Validity of Double Dividend Hypothesis in EU-15 Countries

The increases in climate change, biological diversity loss, ecosystem degradation, effects of chemical pollution on the human health, scarcity of the sources, energy and water security have been experienced in the world in recent years together with globalization and the sharp increases in the mass production. In this regard, especially developed countries have begun to implement the policies such as environmental tax reforms to overcome these problems. This study examines the impact of environmental tax reforms on both environment and employment in EU-15 countries (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, United Kingdom) during the period 1995-2012 by using panel cointegration and Panel FMOLS tests. We found that the double dividend hypothesis was valid in EU15 countries.


Introduction
Today, for many countries both the problems of environment and unemployment have been started to become a major problem.Therefore, including developed countries, many countries have begun to take important steps to resolve these types of problems particularly in recent years.In most of the EU member countries, the heavy tax burden on employment has been suggested as one of the major causes of unemployment seen in the 1990s.In these countries, it has been accepted that the unemployment problem can be resolved through environmental tax reforms.
One of the objectives of environmental tax reform is also to protect of the environment and to increase the employment.For this, with constant of total tax burden, reducing taxes on labor and increasing taxes on the environment have been advocated.Thus, the protection of the environment and the increase in employment have been achieved.The double-dividend hypothesis as the theory suggests that environmental taxes can improve the environment and increase economic efficiency simultaneously.In this case the "double profit hypothesis" (double dividend hypothesis) is expressed.In other words, according to this hypothesis, environmental tax revenues obtained to protect the environment, are used to reduce taxes on labor.In this context the aim of the work is to defend the effectiveness of the hypothesis of a double dividend EU-15 countries and to test whether this hypothesis is valid.
In this context, the work is addressed primarily to the hypothesis that put forward earnings double on environmental taxes.Later literature is devoted to the empirical results with the data and methodology.

Double Dividend Hypothesis
The concept of Double-Dividend Hypothesis, firstly was driven by Tullock (1997), later on this hypothesis Terkl (1984), Lee and Misiola (1986), Pearce (1991) andD. D (1997) suggested more ideas on how the environmental tax revenues will be used (Hur, 2000).Since the 1990s, earning double hypothesis has been discussed by many scholars and important when politicians availability (Günaydın, 1999).
As well as a rise in employment and the quality of the environment by making the environmental tax reform, it is expressed as a double dividend hypothesis (Bovenberg & Ploeg, 1998).Placing a tax on activities affecting the environment adversely will correct existing distortions in economic decisions.Therefore economically useful abolition of taxes on activities will increase transfer efficiency on the activities of these economically harmful taxation (Günaydın, 1999).
There are several approaches achieved through environmental taxes in terms of revenue should be used for this purpose.Lowering the OECD by obtained on the environment according to a report published in 1996, the tax excessive tax burden on labor income in terms of efficiency of use for the reduction of unemployment through is said to be the most effective way (Hur, 2000).
A relatively small number of studies (Takeda, 2007) reached the complex findings.These contradictory results, the role of special factors, labor is due to the choice of tax instrument used to assess the environmental tax revenues and tax interaction effect of supply (Fraser and Waschik, 2013: 283).In this section Which is about the validity of the hypothesis made we will cover the double dividend of some empirical findings.Markandya et al. (2012) for Spain's economy, including the informal labor sector have developed a new general equilibrium model.Besides, undeclared labor force registered in the model is different from other models in the literature included in the labor force, but it is assumed to be missing a substitution between the two parts.Labor mobility between the formal and the informal, unregistered employment continues to the point where real wage is equal to the expected formal charges.Models covered by the CO 2 emissions by between 5 % and 30 % are looking to reduce the CO 2 tax determined for different amounts.CO 2 emissions derived from taxes,  Lump sum payment transfers,  Or reducing taxes on employment  It is recycled to reduce the tax on capital.
All this comes against a neutral tax reform (public fixed income) were compared assuming prosperity effects.As a result of analysis of a portion of tax on their employment with the carbon tax shifting to Spain they have identified could lead to a small decline employment.
Fraser and Waschik ( 2013) computable general equilibrium model (CGE-computable a general equilibrium) is studied using the hypothesis of a double dividend for Australia and the United Kingdom.Especially coal in the production of specific energy commodities, mainly oil and natural gas as primary energy products (fixed) are identified as factors that play an important role.Using a CGI model of the Australian economy, as assessed by a drop in consumption tax revenues obtained from emission tax, they have found that up to 12-13 % decline in a powerful dual gain occurs.The United Kingdom, although not in the direction of the CG model hypothesis double earnings results, led by Australia and discounts up to a much lower level of about 2 %.Orlova et al. (2013) were analyzed using the CGE model sectoral and macroeconomic impact of the carbon tax on the Russian economy by using the CGE model.The results of the study hypothesis of double earnings demonstrate that international mobility of capital have emerged under the following conditions:  The presence of high labor supply elasticity.
 The presence of high-energy elasticity of substitution between labor and capital sum.
 The presence of low elasticity of substitution between capital and energy.
Nerudová and Dobranschi on (2014) 15 EU countries using Granger causality test period 1995-2011 have analyzed the validity of tax interaction effect.Causality test results have determined the only way to the environmental tax to income tax causation, the absence of a two-way relationship between two variables, the emergence of a revenue-conversion effect, so they reached the finding that the second gain is not valid double earnings hypothesis.

