110 Only the financing function could partly be supported: Deficits were correlated with tax burden increases via tax reforms. However, the correlation was not strong and the explanatory impact of the deficit variable for tax reforms was low. Therefore, we conclude that the normative approaches discussed here do not limit the room for our positive analysis as they are largely unable to contribute to our understanding of the pattern of tax reforms. 3 Polit-economic theories Positive polit-economic approaches take – in contrast to the considered normative approaches – the tax policy process (its actors and its institutions) into account. What do we expect if self-interested politicians pursue tax policy? Do parties differ in their tax policy strategies? Do elections play a role for the timing of tax policy? What is the impact of divided government on tax policy? These are important questions of polit-economic approaches which we discuss in the following. To be able to discuss these questions, we must first review the tax policy process and its most important changes in the period analyzed (part V.3.1). This serves as background for the discussion and empirical testing of different polit-economic theories (part V.3.2 to V.3.8). We start with a discussion of the development of the “financial constitution” (part V.3.1.1), which regulates tax revenue assignment and legislative competencies in taxation, and then move on to the political development in Germany (part V.3.1.2). Finally, we consider the role of the constitutional court (part V.3.1.3) and of the EU (part V.3.1.4) in the German tax policy process. 3.1 Legislative competencies in tax policy and political development 3.1.1 The financial constitution – legislative competencies in tax policy Most important for the legislative competencies in tax policy is the financial constitution (Finanzverfassung) which regulates the assignment of tax revenues and of legislative competencies to different levels of government in the Federal Republic of Germany (most importantly in between the federal government and the states). We first shortly review the roots of the German financial constitution and then move on to the development of the tax policy process after 1950. 3.1.1.1 Historical development of the German financial constitution before 1950 We can distinguish four periods in the historical development of the German financial constitution before 1950. From 1871 to 1919 the German financial constitution was very federalistic as a 111 result from Germany’s late consolidation as a nation state in 1871 and a continuingly strong influence of the German states.192 After unification in 1871 the states were hesitant to transfer political power to the federal level – especially in the important area of taxation. The main direct taxes, which had developed so far,193 and their revenues continued to be under the individual legislative authority of the states, while the federal level’s authority was restricted to customs and some indirect (mostly consumption) taxes. As revenues from these taxes were very limited, the federal level relied on transfers from the states to offset deficits.194 The revenue position of the federal level was strengthened somewhat by a general reform of public finances in 1909 which brought major excises (especially on beer, hard liqueur and tobacco) and their revenue under the authority of the federal level, but the states continued to be dominant in tax policy. The second period started with the “Erzberg reform” of 1919/20 which fundamentally changed the public finance system of Germany with respect to the tax code and the financial constitution. The Erzberg reform harmonized the formerly 26 different state income tax laws195 and transferred the political authority for the most important taxes from the states to the federal level. The fiscal authority of the states and municipalities was largely curtailed.196 The municipalities continued to receive the revenues of real estate and local trade taxes and received additionally parts of the revenues of income taxes, corporate profit taxes, and sales taxes in a new revenue sharing system that displaced the former separated revenue system. In many aspects the new system was a reversal of the old system and thereby the states became “boarders of the federal government”.197 In the third period, the reign of the Nazis from 1933 to 1945, the role of the states was reduced to pure administrative districts. On December 1st, 1936 a general tax reform centralized tax legislation for 98% of all tax revenues on the federal level.198 Tax policy became an almost exclusive matter of the federal government. Finally, the fourth period started after the second world war and Germany's defeat (capitulation of the “German Reich” on May 8th, 1945) when the central control agency of the allied forces was funded as central governing council with farreaching legislative competencies (from June 5th, 1945 on).199 As tax legislation was 192 The thirty years' war (1618 to 1648) resulted in a Germany of 234 territorial units and a united Germany did not emerge until 1871. This imperial regime (1871 to 1918) – where Prussia represented more than ? of the population and surface – struggled strongly with the German states for political influence. 