166 VI Summary and conclusion What determines tax policy? What motivations do governments follow in tax reforms? Do voters react to tax reductions and tax increases? Our knowledge of these questions is – based on the current empirical literature – still very limited. This results largely from a lack of reliable data. In this book we employed and transformed a new data-set on tax reforms (based on data from the Federal Ministry of Finance) that offers unique research possibilities on the political economy taxation. Our data-set allowed us to analyze more than 1,000 new regulations in tax policy from 1964 to 2004 including their expected fiscal effects, the date when they were adopted, and the date when they became effective. We showed that the fiscal effects included in our data-set are a reliable quantitative indicator for tax reforms as they are highly significant for explaining revenue developments. Based on our data-set we derived new insights on tax reforms in Germany and developed a quantitative description of reform patterns. We found that tax reductions dominated the number of new regulations, but that tax increases played an important role as well. Overall tax reforms reduced the tax burden on average by 0.13% of GDP per year from 1964 to 2004. Most of the new regulations were small. Regulations with fiscal effects lower than 0.1% of GDP accounted in sum for 89% of all reforms. An analysis of the timing showed that tax laws were passed especially in March, June and December, and the vast majority of tax reforms became effective in January. A detailed study by kind of tax revealed that tax burden reductions were strongly dominated by wage and income taxes while consumption taxes played an important role in tax burden increases. We found the highest tax reform intensity in wage and income taxes, followed by wealth taxes, mineral oil taxes, corporate profit taxes, VAT and the local trade tax. At the core of our analysis was the political economy of taxation, but we analyzed the impact of normative approaches as well. What have we learned about the economics of tax reforms in Germany from 1964 to 2004? What are our main results and contributions to the literature? With respect to the importance of international tax competition for business taxation we found that the reform pattern in corporate profit taxes is strongly influenced by tax competition, while there is no indication for an effect of tax competition in the local trade tax. However, in our overall analysis we were largely able to abstract from tax competition as fiscal effects of corporate tax reforms accounted only for less than 4% of the total fiscal effects of tax reforms covered in our data-set. With respect to normative theories we found that revenue needs, economic stabilization motives, tax smoothing and income tax rate reductions as a reaction to cold progression had only a very limited explanatory impact in our analyses of tax reforms. Only the financing function could partly be supported: High deficits were 167 correlated with tax burden increases via tax reforms. However, the correlation was not strong and the explanatory impact for tax reforms was low in the time period analyzed. Therefore, we conclude that normative approaches discussed here do not limit the room for the positive analysis of tax reforms substantially, as they are largely unable to contribute to our understanding of the pattern of tax reforms. With respect to polit-economic theories the results are more promising. First, we showed that tax policy was – contrary to the inertia hypothesis – characterized by permanent and far-reaching change. Within 39 years the cumulated fiscal effects of all tax reforms accounted for roughly two times the total average annual tax revenues. With respect to the fiscal illusion hypotheses we found that German governments did not try to camouflage tax burden changes by the timing of tax increases and tax reductions. Of all polit-economic approaches we found the strongest support for the opportunism hypotheses. Governments did try to increase their re-election probabilities by scheduling tax burden increases directly after and tax burden reductions in the year before the elections. Thereby governments focused on the timing of implemented and not so much of adopted reforms. As the theory of opportunistic government behavior predicts, wage and income taxes were most important for electoral manipulation while the timing of reforms in most indirect taxes (especially VAT and tobacco) showed no indication of opportunistic motivation. Contrary to most qualitative studies we were not able to find any empirical support for the hypotheses of differences in tax policy depending on the partisan orientation of governments. None of the main partisan hypotheses was reflected in the data. Left-wing governments reduced the overall tax burden stronger than right-wing governments. Furthermore, and as well contrary to the partisan hypotheses, leftwing governments reduced the tax burden stronger in the progressive wage and income taxes than right-wing governments and increased the tax burden stronger in the regressive consumption taxes. Most surprising were our findings with respect to the importance of divided government. Instead of legislative gridlock, which plays a prominent role especially in the case-study based literature on tax reforms in Germany, we found by far more and higher fiscal effects of tax reforms under divided than under undivided government. In our data there was no indication that a majority of the opposition effectively blocked tax reforms nor that tax reforms were larger and more frequent in case of undivided government. Furthermore, there was no evidence that the federal government in case of divided government concentrated on reforms of taxes which did not need approval of the second chamber. Instead we found that especially tax burden increases were substantially lower in case of undivided governments. This shows that undivided governments feared to be blamed exclusively for tax burden increases in the next elections and therefore abstained from tax burden increases. After studying the influence of political and institutional factors on tax policy, we were interested in whether tax reforms affected electoral decisions. Did voters react to different tax policies? 168 Based on the new data we were able to show that tax reductions increased (and tax increases reduced) the re-election probabilities of incumbent governments. A combination of real GDP growth and the fiscal effects of tax reforms was able to explain more than 80% of the vote share changes of the incumbent governments in the ten federal elections analyzed here. Especially implemented reforms influenced vote share changes and voters took tax policy in the whole legislative period before the elections into account. If we review our overall results, we found that electoral motivation played a key role in German tax policy. Undivided governments refrained from tax increases, because they feared to be punished by the voters in the next election. And almost all governments tried to increase their re-election probabilities by scheduling tax increases directly after elections and by reducing the tax burden before elections. But if we combine these findings with our analysis of economic voting, we see that large parts of the opportunistic behavior of governments were in vain. Voters reacted to fiscal effects of tax reforms throughout the preceding legislative period and they were not fooled by tax reductions just before the federal elections. It follows that the revealed patterns in tax policy-making were suboptimal even for a self-interested politician driven by electoral motives: Instead of just scheduling tax burden reductions before elections it would have been far more promising for self-interested governments to focus on the net tax burden reductions throughout the whole legislative term. Many questions remain open. Our study is restricted to a relatively limited time period of 1964 to 2004 and covers only the Federal Republic of Germany. Furthermore, there are important theoretical approaches we were unable to integrate in our empirical testing. And while we shed light on the importance of several (partial) theories and gained some interesting insights, we are far from being able to explain German tax policy and tax reforms satisfactory. However, one thing seems to be very clear and important for future analysis: Public choice plays an important role and cannot be neglected in the study of tax reforms.