This Selected Issues paper on Germany reviews investment trends and business capital stock in Organization for Economic Co-operation and Development (OECD) countries. Sharp wage increases are found to boost capital formation in the short term as employers substitute capital for labor at a rate that adjusts to the higher relative price for labor. To limit the political economy biases to fiscal policy, the paper explores options to strengthen budgetary institutions, notably more transparency; stronger budgetary rules; and more room for Länder governments to mobilize revenue and tailor spending to local circumstances.


This Selected Issues paper on Germany reviews investment trends and business capital stock in Organization for Economic Co-operation and Development (OECD) countries. Sharp wage increases are found to boost capital formation in the short term as employers substitute capital for labor at a rate that adjusts to the higher relative price for labor. To limit the political economy biases to fiscal policy, the paper explores options to strengthen budgetary institutions, notably more transparency; stronger budgetary rules; and more room for Länder governments to mobilize revenue and tailor spending to local circumstances.

IV. Perspectives on Federalism and the Political Economy of Adjustment40

A. Introduction and summary

105. The relation between Bund and Länder is attracting considerable interest, amid questions about how it might affect fiscal policy and structural reforms. Against that background, Section B briefly describes Germany’s political infrastructure. Section C explores empirically the relation between the political environment on the one hand, and fiscal consolidation and structural reforms on the other. Section D advances several proposals to strengthen budgetary institutions, and Section E discusses possible improvements in intergovernmental fiscal relations. The suggested reforms aim to counter political economy distortions of fiscal policy.

106. The empirical evidence suggests that governments with weak parliamentary support spend more and run larger structural deficits but that they do not necessarily raise state intervention in the economy through structural measures. Most economic policies require approval of both the lower and upper houses in parliament. Governments typically begin their terms with majorities in both houses, but they tend to gradually lose support in the upper house as their term progresses. This weakening of power has been associated with higher government spending and weaker fiscal positions. Matters are somewhat different for structural reforms. Over the past decade, policymaking in this domain is better characterized by a lack of consistency rather than by inaction. This might have contributed to uncertainty, which reduces the responsiveness of the economy to structural reforms.

107. Budgetary institutions can play an important role in reducing the political economy bias to fiscal policy. The idea is to foster a better understanding of the cumulative consequences of policy decisions, particularly of those regarding public expenditure, entitlement programs, and taxes and contributions.41 Nothwithstanding Germany’s good international standing, there is room to strengthen budgetary institutions:

  • Each year an independent commission of experts could prepare medium- and long-term projections for the general government based on the policies in place. Such projections are essential for the electorate to determine how budgetary measures or structural reforms would help in achieving targets for the general government deficit while maintaining healthy growth. Currently no body regularly produces such projections.

  • The presentation of all fiscal accounts should be moved to an ESA 1995 basis and they should be better integrated. Presently, it is difficult for the electorate to understand how the federal budget and the budget of each Land fit within the objectives for the general government deficit. Not only is there no aggregation of these budgets, they are also prepared on the basis of accounting standards that differ significantly from ESA 1995.

  • The golden rules governing Bund and Länder budgets could be replaced with rules that are more consistent with the Stability and Growth Pact (specifically, aiming for balance over the cycle). Also, the Internal Stability Pact could be strengthened in various respects.

108. In addition, reforms of intergovernmental fiscal relations could improve the prospects for fiscal consolidation and structural reform. At present, these relations are too complex and the mechanisms that are in place could be redesigned to provide stronger incentives for prudent fiscal management. Potential measures are:

  • Making revenue allocation across Länder more transparent by folding all redistribution into the formal equalization mechanism (the Finanzausgleich).

  • Providing additional scope for the Länder to follow independent, deficit-constrained expenditure and tax policies. At the same time, the interregional equalization mechanism could be simplified and redesigned to support Länder fiscal adjustment efforts.

B. The Political Economy Infrastructure

109. The power to shape economic policy is shared between the federal government (Bund) and 16 Länder governments. Federal parliament is composed of two chambers. The members of the lower chamber (Bundestag)—which selects the Chancellor, the head of the federal government—are chosen in general elections that take place every four years. The members of the upper chamber (Bundesrat) are designated by the state (Länder) governments: at least three and up to six per Land, depending on the population. Members for each Land have to cast their votes en bloc. Länder governments are chosen in Länder elections that are staggered throughout the term of the Bundestag.

110. According to Germany’s Constitution (Grundgesetz), both the Bund and Länder can shape economic policy but in practice the initiative in policy making has largely been taken over by the Bund, with the Länder ensuring the administration of the laws and their enforcement (Spahn, 2000). One key reason is the concern—raised in the Constitution—for establishing the same living standards across Germany.42 However, all laws proposed by the Bund that affect Länder interests—either financially or administratively—in any event need the approval of both the lower and upper chambers of parliament.43

111. Accordingly, Germany’s federalism is highly cooperative rather than competitive. In other federations, e.g., the United States, Canada, and Switzerland, lower levels of government have considerable tax and expenditure powers. In Germany, the federation develops the economic policy framework for all Länder, with the latter implementing and administering the specific policies, including through their own budgets. For instance, reflecting a very strict interpretation of the call for same living standards across Germany in the Constitution, tax law is virtually identical across Länder. Also, revenue is typically shared or apportioned among different layers of government, with an equalization mechanism ensuring that all Länder have very similar revenue per capita.