193 The real estate, wealth, trade and general income tax. 194 The so called “Matrikularbeitraege”. The German chancellor Bismarck therefore called the German Reich a “boarder of the states” and tried to extent indirect taxation to strengthen the revenue position of the federal government. 195 See Moeller (1971), pp. 40 ff. or RGBL (1920), pp. 402 ff. 196 As competing taxation (“konkurrierende Steuergesetzgebung”) – meaning that for example the states introduce income taxes in addition to the federal income taxes – was forbidden. 197 See Respondek (1921) for a detailed discussion of the Erzberg reform. 198 See for a discussion: Terhalle (1952), pp. 318 ff. 199 See for a discussion: Haeuser (1966), pp. 204 ff. 112 not seen to be first priority, most of the previous tax code (which had not been changed fundamentally by the Nazis) continued to be in effect.200 Major changes in tax policy and the financial constitution were postponed to the time after the establishment of the Federal Republic of Germany (West Germany) and the German Democratic Republic (East Germany) in 1949. 3.1.1.2 Development of the financial constitution since 1950 The Federal Republic of Germany (in the following referred to as Germany) was established by the basic law, the German constitution, which came into force in West Germany on May 24th, 1949. Germany became a federal democracy consisting of a federal level and 11 states plus Berlin.201 In the Federal Republic of Germany the distribution of tax revenue and of legislative competencies in tax policy in between the federal level and the states continued to be an important issue. The articles 104 ff. of the basic law include the general rules on tax legislation, tax administration and tax revenue assignment. Changes of these general rules – like changes affecting the tax revenue distribution – require a ? majority of votes in both chambers of parliament. With respect to the financial constitution, two factors are especially important: the revenue assignment and – based thereon – the assignment of legislative competencies. We discuss both separately in the following. 3.1.1.2.1 Tax revenue assignment Figure 52 displays the development of revenue assignment to the different levels of government based on the financial constitution from 1950 to 2005. We can distinguish in between tax revenues assigned exclusively to the federal level, tax revenues assigned exclusively to the states, tax revenues which are assigned jointly to the states and the federal level (joint taxes) and municipal tax revenues. From 1950 to 2005 we observe three periods with different revenue assignments. 200 The most important economic reform of the control council was the introduction of the “Deutsche Mark” in a currency reform on June 20th, 1948. At the same time the liberalization of large parts of the economy especially in the consumption good sector, where government and price controls were abandoned, took place. However, tax reforms were not included in this major reform package, and even the necessary regulation for burden sharing of the costs of the war had to be worked out later by the German parliament. See Mueller (1982), pp. 130 ff. 201 On April 25th, 1952 Baden, Wuerttemberg-Baden and Wuerttemberg-Hohenzollern were unified to the state of Baden-Wuerttemberg reducing the number of states to 9 plus Berlin. On January 1st, 1957 the Saarland became the 10th state of the Federal Republic of Germany, but remained associated economically with France till 1959. 113 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 19 50 19 52 19 54 19 56 19 58 19 60 19 62 19 64 19 66 19 68 19 70 19 72 19 74 19 76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 Joint taxes Federal taxes State taxes Municipal taxes Wage and income tax becomes joint tax VAT becomes joint tax period studied in tax reform data set Re ve nu es / To ta l T ax R ev en ue s PERIOD II PERIOD I PERIOD III TAX REVENUE ASSIGNMENT 1950-2004 Data Source: Federal Statistical Office(2007). Figure 52: Changes in tax revenue assignment 1950-2006 In the first period (from 1950 to 1955), the largest part of tax revenues was assigned to the federal level, but tax revenues of the state level (including wage and income taxes) and of the local level (including the local trade tax202, the real estate acquisition tax and the real estate tax) played an important role as well. In 1955 – the starting point of the second period – the revenues of income and wage taxes started to be shared by the federal level and the states (joint taxes).203 This strongly increased the importance of joint taxes which – caused by a strong expansion of revenues in wage and income taxes – became the most important revenue category by the end of the second period (1969). In 1969 – the beginning of the third period – a large reform of the German financial constitution took place.204 From 1969 on the revenues of the value-added tax 202 The local trade tax had a federally unified regulation since 1936 and the municipalities were only able to levy multipliers on a federally unified tax base. See Maunz (1954), p. 192. and Hansmeyer (1981), p. 619 for a discussion. 