Data and Methodology
In this study, we investigated the validity of double dividend hypothesis in EU-15 countries (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden and United Kingdom) during the period 1995-2012.We estimated two models in the study.We used the CO 2 emissions as dependent variable and environmental tax revenues (envtax), energy consumption (enecons) and real GDP (GDP) as independent variables in the first model.In the second model we used the unemployment rate (unemp) as dependent variable and environmental tax revenues (envtax) and real GDP (GDP) as independent variables.We estimated two models in the study, because we want to test the validity of double dividend hypothesis by examining the impact of environmental tax revenues on the environmental pollution and unemployment.Our models were presented in (1) and ( 2) numbered equations: The variables used in the study were presented in Table 1 and the variables were obtained from European Community Statistical Office (Eurostat).Firstly we tested the time series properties of the panel data by Im, Pesaran and Shin (IPS) ( 2003) panel unit root test, then we estimated the long run relationship among the variables by Kao (1999) cointegration test and estimated the cointegrating coefficients by panel FMOLS.

Empirical Analysis
In the empirical analysis, we firstly tested the stationarity of the panel data by Im, Pesaran and Shin (IPS) (2003) panel unit root test, then we estimated the long run relationship among the variables by Kao (1999) cointegration test and the direction of the relationship by panel FMOLS.

5.1.Panel Unit Root Test
We tested the stationarity of the variables by Im, Pesaran and Shin (2003) panel unit root test.We selected Schwartz information criterion as the optimal lag length which eliminated the autocorrelation problem among the error terms.Also we applied unit root test with constant and trend, because the series included trend.The results of panel unit root test were presented in Table 2.The results of unit root test showed that the variables in Model 1 and Model 2 were not stationary at their level.So, we applied the unit root test after taking the first differences of the variables and the variables became stationary after first differencing.

Results of Kao Cointegration Test
We examined the long run relationship among the variables in Model 1 and Model 3 by Kao cointegration test and the results of cointegration test were presented in Table 3.We rejected the null hypothesis (there was no cointegration relationship among the series) in both models and therefore, there was long run relationship among the series in Model 1 and Model 2.

Results of Panel FMOLS
We estimated the long run cointegrating coefficients and their direction by panel FMOLS and the results of the test were presented in Table 4.The results showed that there was negative relationship between environmental tax revenues and environmental pollution and unemployment.So we concluded that the double dividend hypothesis was valid for this group of countries.

Conclusion
It is aimed to shift the taxes obtained from labor and capital to the emission which causes the environmental pollution with the double dividend hypothesis by pegging the total tax revenues.Thus, the environment will be protected and increases in the employment will be realized.Consequently, there will be increases in the welfare after the decreases in the tax burden on the labor and capital.In this regard, the validity of the double dividend hypothesis is very important.Therefore, we conducted econometric analysis to find out whether double dividend hypothesis is valid or not.We reached the finding that the double dividend hypothesis is valid for EU-countries and this finding is consistent with the general trend in the literature.We found that the increases in the environmental tax revenues decreased both the environmental pollution and the unemployment in this group of countries.We saw that the environmental tax reforms by developed countries contributed to the decreases in the environmental pollution and unemployment.Therefore, all the countries should consider to implement environmental tax reforms and to decrease the tax burden on the labor which may be a solution to overcome the problems of environmental pollution and unemployment.

Table 1 :
Variables used in the analysis and their symbol

Table 2 :
Model 1 and 2 Results of Panel Unit Root Tests