112. With this strong consensus approach, broad support within the population is almost always necessary to achieve important economic reforms. There are two key reasons. First, a majority of the members of the upper chamber (Länder representatives) need to approve all major reforms. However, the membership of the upper chamber changes frequently as a result of Länder elections that are staggered throughout the four-year general election cycle. The outcome of the Länder elections then represents, to varying degrees and over time, a view on the policies followed at the national level. Second, electoral rules favor coalition governments. Elections for the lower chamber and federal government feature direct voting for roughly half of the seats. The remainder is distributed across parties with a view to securing a representation that is broadly in line with the proportion of votes cast for the various parties.44 As a result, no single party has been able to command an absolute majority in the lower house of parliament over the past three decades. Instead, the country has been governed by coalition governments, headed either by the Christian Democrats (CDU) or the Social Democrats (SPD). Coalition governments are also the rule rather than the exception in the Länder.

C. The Role of the Political Economy Infrastructure in Fiscal Adjustment and structural Reform

113. The government’s political support can have significant implications for fiscal adjustment and structural reforms. This section explores how the political infrastructure in Germany might have influenced economic policy. Furthermore, it analyzes empirically the relation between fiscal adjustment and structural reform on the one hand and political economy indicators on the other.

Insights from the political economy literature

114. A growing literature highlights the many ways in which the political infrastructure of a country can affect fiscal policy. Optimal fiscal policy is frequently equated with intertemporal tax smoothing, where the net present value of spending has to be equal to the net present value of taxes. The budget is maintained in structural balance but deficits or surpluses can arise from the free play of automatic stabilizers. Such an optimal policy might not be pursued by policymakers for various reasons related to fiscal illusion among voters and the political infrastructure. In this regard, periodic elections and their related uncertainty, the nature of party competition, and the degree of information and polarization of the electorate, can play important roles.45

115. Germany’s political infrastructure might favor the emergence of deficits and a large public sector, according to the literature.

  • The emphasis on proportionality in Germany’s electoral rules means that legislators need to appeal to a broader spectrum of the population than in countries with majoritarian rules. Evidence suggests that countries with proportional electoral rules have larger governments and welfare programs (Persson and Tabellini, 2004).

  • Fragmentation is a greater risk when many decision makers need to be brought on board to approve major reforms. Empirical evidence suggests that economic shocks prompt action but that more fragmented governments tend to need more time to deal with fiscal adjustments (Roubini and Sachs, 1989a b and Poterba, 1994). Also, more fragmented governments have been associated with larger public sectors, particularly welfare programs (Perotti and Kontopoulos, 2002). In Germany, fragmentation between decision makers has been an issue: (i) one coalition government has broken up during its term; and (ii) Länder elections have often significantly changed the support of federal governments in the upper chamber. Thus, policy makers must take account of the frequent Länder elections throughout a federal government’s term to maintain sufficient support in the upper chamber.46

116. Moreover, different but related political economy issues concern the emphasis in the Basic Law on attaining the same living standards in fairly dispersed Länder economies.47 This emphasis is reflected in the nature of interregional fiscal relations. Evidence in the literature suggests that representatives overestimate the net benefits of local spending if the revenues also come from other districts as a result of interregional redistribution (Weingast, Shepsle, and Johnson, 1981). Section E explores whether the interregional redistribution mechanism in Germany may create a bias toward higher spending and deficits.

Indicators of the political economy in Germany

117. The federal government and the head of government change infrequently. The administration changed only twice during the last 35 years: in 1982, from a left-leaning to a right-leaning government, following the break-up of the coalition between social democrats (SPD) and liberals (FDP); and in 1998 back to a left-leaning coalition of the SPD and the Green Party, as a result of the general election. While sometimes narrow, the governing coalition’s margin in the Bundestag was always sufficient to ensure the stability of the national government (Figure IV-1). This stability in government should have been conducive to reform.

Figure IV.1.
Figure IV.1.

Germany: Central Government Margin in Bundestag

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

118. Nonetheless, the power base of federal governments can be fragmented and thus weak. Many laws require the approval of the Länder in the upper house. Länder elections are much more frequent than national ones and can tip the balance of power during a government’s term. In fact, all three governments over the last 35 years suffered a loss of their majority in the Bundesrat. Figure IV-2 shows the governing coalition’s margin of seats in the Bundesrat, using information from Länder statistical and electoral offices. Coalition governments in the Länder were classified according to their voting record. As a rule, the largest coalition party dominated (SPD or CDU). “Grand Coalition” state governments of SPD plus CDU usually abstained from voting and thus were not counted for in the margin.

Figure IV.2.
Figure IV.2.

Germany: Central Government Margin in Bundesrat

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

119. The fragmentation of the power base via a loss of majority in the Bundesrat seems to occur in a regular fashion, and has preceded the loss of majority in national elections. The support for the typical German government thus weakens predictably over the electoral cycle. The government is strong in the first term but the governing coalition increasingly needs to seek consensus in later terms, as it loses its combined majority in Bundestag and Bundesrat—which will be the measure of government support (or power) used in the analysis (Figure IV-3). Interestingly, the loss of majority accelerated between the 1970s and now, suggesting some loss of loyalty on the side of voters.