203 The “financial equalization law” (see BGBl I (1955), p. 817) fixed the assignment on revenues from income and corporate taxes on the federal level and the states with a ratio of ? to ?. 204 For a discussion of the fundamental reform of the financial constitution in 1969 see Blankart (2006), pp. 657 ff., Lehmbruch (2000) pp. 112 ff. or Renzsch (1991). For the tendency of 114 (VAT) – which had been assigned to the federal level exclusively – became joint revenues. Furthermore, parts of the local trade tax revenues were assigned to the federal level, while the municipalities received larger parts of the wage and income tax revenues.205 Together these fundamental changes of the financial constitution led to a dominance of joint tax revenues after 1969. Since then around ¾ of all tax revenues have been joint taxes. 3.1.1.2.2 Legislative Competencies in tax policy BT = Bundestag BR = Bundesrat (representation of state governments) M an da to ry co -d ec isi on M an da to ry co -d ec isi on Federal Government BR new policy no policy change no policy change consent no consent approval no approval proposal no proposal BT no policy change No nm an da to ry co -d ec is io n N on -m an da to ry co -d ec is io n Federal Government BR no policy change new policy rejection no rejection approval no approval BT no policy change New policy no policy change overrule BT no overrule Source: Braeuninger/Koenig 1999, p.213. LEGISLATIVE PROCESSES IN TAXATION proposal no proposal Figure 53: Legislative processes in tax policy After reviewing the revenue assignment, we now move to a discussion of legislative competencies in tax policy. We first review general legislative processes in taxation and then discuss the most important changes within the period analyzed. centralization within the German federal system see Blankart (1999). For a general analysis of German federalism see Blankart (2007). 205 See BGBl I (1969), pp. 1587 ff. 115 The basic law established two legislative chambers in charge of federal legislation: the Bundestag and the Bundesrat. While the members of the Bundestag are elected in federal elections, the Bundesrat consists of representatives of the state governments which have voting rights depending on the population size of their states. Generally Bundestag and Bundesrat are together in charge of federal legislation. But the basic law distinguishes in between two general legislative processes: mandatory co-decision and non-mandatory co-decision (see Figure 53).206 In case of mandatory co-decision, a bill which was passed by the Bundestag has to be agreed upon by the Bundesrat to come into force (veto-right of the Bundesrat). In case of nonmandatory co-decision, an agreement of the Bundesrat is not necessary, as the Bundestag can overrule a rejection of the Bundesrat.207 Legislative competencies in tax policy and the type of legislation to be applied (mandatory co-decision or non-mandatory co-decision) are linked to the revenue assignment discussed in the previous part. For taxes with joint tax revenues mandatory co-decision applies while for taxes assigned exclusively to the federal level approval by the Bundesrat is not necessary for policy changes. However, legislation on the federal level is not restricted to joint and federal taxes. After 1950 almost all state taxes and municipal taxes were regulated uniformly on the federal level and here mandatory co-decision applies as well. If we analyze the development of legislative competencies in the German tax system, we find that since 1950 federal legislation applied to on average 97% of all tax revenues from 1950 to 2005 (see Figure 54). With respect to the influence of the Bundesrat, we see that the importance of mandatory co-decision increased gradually from 1950 to 1969 (resulting especially from the increasing importance of the wage and income tax revenues and the harmonization of the tax base of the real estate tax in 1951) and was pushed strongly up in 1969 when VAT became a joint tax. After 1969 the distribution of revenues in between mandatory and non-mandatory codecision was very stable. The most important change after 1969 was the introduction of a federally unified tax base in the real estate acquisition tax, which reduced the importance of the states in tax legislation slightly further.208 206 For a discussion see Braeuninger/Koenig (1999). 207 Legislative matters of the states are directly decided in each state parliament. 208 Since 1970 the real estate acquisition tax belonged to the “competing legislation”. From 1983 on regulation was federally unified. 