مالیات سبز و سود تقسیمی دو بل در یک اقتصاد پویا
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23314||2008||14 صفحه PDF||سفارش دهید||5570 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Policy Modeling, Volume 30, Issue 1, January–February 2008, Pages 19–32
This paper examines a revenue neutral green tax reform along the lines of the double dividend hypothesis. Using a dynamic general equilibrium model calibrated to the US economy, we find that increasing gasoline taxes and using the revenue to reduce capital income taxes does indeed deliver both types of welfare gains: from higher consumption of market goods (an efficiency dividend), and from a better environmental quality (a green dividend), even though in the new steady state environmental quality may worsen. We also find that, given the available evidence on how much households are willing to pay for improvements in air quality, the size of the green dividend is very small in absolute magnitude, and much smaller than the efficiency dividend.
Green tax reform has become a major policy issue in the OECD countries. A number of countries such as Sweden, Denmark, the Netherlands, the United Kingdom, Finland, Norway, Germany and Italy all have implemented explicit environmental tax reforms. The stakes are large. Tax revenues raised from green taxes average about 2% of GDP, but exceed 4% of GDP in some OECD countries. According to Baker and Elkins (2003), the estimates of the impact on US GDP by complying with the Kyoto Protocol vary between a decrease of about 2.5% and an increase of 3% of GDP. The Intergovernmental Panel on Climate Change (in IPCC, 2001) asserts that, in order to comply with the prescribed limits of the Kyoto Protocol, the carbon tax required in the US would be associated with a .45–1.96% decrease in GDP by the year 2010. Results from a poll conducted in the US in 1998 indicate that revenue recycling, that is, using new revenues from green taxes to decrease pre-existing distortionary taxes, may make green taxes politically feasible (see International Communications Research, 1998). Revenue recycling raises the possibility that green tax reform may yield a double dividend. The double dividend hypothesis is nicely exposited in Goulder (1995a) and Bovenberg (1999). Apart from increasing welfare due to lower pollution externalities, a ‘green’ dividend, environmental taxes raise revenue that can be used to lower other pre-existing tax distortions, resulting in welfare gains from a smaller deadweight loss of the tax system, or ‘efficiency’ dividend. Because of its appealing nature, environmental tax reform has been labeled a ‘no regret option.’ This paper examines the effects of a particular environmental tax reform in the US along the lines of the double dividend hypothesis. For policy purposes there are two strands of literature that are of direct relevance to this paper, one is normative, the other is positive. The first strand is concerned with what is the optimal tax structure. In particular, it examines whether in the presence of preexisting distortions, the optimal environmental tax lies above its Pigovian level. Here, the distortionary effect of increasing green taxes above the level at which the marginal pollution damage is internalized should be compared to the efficiency gains from reducing other taxes. In an influential paper, Bovenberg and de Mooij (1994) find that, although environmental quality improves, the efficiency dividend does not materialize. This important result has become a stepping stone, and has proved robust to a number of extensions, including capital accumulation dynamics (e.g. Lans Bovenberg and Goulder, 1996 and Lans Bovenberg and Smulders, 1996). A second strand of this literature is positive in nature. It asks what are the specific economic effects of a particular, perhaps hypothetical, policy reform. In a very influential paper in macroeconomics, Lucas (1990) finds that shifting capital income taxation completely to labor income taxation has negligible effects on long-run economic growth in a model of endogenous growth which is calibrated to the US economy. In environmental economics, the papers that fall in this category are Jorgenson and Wilcoxen (1993a) and Jorgenson and Wilcoxen (1993b). Jorgenson and Wilcoxen (1993a) estimate a model for the US using post war data. Simulations from this model suggest that a carbon tax would have qualitatively different impacts on long-run GDP depending on the preexisting taxes that are reduced. The authors also note that the costs of keeping CO2 emissions below predetermined standards would increase with higher levels of GDP growth. A similar possibility was already mentioned by Koskela and Schob (1999), and considered in more detail by Bayindir-Upmann and Raith (2003), who showed that, in a distorted labor market, substituting green taxes for labor taxes would increase employment and output and eventually produce a detrimental effect on the environment. Goulder (1995b) used a calibrated model to consider different tax recycling policies after a carbon tax was imposed, and found that green tax reform will invariably reduce the efficiency of the tax system. Zhang (1998) uses a dynamic general equilibrium model to assess the macroeconomic effects of reducing carbon dioxide emissions in China. The current consensus of the effects of green tax reform is summarized by Bovenberg in his preface to de Mooij (2000), “Whereas the second dividend may be in doubt, the first dividend (i.e. a cleaner environment) remains a powerful reason for the introduction of pollution taxes.” This claim is strengthened by a report by the OECD (2001), in which in 65% of all simulations where additional green tax revenues were used to cut social security contributions, GDP rose as a consequence of green tax reform, however, when extra tax revenue was used to cut personal income taxes, only 23% of all simulations resulted in a higher GDP. For any policy discussion it is absolutely critical to know how the green tax revenue is recycled. Much of the literature on the double dividend hypothesis assumes that green tax revenues are recycled through a decrease in labor taxes. In the models used, labor is often the only primary factor of production (see Parry, 1998, for example for a review). While this assumption may be very sensible in the European context where labor markets are distorted by a variety of factors, in the US context, where labor markets are relatively unfettered, it is worthwhile to consider the effects of recycling the green tax revenue through lower capital income tax rates. We know from the literature on optimal taxation (see Atkeson and Kehoe, 1999, Chamley, 1986, Judd, 1987 and Jones et