Figure IV.3.
Figure IV.3.

Germany: Central Government Overall Margin

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

Government ideology, support, and the deficit

120. A visual inspection suggests that the structural fiscal deficit is highly correlated with a government’s power but not with its ideological bent (Figure IV-4 and Box IV-1). “Left-leaning” governments have not run different deficits than “right-leaning” ones, probably reflecting a broad consensus for a degree of fiscal restraint and relatively strong budgetary institutions.48 But the correlation between support and the deficit is fairly close: the deterioration of a government’s base leads to a degeneration of fiscal discipline, typically owing to expenditures (Figure IV-5).49 As a government loses its majority in the upper house, it has to reach out to the opposition and special interests. The institutional setting enforces a consensual, drawn-out decision-making process. While this is sometimes seen as a political characteristic of Germany, it mainly applies to the final years of government. The loss of control seems to occur in an abrupt way, sometimes in conjunction with exogenous shocks: the oil crisis of the 1970s, reunification in the 1990s, and most recently the bursting of the asset prices.

Figure IV.4.
Figure IV.4.

Germany: Government Margin, Structural Balance

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

Figure IV.5.
Figure IV.5.

Germany: Structural Revenue and Expenditure

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

The Political Economy of Fiscal Adjustment

This box formalizes the text by regressing the structural fiscal balance on institutional variables for 1970-2004: the “ideology” of parliament (a higher number indicates a more “left-leaning” parliament); the majority margin of the government; a unification dummy; and the misery index (the sum of inflation and the unemployment rate—it stands for economic pressures on the government). A two-period lag (AR(2)) was included so as to have the same set-up as for the regression explaining structural reforms where it matters (Box IV-2): 1/

article image

Indicates significance at the 5 percent level.

The results suggest that the “majority” variable is the most important factor in determining the structural fiscal balance. It has a positive sign, which means that a stronger majority in parliament will deliver stronger fiscal results. The unification dummy is also significant, while ideology is not. The latter may indicate a consensus across parties against excessive deficits. Also, there is no evidence for backtracking in fiscal adjustment—the two-period lag does not enter significantly—unlike for structural reforms (see Box IV-2); and the structural fiscal balance is not sensitive to overall economic pressures as captured by the misery index.

1/ Another specification also included the output gap (to capture the effect of the cycle on social security contribution rates) but it did not enter significantly.

Government ideology, support, and structural reforms

121. Structural reforms seem to be related to economic pressures and ideology, in addition to government support. Constructing an index of the structural reform efforts of the past 30 years is not easy. Descriptive information on the reforms reviewed here was mainly drawn from the Annual Reports of the Bundesbank for 1971-2003 (Table IV-1). First, the scope of each reform measure was gauged (1-4), with a higher number indicating a more far-reaching reform. Second, the direction—more (+1) or less state intervention (-1)—was determined. The resulting index does not necessarily mirror government spending, given that some measures have large financial repercussions in bad economic times only (e.g., more generous unemployment benefits), or mainly in the future (e.g., long-term care programs), or they might have no direct repercussions at all (e.g., job protection legislation, product and financial market reforms). A simple bivariate analysis of government ideology, support, and the structural reform index reveals the following, tentative conclusions (Figure IV-6 and Box IV-2):

  • Until German reunification, left-leaning governments typically expanded the role of the state in the economy while right-leaning governments did the opposite.

  • Following German reunification, economic pressures came to dominate ideological orientation. Slow growth led to a steady upward drift of unemployment and impressed on all governments the need to improve the efficiency of the economy. Also, reform initiatives became erratic, changing direction within the same government term. While the right-leaning government cut social benefits and privatized state enterprises in the 1980s, it expanded the welfare state in 1995 by establishing a public long-term care insurance. However, it cut benefits again in 1997 by introducing a demographic factor to pensions, co-payments to health care, and subtracting separation grants from unemployment benefits. The incoming left-leaning government repealed these measures in 1999, only to reintroduce them under different headings in 2003-04. As a result, the public may well have become disoriented about the general direction and objectives of structural policy.

Table IV.1.

Germany: Chronology of Reforms, 1970 - 2004

article image
article image
Sources: Annual Reports of Bundesbank and IMF staff assessment of reforms (scope=1,..,4 with a larger number indicating a larger reform; role of state = +1 if the reform increased state intervention or -1 if it narrowed intervention).
Figure IV.6.
Figure IV.6.

Germany: Government Margin, Structural Reforms

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

The Political Economy of Structural Reforms

This box formalizes the text by regressing the indicator of structural reforms on institutional variables (1970-2004). The change in the reform indicator is regressed on the ideology of parliament, the majority margin of the government, a unification dummy, the misery index, and a lag term (this was added to remove autocorrelation found in the regression without this term).

article image

Indicates significance at the 5 percent level.

The results suggest that, since the 1970s, the ideology of parliament tilted reforms towards an expansion of the state under SPD-led coalitions and towards a reduction under CDU-led coalitions. These are average influences, however. The late 1990s saw different policies being pursued on both sides of the political spectrum. For example, in 1995, a conservative government introduced public long-term care, an important expansion of the welfare state. In 2004, the current left-leaning administration reduced benefits in health care, pensions, and assistance for the unemployed.