116 LEGISLATIVE COMPETENCIES IN TAX POLICY * Revenues assigned to mandatory co-decision include municipal taxes with federally unified tax base, although tax multipliers are set by the municipalities. Data Source: Federal Statistical Office. *2004 = end date of our data-set. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 19 50 19 52 19 54 19 56 19 58 19 60 19 62 19 64 19 66 19 68 19 70 19 72 19 74 19 76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 Federal legislation Mandatory codecision Non-mandatory codecision VAT becomes joint tax period studied in tax reform data set * Re ve nu es / To ta l T ax R ev en ue s Figure 54: Importance of mandatory co-decision in tax legislation 3.1.1.2.3 The current tax policy process In the following we review the status quo of the tax policy process at the endpoint of our data-set (2004). Reunification in 1990 (which increased the number of German states from 11 to 16) did not lead to fundamental changes in the financial constitution. Figure 55 shows the revenue assignment and legislative competencies in 2004. With exception of some minor taxes209, which altogether amounted for only 0.1% of total tax revenues, all tax legislation took place at the federal level in 2004. Especially based on the high importance of joint taxes, mandatory co-decision (with the necessity of the Bundestag and the Bundesrat to agree on tax policy changes) applied to 82.1% of all tax revenues in 2004 while in case of 17.8% of revenues nonmandatory co-decision was applied. 209 As e.g. a tax on beverages, a tax on dogs and on hunting, an amusement tax and some others. 117 Data Source: Federal Statistical Office. *2004 = last year of our data-set. Revenue assignment (2004)* i t ( ) Legislative competencies (2004)*i l ti t i ( ) % of all tax revenues (2004) % of all tax revenues (2004) REVENUE ASSIGNMENT AND LEGISLATIVE COMPETENCIES Joint Taxes 69,9% Federal Taxes 17,7% State Taxes 4,2% Municipal Taxes 8,2% Bundestag 17,8% States only 0,1% Bundestag and Bundesrat 82,1% Figure 55: Revenue assignment and legislative competencies (2004) If we analyze the distribution of competencies in the main taxes (see Table 7), we see that the Bundestag does not depend on approval of the Bundesrat in mineral oil and tobacco tax regulation while for all other major taxes mandatory co-decision applies. 118 Tax assignment Tax legislation Changes 1950-2005TAX Joint revenue (federal level, states, municipalities) Bundestag and Bundesrat Tax revenue assignment of wage and income tax to federal level before 1955 (legislation: only Bundestag)Wage and income tax Joint revenue (federal level, states, municipalities) Bundestag and Bundesrat Tax revenue assignment of VAT/Sales Tax to federal level before 1969 (legislation: only Bundestag) Value-added tax (till 1968 sales tax) Joint revenue (federal level, states) Bundestag and BundesratCorporate profit tax Joint revenue (federal level, states, municipalities) Bundestag and Bundesrat (multipliers by municipalities) Until 1970 tax assignment exclusively to the municipalities (but legislation by Bundestag and Bundesrat) Local trade tax Federal level BundestagMineral oil taxes Federal level BundestagTobacco taxes Municipalities Bundestag and Bundesrat (Multipliers by municipalities) Real estate tax States Bundestag and BundesratInheritance tax MunicipalitiesReal estate acquisition tax Bundestag and Bundesrat Before 1983: individual state regulations. TAX ASSIGMENT AND LEGISLATIVE COMPETENCIES IN MAIN TAXES Table 7: Importance of bicameral legislation in main German taxes 3.1.1.2.4 Implications for our study of the political economy of taxation What are the implications of the discussed developments of the financial constitution from 1950 to 2004? First, we found that tax policy on the federal level was most important and covered nearly all tax revenue throughout the period studied in our data-set (1964 to 2004). Second, we had to distinguish in between taxes where mandatory co-decision applied and taxes where the Bundestag was ablet to pass changes without approval of the states. This has two implications: first, the second chamber and its majority constellation were important for tax policy. Second, we have to take an important change within the period studied in our data-set into account. The importance of the second chamber in tax policy has increased strongly from 1969 on when VAT was transformed into a joint tax. This becomes important especially when we study the effects of divided and undivided government on tax policy in part V.3.7. 