al., 1993) that under certain robust conditions the optimal capital tax rate in the long-run is zero. For purposes of welfare economics, we know from Lucas (1990) and others that changes in capital incomes taxation have strong welfare effects. In this paper, we study the effects on national income and welfare of a green tax reform that raises a tax on a polluting good. This tax reform is revenue neutral; we lower a sizeable pre-existing distortion by decreasing the capital income tax in order to keep government’s share of the GDP constant. We consider both the steady state implications and the effects along the transitions. We use a version of the Cass-Koopmans economy where output is produced with capital and fuel and where fuel is produced with capital. The consumption of fuel adds to the stock of pollutants. Individual utility depends on a standard consumption good, on the consumption of fuel, and on the stock of health, which is inversely related to the stock of pollutants. The main results of our paper are based on the response of investment to changes in the capital tax rate. Empirically, there is now clear evidence that corporate taxes have strong effects on capital accumulation. Using a panel of US firms spanning 36 years, Cummins et al., 1994 and Cummins et al., 1996 estimate the responses of investment to changes in corporate taxes, using major tax reforms as natural experiments, and find a robust and large response of firm level investment to changes in tax rates. This paper contributes to the literature on the double dividend hypothesis in several ways. First, we focus on a very particular tax, namely a gasoline tax, rather than a general carbon or green tax. It is noteworthy that according to the green tax German memorandum (see FOSEV, 2004) ecotaxes in Germany entail the least amount of red tape and have the lowest adminstrative costs of all German taxes. Policy changes such as increasing the gasoline or fuel tax as considered here are thus very easy to implement. Second, our results are based on a model calibrated to the US economy. Our calibration includes two novel features. Most of the literature that uses calibrated models to study the effects of particular policy reforms uses CES utility functions (see, for example Lucas, 1990). Such utility functions do not match observed price and income elasticities for the demand for gasoline. We, thus, consider a wider class of utility functions which are consistent with these observations. Moreover, we bring into our calibration the large literature on the valuation of environmental amenities (see, for instance Freeman III, 1993). This literature allows for the calibration of preferences for a cleaner environment and, therefore, for evaluating the environmental and market effects of green tax policy reform in a unified framework. Third, our analysis shows the importance of the transition dynamics in evaluating the welfare effects of tax reform. All along the transition, we examine the effects that higher levels of capital accumulation resulting from a lower tax on capital earnings will have on environmental quality as well as on the other economic variables, a point first raised by Jorgenson and Wilcoxen (1993a). Our results show that, because a lower tax rate on capital encourages capital accumulation, the new steady state levels of capital and consumption of the clean good are higher than their pre-reform levels and, as a result, the quality of the environment may worsen in the new steady state. However, in all the cases we consider, at the beginning of the transition a cleaner environment is obtained, and consumption has to be sacrificed in order to build up capital, so that accounting for transition dynamics is necessary in order to access the welfare effects of this policy change. Our results show that both dividends are likely to materialize under relatively general conditions. We also find that the green dividend, or higher discounted utility from a cleaner environment, is much smaller than the efficiency dividend, or higher discounted utility from the consumption of market goods. These results are broadly consistent with those found in the literature. They complement those found in Lans Bovenberg and de Mooij (1994) and most of the double dividend literature in showing that, given current levels of taxes, a green tax reform of the type examined here would achieve both dividends. These results also show that once transitional dynamics are accounted for, the negative impact of growth on the environment as suggested in Bayindir-Upmann and Raith (2003) is not sufficient to reverse the welfare gains obtained from a better environment quality at the beginning of the transition, a point on which the policy literature is silent. In Section 2, the model is presented. In Section 3, functional forms and parameter values are chosen. Section 4 presents the results and Section 5 concludes the paper.
نتیجه گیری انگلیسی
In our model, raising a green tax does indeed allow a pre-existing tax to be decreased, here a tax on capital income. Cutting the highly distortionary capital taxes does reduce the deadweight loss from the tax system given current tax levels, so green tax reform does yield one dividend. If fuel is an input in the production of capital, however, increasing the capital stock raises the demand for fuel which may offset any decline in fuel use due to higher fuel taxes. While this offsetting effect is important in steady state comparisons, it is dwarfed by substitution effects that decrease the consumption of fuel and thus deliver a better environmental quality for a very long period along the transition path. This result certainly depends on the elasticity of substitution in production between capital and energy. If this elasticity of production were to be substantially smaller than the value we use in our model, this result may be overturned. For the estimates available for the US economy, however, the green dividend is indeed achieved. It is worth noting that given the low value that households show for the quality of the environment, the size of this welfare gain (the green dividend) is very small in absolute terms, and much smaller than the efficiency dividend. Our results suggest that the green dividend may not be after all a strong argument in favor of the implementation of green tax policy reform. Our analysis does suggest however that policymakers who are contemplating a green tax reform should give serious consideration to how the extra revenue should be recycled. In our paper the largest pre existing tax distortion is a tax on capital income. The welfare gains of green tax reform from being able to correct this particular distortion are large.