The variable “majority” has an unexpected positive sign. Governments with large majorities in parliament apparently engage in an expansion of the economic role of the state, not in market-oriented reforms. As Box IV-1 illustrates, however, the same strong governments carry out fiscal adjustment. Perhaps governments perceive a trade-off between fiscal adjustment and structural reform, as both cannot easily be accomplished at the same time. Alternatively, there might be feedback from the structural reforms to the majority—this concern may be addressed by specifying a system of two equations, one of which models the political process, but this would require additional data.

Unification temporarily increased the role of the state, as one-off expenditures were financed by tax surcharges. Notice that with increasing economic pressures (misery index) governments reduce distortions and make the economy more efficient. Finally, reforms were found to have some negative autocorrelation, the AR(2) term. This is evidence for backtracking, as modeled e.g. by Wyplosz (1993). Reforms generate gains that are unevenly distributed. The actual distribution of winners and losers may only become apparent after the fact. In the absence of compensation schemes, the losers may extract concessions from the government that lead to a partial reversal of the reforms. The concessions could also be in other areas of economic policy, but they would appear in any case as a renewed increase in the reform indicator.

122. Overall, governments have faced difficulties in advancing fiscal consolidation as their support in parliament has fallen over time, while structural policies have lacked a consistent direction over the past decade. It is hard to disentangle analytically whether the falling support reflects a normal political cycle, of the type that can be observed in many countries, or whether it is related to special features of the political infrastructure in Germany. But two features—the power vested in the Länder and the frequency of Länder elections—might allow a more rapid feedback between economic policy and the electorate. The inconsistency of structural policies might reflect policymakers’ struggle to adapt the welfare state to major shocks, such as unification, globalization, and the recent growth slowdown in Germany. The next sections explore some avenues to improve the prospects for fiscal consolidation and structural reform: strengthening budgetary institutions (Section D) or changing intergovernmental fiscal relations (Section E).

D. The Role of Budgetary Institutions

123. Empirical evidence suggests that budgetary institutions can have an important effect on fiscal policy performance. According to Alesina and Perotti (1995), “Germany’s voting rules are actually among the least compatible (at least on paper) with fiscal responsibility.” However, the country ranks relatively high on the efficiency of its budgetary institutions, which helps also in overcoming the biases stemming from fiscal illusion. This section briefly describes the key features of current budgetary institutions and options for reform.

Budgetary institutions: The status

124. Many aspects of the budgetary institutions in Germany promote fiscal responsibility.50 The IMF fiscal ROSC concluded that Germany has achieved a high level of fiscal transparency (IMF Country Report No. 03/286). In particular, sound standards for budgeting, accounting, and reporting apply to all levels of government; multiyear budget preparation is an integral part of the process; and fiscal reporting includes contingent liabilities, guarantees, tax expenditures, and equity holdings. Nonetheless, the fiscal ROSC pointed to significant scope for strengthening the budgetary institutions.

125. Reliance on outside experts for elaborating the macroeconomic framework and the tax revenue projections plays an important role in stemming fiscal illusion. Each January, the Ministry of Economy and Labor elaborates a macroeconomic framework in its Annual Economic Report (Jahreswirtschaftsbericht). This is followed, typically in March, by the consensus macroeconomic projections of Germany’s main economic research institutes. This consensus framework serves as input for the medium-term tax forecast, another consensus projection prepared by the Working Group for Tax Estimates (Arbeitskreis Steuerschätzung). This group includes representatives from the Federal and Länder Ministries of Finance, local government, the Bundesbank, the main research institutes, and the Council of Economic Experts.51 The tax estimates are incorporated in the Bund and, partly, in the Länder budgets. All estimates are updated in the fall, before the budget’s final reading and adoption (typically) in December.

126. Furthermore, the Basic Law and the Law to Promote Economic Stability and Growth commit all levels of government to coordinated fiscal management. For example, borrowing by the Bund is limited to the amount of the gross investment in the budget (golden rule) unless the government declares that the economy is out of equilibrium and that more borrowing will help in redressing the disequilibrium. The same holds for many Länder as well as, in a tighter form, for all the municipalities, whose budgets are approved by their respective Land parliament. Furthermore, policies should be coordinated between all levels of government. In practice this is done by the Joint Financial Planning Commission (Finanzplanungsrat) that meets twice a year following the meeting of the Working Group for Tax Estimates. The commission’s task is to reach an understanding on the broad budgetary targets for Bund, Länder, and Municipalities.

Budgetary institutions: Scope for reform

127. These budgetary institutions were strong enough to forestall major fiscal excess in any given year but did not stem persistent fiscal deficits and rising public debt. A continued increase in the public debt-to-GDP ratio and a growth-stifling tax and contributions burden—which is a likely scenario for the long run absent any further policy change—would not be in the spirit of the laws covering fiscal responsibility.52 Better budgetary institutions might help in forestalling such a scenario, by fostering an earlier adoption of reforms. Three avenues for reform could be considered: more transparency and accountability; better coordination; and stronger rules.