119 3.1.2 Political development In the previous part we discussed the legislative process in tax policy and the role of Bundestag and Bundesrat. In both chambers the political parties, which participate in “forming the political will of the people”,210 play a major role. All federal and state governments since 1950 have been party governments. In all governments at least one of the major parties, the CDU/CSU or the SPD, dominated.211 The socialdemocratic party SPD is a left-wing party, while the conservative CDU/CSU is more on the right wing.212 POLITICAL DEVELOPMENT 1950-2005 19 51 19 53 19 55 19 57 19 59 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 = federal elections Left-Wing Government (with majority in the 2nd chamber) Left-Wing Government (without majority in the 2nd chamber) GRAND COALITION (with majority in the 2nd chamber) Right-Wing Government (with majority in the 2nd chamber) Right-Wing Government (without majority in the 2nd chamber) period studied in tax reform data set Figure 56: Political development in Germany 1950-2006 210 As article 21 of the basic law says. 211 We abstract from the fact that all governments (except the CDU/CSU government from 1960 to 1961) were coalition governments of CDU/CSU or SPD with a substantially smaller coalition partner. A discussion of the influence of coalitions on partisan politics can be found in Schofield (1993), Austen-Smith/Banks (1988), Laver/Schofield (1990) or Laver/Shepsle (1996). 212 See for a discussion of partisan preferences in general and in Germany Budge et al. (2001), Castles/Mair (1984) or Laver/Hunt (1992). 120 As both chambers of parliament played an important role in tax policy and politics were strongly influenced by political parties, we have to take the majority constellations in both chambers and the partisan orientation of government into account when we review the political development. If the parties of the federal government dispose of a clear partisan majority in the second chamber, the Bundesrat, we define such a situation as undivided government as the opposition parties in the Bundestag are not able to veto government projects (which need approval in the second chamber) in the Bundesrat. In all other cases, we speak of divided government.213 Figure 56 shows the development of party governments and divided/undivided government in the Federal Republic of Germany from 1950 to 2005. As our data-set on tax reforms starts in 1964/65, we focus on the years thereafter.214 On a right-wing (CDU/CSU-led) government (in power from 1949 on) followed the grand coalition of CDU/CSU and SPD from 1966 to 1969. Thereafter a SPD/FDP government ruled without a majority in the second chamber until 1982. From 1982 till 1998 a CDU/CSU-FDP government was in power until 1990 with a majority in the second chamber.215 In 1998 an SPD/GREENS government came into power but never had a majority in the second chamber. Altogether we observed undivided governments for 13 years while government was divided in 26 years.216 213 We apply the official definition of a majority of votes in the second chamber based on the “Rstates” in the Bundesrat. R-states are those states, in which the state governments consists only of the parties governing as well on the federal level. 214 Our analysis starts with undivided right-wing government in the year 1965. We coded the year 1966 as undivided government although the FDP dropped out of the federal government in October 1966 (leading to a situation of divided government). The grand coalition government was constituted in December 1966 and governed in 1967, 1968 and 1969 as undivided government. 215 After the “vote of no confidence” and the government change to a CDU/CSU and FDP government in October 1982, the CDU/CSU-led government disposed of a majority in the second chamber until June 1990 and then again from November 1990 to April 1991. As the CDU/CSU-led government disposed of a majority in the second chamber for ? of 1990 and had the chance to pass laws, which might have been blocked over the summer, in November and December with a majority in the second chamber, we coded 1990 as undivided government leading to a period of undivided right-wing government from 1983 to 1990. From April 1991 on, no government disposed of a majority in the second chamber. This led to a period of divided government from 1991 to the end of our data-set in 2003/2004. 216 We count only those years which are covered completely in our data-set (1965-2003). In 15 out of these 26 years of divided government the opposition had the majority in the second chamber while during 11 years the majorities were unclear (as government coalitions on the state level often included only one of the federal coalition partners). 