similar model, but supplement with a detailed survey to generate input–output coefficients specific to the renewable electricity sector. They find that the net effect of the renewable energy support policies in Germany is positive, such that the policy lowers the long-run rate of unemployment. This conclusion is reinforced by Ragwitz et al. (2009), who estimate that EU-wide renewable energy support policies have generated a net positive impact on employment. Their study again uses an input–output framework, and this is coupled with a macro-economic model. Other studies employ a computable general equilibrium model to assess the impacts of renewable electricity policies on employment. For example, Böhringer et al. (2012) implement a model of the feed-in tariff policy in the Canadian province of Ontario, and conclude that employment is likely to decrease, and unemployment is likely to increase, as a result of policy implementation. Kuster et al. (2007) examine the impact of renewable energy investment subsidies in EU countries on a variety of economic variables including the level of unemployment, using a multi-sector, multi-region, recursive dynamic computable general equilibrium model. They find renewable energy subsidies increase unemployment rates in each of the countries they study. One disadvantage of these studies is that they all use relatively detailed and consequently opaque models, making the mechanism by which renewable energy policies increase or decrease unemployment rates unclear. In contrast, there exists an established theoretical literature that explores the link between environmental taxation and employment using stylized general equilibrium models. For example, Bovenberg and de Mooij (1994) use a simple general equilibrium model with labor-leisure choice (but without involuntary unemployment) to explore the impacts of a labor to dirty goods tax shift on employment and non-environmental welfare. They find that the relatively narrow base of the environmental tax implies that the dirty goods tax is more distorting than the labor tax, and results in a fall in the real wage, and consequently a reduction in employment. Schneider (1997) uses a similar model but includes involuntary unemployment, and concludes that a shift towards environmental taxation is likely to reduce unemployment (but see comment by Scholz (1998)). A series of papers by Bovenberg and van der Ploeg, 1996, Bovenberg and Van der Ploeg, 1998 and Bovenberg and Van Der Ploeg, 1998 as well as a summary by Bovenberg (1995) is based on similar types of models that include various specifications for involuntary unemployment and consider the effect of mobile factors, substitution elasticities, initial tax rates, and factor shares. This literature generally concludes that shifting the tax burden from broadly-based income taxes to dirty goods taxes is unlikely to boost employment. This paper aims to bridge the two literatures by developing a simple general equilibrium model with some detail around electricity generating technologies to predict the impact of policies that support renewable energy on long-run levels of employment and unemployment. Additionally, the model developed here admits the comparative analysis of alternative renewable energy support policies. The model is deliberately kept simple in order to maintain its transparency. In the model, electricity is generated from either renewable or conventional sources. Conventional electricity requires fossil fuel inputs, which produce pollution when consumed. In contrast, renewable electricity requires no fossil fuel inputs, and produces no emissions. In addition to the electricity sectors, the model includes a sector to capture economic activity in the rest of the economy. Labor is supplied by households to firms and used by the government sector. Additionally, equilibrium unemployment is modeled, using a reduced-form wage curve relationship between the real wage and the rate of unemployment. The model includes pre-existing taxes, which makes it possible to capture the effect of interactions between renewable electricity support policies and pre-existing distortions, which is shown to be important by Bovenberg and de Mooij (1994) and others. The model is used to project the impact of alternative renewable energy support policies on the rate of equilibrium unemployment. Two main types of policies are considered: a tax on fossil fuels, and a subsidy on renewable electricity generation. Tax reforms are conducted in an equal-yield environment, where increases or decreases in revenue related to renewable electricity policies are offset by changes in the labor tax rate. In a simplified version of the model where labor and renewable energy sites (land) are the only primary factors of production, the subsidy on renewable electricity production raises the equilibrium level of unemployment. However, when more complexity is introduced into the model in the form of a second primary factor of production (capital), it is no longer possible to draw unambiguous conclusions. In particular, it is shown that when the substitutability between capital and labor is limited, when capital is internationally immobile, and when the renewable technology is relatively labor-intensive compared to the conventional generation technology, it is possible that renewable energy support policies lower the rate of equilibrium unemployment. After developing the analytical model, the model is paramaterized numerically, to roughly represent renewable electricity policy in the United States. The central-case parameter choice suggests that increasing the support for renewable electricity generation throughout the US in order to reduce fossil fuel consumption in the electricity sector by 10% could increase the rate of equilibrium unemployment by up to 4% (or roughly 0.3 percentage points, given a benchmark equilibrium unemployment rate on the order of 7%). The rest of the paper proceeds as follows. In Section 2, the key features and assumptions that underlie the model are described. Section 3 describes the process by which the model is linearized such that it admits an analytical solution. In Section 4, the model is solved to produce estimates of the impact of renewable energy support policies on the rate of equilibrium unemployment. In Section 5, a numerical version of the model is parameterized and used to produce numerical estimates of the impact of renewable energy support policies on the rate of equilibrium unemployment. Finally, Section 7 concludes.

Using a simple analytic general equilibrium model, this paper aims to characterize the relationship between renewable energy support policies and the rate of unemployment. The simplest version of the model includes only one primary factor of production, and suggests that subsidies to promote the adoption of renewable energy, as well as taxes to discourage the consumption of fossil fuels used by conventional electricity generators, will increase the rate of unemployment. A more detailed version of the model that includes capital as a second primary factor of production produces slightly more nuanced results. In particular, when substitution possibilities between the capital and labor are limited, when capital's international mobility is restricted, and when the renewable electricity generation technology has a relatively high cost share for labor, it is possible that renewable electricity promotion policies could reduce the equilibrium rate of unemployment. It is important to consider the policy relevance of the model results. Debate around renewable energy policy often focuses on employment impacts of such polices. On one side, proponents claim large job creation benefits from adoption of renewable support policies. On the other side, detractors suggest that such policies can lead to significant job losses. The results in this paper suggest neither is correct: overall changes in employment and unemployment rates resulting from adoption of renewable energy policies appear very small. The focus of the debate around such policies should shift from employment impacts to other more relevant metrics.