Transparency and accountability

128. A promising first step might be to raise transparency and accountability. This could be achieved through several measures that combat fiscal illusion and clarify intra- and intergenerational redistribution:

  • Shifting budgetary accounting and planning to an ESA 1995 national accounts basis. Currently, the budget is prepared on the basis of cash-based accounting dating back to 1969. The ESA 1995 (accrual) presentation of the general government’s accounts is considered the broadest and most accurate measure of a country’s general government balance, expenditures, and revenues. Moreover, Germany’s commitments under the Stability and Growth Pact (SGP) are monitored on the basis of ESA 1995 data. Accordingly, it would be natural to prepare Bund and Länder budgets also on that basis.53

  • Producing long-run projections (say through 2050) for general government deficits, expenditure, revenue, and tax and contribution rates on a “current services” basis. While outside experts produce medium-run projections for tax revenue on such a basis (Working Group for Tax Estimates), they do not do so for the long run, for social security receipts, or for expenditure. Such projections for the Bund, Länder, and the social security system would reveal the full future burden of current fiscal and social policies, fostering a more informed discussion about reform. They would also facilitate assessing the consistency of current policies with sustained and balanced growth, which is stipulated by the laws covering fiscal responsibility. Such projections should ideally be produced for parliament by independent experts.

  • Explaining in the budget documents how the budgetary and other measures link the official targets for the general government with the long-run current services projections.


129. The coordination of Bund and Länder budgeting could be increased, through reforms to the Internal Stability Pact. In 2002, Germany adopted an Internal Stability Pact (ISP) to improve Bund-Länder fiscal policy coordination. However, this pact could be made more transparent and binding in several respects. This is particularly important when the majorities in lower and upper chambers differ.

130. The ISP established several objectives. First, Bund and Länder (including municipalities) agreed on the need to return budgets back to balance and this objective was made part of the laws defining fiscal responsibility. Second, they agreed that the deficit target of the Bund and all Länder combined could reasonably be split 45:55 percent between Bund and all Länder, respectively. Third, under the pact the Bund would reduce its expenditure by ½ percent per annum during 2003-04 while the Länder (as a group and including the municipalities) would keep annual expenditure growth within 1 percent. However, the pact did not spell out specific deficit or expenditure targets for each Land (Table IV-3). This, in turn, makes it difficult for the Joint Financial Planning Commission to issue recommendations for corrective action, except in egregious cases. Furthermore, its recommendations would not be binding.54

Table IV-2.

Germany: General Government Balance and Public Investment, 1992-2003 1/

(In percent of GDP)

article image
Source: Federal Statistical Agency.

Data for net investment for 2000-03 assume same depreciation (in percent of GDP) as for 1999. Data through 1999 are taken from Wendorff (2001).

Table IV-3.

Germany: Distribution of Cash Deficit, 1992-2002

(In percent)

article image
Source: Federal Statistical Office, ESA 1995 data.

In 2000, excluding UMTS receipts.

131. The ISP offers a useful starting point to improve coordination between Bund and Länder. The roles and responsibilities of various levels of government in fiscal policy need to be agreed on and spelt out clearly. Specifically, the annual and medium-term budget planning of the different levels of government needs to be better integrated, which may require some Länder to shift from two-year budgets to one-year budgets. Based on a common macroeconomic framework, specific expenditure and deficit targets should be proposed by the Bund and each Land in support of the official objectives for the general government balance, and all should be held accountable for the achievement of these targets. Furthermore, the proposed expenditure and deficit paths would have to be approved by the federal and regional parliaments to ensure ownership.55 Mechanisms to sanction Bund or Länder that do not respect their ISP commitments might have to be considered if increased transparency fails.

Budgetary rules

132. An alternative avenue for reform would be to strengthen the budgetary rules that already apply to Bund and Länder. The golden rules are not compatible with the rules under the SGP. The SGP calls for general government fiscal balance over the cycle; deficits under 3 percent of GDP, except under certain circumstances; and a debt ratio that, if not falling, is under 60 percent of GDP. While the SGP has already found its way into German budgetary law, it has done so only in the form of a general call on Bund and Länder to return to their budgets to balance. Perhaps the golden rules that govern budgeting at Bund and Länder level should be replaced with the SGP’s rules, although this raises complexities on account of the intergovernmental fiscal relations. Alternatively, they could be tightened, by focusing on ESA 1995 government deficits and by adopting a stricter definition of investment expenditure, namely net investment as defined under ESA 1995. Such a definition would better capture the spirit of a golden rule. Wendorff (2001) points out that on such a basis the golden rule for Germany need not be inconsistent with a balanced budget rule: the ESA 1995 data suggest that net investment by the public sector was close to zero ever since the mid-1990s (Table IV-2).

133. The experience of other countries suggests that binding fiscal rules can be helpful in stemming expenditure growth and deficits.56 In some countries, local governments have independently chosen fiscal rules. In others, the rules have grown out of a cooperation between federal and regional governments and the monitoring and potential sanctioning for breaching rules is also done in a cooperative manner. Daban and others (2003) review the ISPs or ISP-like frameworks that have been adopted by Austria, Belgium, Italy, and Spain. All these frameworks set ceilings on deficit or debt and they usually do so for the federal government and each local government. Austria has chosen to allocate the Maastricht deficit between the federal government and the regions, mainly as a function of their populations. All these frameworks, except that of Spain, provide for transparent sanctions in case of noncompliance with the targets (in the case of Italy, a benefit for compliance), although they typically can only be invoked by a unanimous decision of all parties.