121 3.1.3 The role of the constitutional court The Federal Constitutional Court (Bundesverfassungsgericht), based in Karlsruhe, has the power to exert far-reaching influence on German tax policy as the Basic Law provides wide scope for judicial review. Individual state governments or a defeated minority in the Bundestag217 can challenge a law in the court. Individuals can also demand an examination of legislation that affects them directly, and lower courts can ask for legislation relevant to a case to be tested by the court for compliance with the constitution. If the court finds a law to be unconstitutional, it can deactivate it. Within the period studied there have been several judgements which directly affected tax policy.218 One example is a ruling in 1990 which demanded the tax exemption of child raising costs and led to an increase in child allowances and child benefits in the tax code. Another is the decision in 1991 that compliance of interest taxation has to be enforced. This led to the introduction of a withholding tax on interest income in 1993. Furthermore, the court decided in 1992 that the basic allowance in the wage and income tax was much too low, because the subsistence level (measured by the level of social assistance payments) was not completely taxexempt. In the following years, this ruling led to an increase of the tax exemption level. A fourth important decision with direct effects on tax policy was that the constitutional court ruled the different treatment of real estate219 and other assets in the net wealth tax to be unconstitutional. The constitutional court set the parliament a deadline for a new regulation by December 31st, 1996, but the government decided to abolish the tax completely and increased the real estate and real estate acquisition tax instead.220 As these examples show, the constitutional court initiated several important changes of tax policy which are part of the data-set we want to analyze. Does this mean that we have to include the constitutional court explicitly as an actor in tax policy in our analysis? First, we have to note that the constitutional court did not establish new regulations but made new regulations by the legislative necessary. Thereby the constitutional court initiated tax policy changes but did not formulate tax policy. Therefore, we argue that the direct influence of the constitutional court on the tax policy decisions in our data-set was nonetheless limited as the chambers of parliament were still free to decide in between different options. 217 Provided it comprises at least one-third of the members of the Bundestag. 218 For the mentioned rulings, analyses of the role of the constitutional court in German tax policy and decisions affecting tax policy see especially Obermeier (1999) but as well Herter (1992), Landfried (1992), Huber (1999), Vogel (1999) and Wagschal (2005). 219 Which is taxed based on standard values calculated for the last time in 1964 – see discussion of property taxes in part IV.2.5. 220 A current (January 31st, 2007) ruling of the constitutional court on the inheritance tax, which made an inheritance tax reform before 2009 necessary, is not included as 2007 is not covered in our data-set. 122 A second important argument is that although the constitutional court initiated important policy changes, the share of regulations with no influence of the constitutional court strongly dominated. Even a generous counting of new regulations, which can be directly related to rulings of the constitutional court,221 leads to a share of less than 5% of the fiscal effects of all reforms covered in our data-set. Based on these two arguments and our goal to keep the quantitative analysis as parsimonious as possible we decided to exclude the constitutional court as a separate actor from our analysis. 3.1.4 The role of the EU The ongoing economic and political integration of Germany in the European Union has increased the possibilities of the EU to influence national policies. Attempts of tax coordination and harmonization222 have the potential to affect German tax legislation. Has the EU had an important influence on German tax policy so that we need to integrate the EU as an actor in our analysis of tax policy? In the past, the EU focused especially on consumption taxes and special excises, while harmonization played only a very minor role in the other taxes like corporate income or wage and personal income taxes. With respect to German tax policy analyzed here, the effects of the EU on national regulation were only marginal. Some taxes of minor importance were abolished within a process of tax harmonization in Europe.223 However, these yielded only negligible revenues. In consumption taxes, which are important with respect to revenues (like especially the VAT, the mineral oil and the tobacco taxes), the EU tried to establish a limited bandwidth.224 While this directly affected tax policy in some EU countries, there was never a substantial direct influence on German regulation in these taxes as German tax rates were always higher than the agreed minimum levels and lower than the agreed maxima. Therefore, we are able to widely abstract from the influence of the EU on German tax policy in our data-set. 221 Beyond the examples discussed here. 222 See for an overview EU (2002), Gentschel (2002) and European Commission (2000). 223 See the discussion in part II.3.3. 224 Both the tobacco and the mineral oil tax are partly harmonized within the EU. With respect to mineral oil taxes, the EU defines the products on which a mineral oil tax can be levied, and there is a minimum taxation of around €0.30/l for fuel (€0.337 for leaded and €0.287 for unleaded fuel (see EU (2002), pp. 51ff.)) wich is much lower than the German rates (see part IV.2.4). With respect to the tobacco tax, the current minimum rates are for example 57% (tobacco tax plus VAT) of the final sale price for cigarettes. However, in Germany tobacco tax rates are with more than 70% much higher. 123 3.1.5 Summary What are the implications of our analysis of the tax policy process and its most important changes for the polit-economic analysis of tax policy? First, we found that the federal government dominated tax policy. Throughout the period studied in our data-set (1964 to 2004), federal legislation decided upon almost all tax revenues. Second, we have to distinguish in between taxes where codecision of the two German chambers of parliament was mandatory and where it was not (as especially in the mineral oil and tobacco taxes). Third, partisan interests are possibly important in tax policy as all German federal governments were party governments led either by the conservative CDU/CSU or by the social-democratic SPD. The combination of partisan governments and the importance of bicameral legislation especially in mandatory co-decision, which leads to a veto-right of the second chamber, points at a possibly important influence of divided government on tax policy. The fifth point is that we – with respect to the importance of divided government – have to take the important structural change in our data-set in 1969 into account. When VAT became a joint tax in 1969, the possible importance of divided and undivided government increased strongly. Finally, we found that rulings of the constitutional court initiated important changes of tax legislation. Nonetheless, we abstract from the constitutional court in our analysis as the court did not directly change the regulation but just initiated changes. With respect to the EU we found that the influence from the European harmonization of consumption taxes on German tax policy had been only marginal as German tax rates were always within the bandwidth established by the EU. 124 3.2 Polit-economic approaches to tax policy – overview There is not one unifying theory of the political economy of taxation but many different approaches that try to explain tax policy. Some of these approaches can be combined, but some are also competing. Furthermore, the approaches differ largely in their scope. While some are only focusing on one aspect of tax policy (for example the government’s motivation to try to influence re-election probabilities), others are far more encompassing. This makes the models are often hard to compare. Furthermore, there are barely any empirical tests which compare the theories and their empirical relevance directly. This is probably the decisive reason, why the field has stayed so fragmented. Therefore, the opportunities for empirical testing based on our new data-set are especially valuable for comparing the empirical relevance of different approaches. In this section we review five major approaches from which we can directly derive testable hypotheses. We start with the theory of inertia, followed by a discussion of the concept of fiscal illusion. Then we move on to approaches with opportunistic and partisan motivation in tax policy and analyze the effects of divided government within the German bicameral legislature. Finally, we test combinations of several hypotheses statistically (part V.3.8) and summarize our results. Overview – polit-economic theories of tax policy and tax reform Opportunistic behavior Governments try to increase their re-election probabilities by tax policy Tax burden increases will take place after, tax burden reductions before elections Can be tested directly based on tax reform data Alternative approaches Fiscal Illusion by timing of tax reforms Governments use timing of tax reforms to camouflage real tax burden changes A larger share of tax increases (compared to tax reductions) will be enacted retrospectively (and with a larger time-lag) Can be tested directly based on tax reform data Inertia/theory of reform Incentive to avoid tax reforms (and especially tax increases) because of high political costs Minimize political costs and maximize revenue by self-restriction to automatic tax revenue increases based on cold progression (inertia in tax policy) Can be tested directly based on tax reform data Partisan behavior Ideologically motivated redistribution via tax reforms Right wing: Reduce the tax burden and shift it to regressive taxation Left-wing: Increase the tax burden and shift it to progressive taxation Can be tested directly based on tax reform data Institutional factors (divided government) Divided government leads to political gridlock Extent and frequency of tax reform activity will be larger in times of undivided government M ec ha ni sm Pr ed ic tio ns Te st in g Can be tested directly based on tax reform data Models of bureaucracy Interest group models Probabilistic voting on taxation Median voter model Leviathan model Table 8: Polit-economic approaches to the study of tax reform