E. Reforming Bund-Länder Economic Relations

134. Broader reforms to the Bund-Länder relations could also help in stemming expenditure pressures and fostering fiscal consolidation. They can do so if they: (i) raise the transparency of the intergovernmental relations; (ii) better align expenditure with taxation responsibilities; (iii) inject some scope for tax and expenditure policy competition between the Länder; and (iv) allow for some Länder-specific experimentation with reform that, if successful, can foster progress across the entire country (Roland, 2001). Moves in such a direction are being explored by a parliamentary commission.

Interregional redistribution

135. Interregional redistribution is common in federal states for reasons related to equity and economic efficiency. Resources are redistributed vertically (from the central to the local governments or vice-versa), for example, when tax receipts accruing to different levels of government do not match expenditure mandates. Resources might be redistributed horizontally (between various local governments) to support an efficient distribution of public goods, to insure against region-specific income fluctuations, or to align living standards. In Germany, equity is a key consideration guiding redistribution.

136. However, the mechanism by which fiscal resources between Bund and Länder are apportioned does not provide strong incentives to contain expenditure or raise revenue (Box IV-3). Many sources provide a detailed review of the functioning of the mechanism (e.g., Federal Ministry of Finance, 2003; Lenk 2003; and Baretti and others, 2000). Its key features have undesirable incentive effects, for various reasons:

  • The vertical revenue sharing is guided by past expenditure. Thus higher spending might be rewarded with more revenue. In addition, it includes a horizontal equalization component. This detracts from the transparency of the redistribution that takes place and thereby potentially fosters fiscal illusion.

  • Redistribution almost completely equalizes the per capita resources of all the Länder (Table IV-4). While some differences remain—largely because municipal tax revenue is not fully considered in a Land’s revenue capacity—these differences are small.57 Furthermore the rationales for excluding part of the municipal taxes in measuring revenue capacity and for the special treatment of city states are not clear.

Intergovernmental Redistribution

Tax sharing. About 70% of all fiscal revenues are shared between Bund and Länder (Table IV-5). The sharing coefficients for income and savings taxes are fixed while that for valued added tax is determined periodically, so as to ensure that each level of government can cover “necessary expenditure” (Table IV-6). Because there is no objective definition of “necessary expenditure,” in practice the sharing of the value added tax receipts is a function of the history of expenditure and Bund-Länder bargaining.

Table IV-4.

Germany: Revenues Before and After Redistribution and GDP, 2001

(Per capita revenue, in percent of national average)

article image
Source: Federal Ministry of Finance. Bold face indicates a city state.
Table IV-5.

Germany: Tax Receipts of Bund and Länder, 2002

article image
Source: Federal Ministry of Finance, 2003.
Table IV-6.

Germany: Bund-Länder Repartition of Tax Revenue, 1992-2001

article image
Source: Federal Ministry of Finance, 2003.

For Länder, including Erganzungszuweisungen; excluding payments into Fonds “DE”; reverse for Bund.

Umsatzsteuervorwegausgleich. Income taxes are distributed across Länder according to the “residency” of the tax payers. For value-added tax: (i) at least 75 percent is distributed according to population; (ii) at most 25 percent is distributed to Länder with a lower-than-average fiscal revenue per capita—computed including receipts from shared taxes and Länder taxes but excluding VAT and municipal taxes. Each Land is to reach at least 92 percent of the average per capita revenue so defined. As of 2005, Länder with less than 97 percent of average per capita revenue will have 95 percent of their gaps filled; those with between 97 percent and less than 100 percent will have their gaps filled between 95 and 60 percent.

Finanzausgleich. Formal horizontal redistribution aligns the revenue-raising capacity (RC) of each land with an equalization index (EI). RC is the revenue per capita except that only 50 percent of the municipal tax receipts are considered. Starting in 2005, 64 percent of municipal tax receipts will be considered (and some 12 percent of the receipts stemming from above-average revenue growth will no longer be included in the measurement of the RC). The EI essentially corresponds to the average level of revenue per capita but with city states receiving a 135 percent population weight.

Thus far, for Länder with RC<EI, equalization transfers bring them up to 92 percent of the EI; in addition, they fill 37.5 percent of the gap between 92 and 100 percent of the EI. The Länder with RC>EI and 1 percent (10 percent) higher than average RC surrender 33 percent (80 percent) of the excess. Starting 2005, horizontal equalization is more complex: (i) for those Länder with RC<0.8EI, 75 percent of the gap between RC and EI is filled; (ii) if 0.8EI<RC<0.93EI, then 75 percent down to 70 percent is filled (linearly falling); and (iii) if 0.93EI<RC<EI, then 70 percent down to 44 percent is filled (linearly falling). Furthermore, between 44-70 percent of the difference between RC and EI is surrendered if RC>EI, with the exact amount rising gradually between a revenue capacity of 100 to 120 percent.

Bundesergänzungszuweisungen. Additional transfers from Bund to weaker Länder ensure that they reach at least 99.5 percent of the average RC. As of 2005, the transfers will only cover 77.5 percent of the gap that remains relative to 99.5 percent of RC. Furthermore, special transfers are made, mostly to the new Länder to overcome burdens that remain from unification.

Tax policy coordination versus competition

137. Aside from the limited incentives to do so, virtually no scope exists for the Länder to practice an independent tax policy. Most tax revenue is shared between Bund and Länder, levied on common tax bases at common tax rates. Some tax receipts flow only to the Länder (see Table IV-4). But tax bases and rates are the same across Länder. Municipalities levy two taxes: a local business tax (the Gewerbesteuer) and a tax on land (the Grundsteuer). For both the definitions of the tax bases are the same across municipalities but the rates can be set independently. In 2002, receipts from these taxes amounted to about 1½ percent of GDP.

138. The reason for harmonizing taxation across the country is to avoid tax competition and complicating tax administration. Tax competition might lead to falling taxation on mobile factors of production. The result might be an underprovison of public goods. Furthermore, different taxes, tax bases, and tax rates can complicate tax administration, resulting in efficiency losses.

139. Nonetheless, there are also good reasons for allowing local governments somewhat greater autonomy in setting tax policy. The degree of tax harmonization across jurisdictions is unusually high in Germany compared with other countries with a federal system (e.g., Canada, and Switzerland). First, more autonomy would allow a closer link between tax policy and the provision of public goods, many of which might not have effects that spill across several jurisdictions. Second, some tax competition might be a desirable counterweight to the “political economy” biases toward a larger government. Third, tax autonomy might help the new Länder to attract business, arrest the emigration of their workforce, and thereby foster a more efficient use of land and existing infrastructure.58

Region-specific expenditure policy

140. More tax autonomy makes sense only together with more freedom for expenditure policy.59 Large expenditure components that concern all levels of government are determined by national policy (e.g., pay of public sector employees and social policies). Nonetheless, Bund and Länder are autonomous in drawing up and implementing their budgets, although Bund budgets often hinge on tax or structural changes that require Länder approval. The Bund takes charge exclusively of national defense and external affairs (policymaking and implementation), employing about 12 percent of all government personnel and accounting for about one third of government expenditure (including federal special funds). The Länder handle exclusively law enforcement, education, culture, sports and leisure, residential dwellings, zoning, and various services. Several programs are decided and implemented jointly by Bund and Länder. These concern regional infrastructure and education and their total cost is about ½ percent of GDP.60 Otherwise, most programs are designed jointly at the national level but are implemented either exclusively by the Bund or exclusively by the Länder.

141. Recently, more scope for regional policy differentiation has been introduced and further steps in such a direction might be desirable. Recently the Länder began to differentiate the wage supplements of public employees (e.g., holiday pay).61 Since living costs and labor market conditions differ across Länder, additional scope for wage differentiation appears appropriate. More scope for differentiating welfare policies has also been brought up for discussion. Differences in welfare policies exist between different jurisdictions in some countries. They might help structurally disadvantaged Länder in charging lower taxes to boost economic development. In short, they would allow tailoring the social policies to local circumstances. In so doing, they might also help to limit the aforementioned “political economy” biases toward higher government spending.

Potential avenues for reforming Bund-Länder fiscal relations

142. Reforms can contribute to improving the prospects for fiscal consolidation, structural reforms, and the interregional convergence of living standards. Over the past couple of decades there has been no significant convergence among the old Länder (Figure IV-7).62 Reform options may include:

  • Increasing the transparency of revenue redistribution and redesigning it to provide better incentives for fiscal consolidation.

  • Raising the scope for the Länder to follow independent expenditure and tax policies. More room for wage differentiation in the public sector would lead to compensation packages that better fit local living and labor market conditions. The same holds for welfare policies. Regarding taxes, Länder could be allowed to levy surcharges or offer discounts on personal income tax rates, possibly within some broad ranges to avoid harmful tax competition.

Figure IV-7.
Figure IV-7.

Germany: Convergence

(Income per capita, in percent of national average)

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004

Source: Federal Statistical Office.

143. The high degree of cooperation that has emerged is not considered the only form of federalism that is compatible with Germany’s Basic Law. As noted, the interpretation of the “same living standards” condition in the Basic Law and its economic content are subject to debate. Many argue that the Basic Law does not seek uniformity across Länder, nor does it tolerate divergences across Länder that risk undermining their legal and economic union. For economic policymakers this raises the issue of striking the right balance between equity and efficiency considerations, with the latter arguing for a measured degree of competition in fiscal federalism.

144. In conclusion, staggered regional elections have changed the majorities in the upper chamber between general elections. This has had implications for both fiscal consolidation and also structural reforms. This phenomenon puts greater emphasis on ensuring that budget institutions and intergovernmental fiscal relations provide better incentives for good fiscal management.

Appendix Indicator of Structural Reforms

145. To quantify the forces that influence reforms in Germany, an indicator of structural reform was constructed as a first attempt that can be refined over time. The sources are the chronologies of economic policy in annual reports of the Bundesbank, the Sachverstandigenrat, Heilemann and others. (2003), and Steffen (2000). Laws and decrees that affect primarily the supply side of the economy are considered structural reforms. This sets them apart from changes, e.g., in interest rates and public spending that target the demand side. To quantify the scope of a reform, each measure is given a value from 1 to 4, where a higher number means a more wide-ranging reform. Finally, all reforms are classified as to whether they expand the role of the state (positive sign) or expand the role of markets (negative sign).

146. Table IV-1 shows a comprehensive list of structural reform measures from 1970 to 2004. The list is visualized in the figures below, which show the changes and the level of the reform indicator, respectively. While the 1970s experienced the last major expansion of the welfare state, the tide began to turn in the early 1980s. Since then, reforms mostly focused deregulating markets, privatizing and more generally reducing the role of the state. However, this trend was not uniform, and structural reforms became more erratic in the mid-1990s.


Reforms: Changes

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004


Reforms: Level

Citation: IMF Staff Country Reports 2004, 340; 10.5089/9781451810455.002.A004


  • Alesina, A., and R. Perotti, 1995, “The Political Economy of Budget Deficits,IMF Staff Papers, Vol. 42, pp. 131.

  • Alesina A., G. Cohen, and N. Roubini, 1992, Macroeconomic Policy and Election in OECD Economies,Economics and Politics, Vol. 4, pp. 30.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Alesina A., G. Cohen, and N. Roubini, 1993, “Electoral Business Cycles in Industrial Democracies,European Journal of Political Economy, Vol. 9, pp. 125.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Baretti C., R. Fenge, B. Huber, W. Leibfritz, M. Steinherr, 2000, Chancen und Grenzen des Föderalen Wettbewerbs, IFO.

  • Bundesbank, Annual Reports 1970-2003. Frankfurt.

  • Daban T., E. Detragiache, G. di Bella, G. M. Milesi-Ferretti, and S. Symansky, 2003, “Rules-Based Fiscal Policy in France, Germany, Italy, and Spain,IMF Occasional Paper No. 225 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Federal Ministry of Finance, 2003, Bund-Länder Finanzbeziehungen auf der Grundlage der Geltenden Finanzverfassungordunung.

  • Friedman, M., Should There be an Independent Monetary Authority,” in L.B. Yeager (ed.) In Search of Monetary Constitution, (Harvard University Press, 1962.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Heilemann, U., Gebhardt, H., Loeffelholz, H., (2003): Wirtschaftspolitische Chronik 19492002. (Stuttgart: Lucius & Lucius).

  • Lenk T., 2003, Finanzausgleich in der Bundesrepublik Deutschland, mimeo, University of Leipzig.

  • Perotti R., and Y. Kontopoulos, 2002, “Fragmented Fiscal Policy,Journal of Public Economics, Vol. 86, pp. 191222.

  • Persson T., and G. Tabellini, 2004, Constitutional Rules and Fiscal Policy Outcomes, American Economic Review, Vol. 94, pp. 2545.

  • Poterba J., 1994, “State Responses to Fiscal Crisis: The Effects of Budgetary Institutions and Politics,Journal of Political Economy, Vol. 102, pp. 799821.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roland G., 2001, “The Political Economy of Transition,” mimeo, University of California, Berkeley.

  • Roubini N., and J. Sachs, 1989a, “Political and Economic Determinants of Budget Deficits in Industrial Countries,European Economic Review, Vol. 33, pp. 903938.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roubini N., and J. Sachs, 1989b, “Government Spending and Budget Deficits in Industrialized Countries,Economic Policy, No. 8, pp. 99132.

    • Search Google Scholar
    • Export Citation
  • Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung: Jahresgutachten, 1971 to 2003. (Wiesbaden: DeStatis).

  • Spahn, P.B., 2000, Consensus Democracy and Interjurisdictional Fiscal Solidarity in Germany, mimeo, IMF.

  • Steffen, J., 2000, Sozialpolitische Chronik. (Bremen: Arbeiterkammer).

  • Von Hagen, J., 1992, “Budgeting Procedures and Fiscal Policy Performance in the European Communities,Economic Papers, European Commission.

    • Search Google Scholar
    • Export Citation
  • Wendorff, K., 2001, “The Discussion on a National Stability Pact in Germany,in Fiscal Rules, Proceedings of the Research Department Public Finance Workshop on Fiscal Rules, Bank of Italy.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wyplosz, C., (1993): After the Honeymoon: On the Economics and the Politics of Economic Transformation. European Economic Review, pp. 379386.

    • Search Google Scholar
    • Export Citation

Prepared by Jörg Decressin and Benedikt Braumann.


A key objective is to combat “fiscal illusion.” See Friedman (1962) for a fuller development of this argument in the context of monetary policy rules.


The interpretation of the “same living standards” wording is subject to considerable debate (see para. 143).


The exact distinction between laws that require approval of both chambers and those that do not is subject to some debate in Germany and a Bund-Länder commission is investigating the matter.


Parties need to have either three direct seats or 5 percent of the votes cast to be represented as a parliamentary group.


For a broad review, see Alesina and Perotti (1995).


They may do so either systematically or only ahead of key elections, giving rise to political business cycles (Alesina, Cohen, and Roubini, 1992 and 1993).


During the 1970s and 1980s, Länder with a coastal line (e.g., Bremen, Niedersachsen, Schleswig-Holstein) and smokestack industries (e.g., Saarland) struggled with structural change. The challenges raised by reunification are considerably larger. Throughout the postwar history the interests of city states, for obvious reasons, differed from those of the other Länder.


By and large, governments in Germany have been fairly centrist. The use of the terms “left leaning” and “right leaning” in this paper therefore is simply convenient shorthand for denoting relative tendencies at the margin rather than an absolute judgement.


In contrast, strong (united) governments have been able to cut expenditures (1980s and 2000), or to raise taxes (1970s)—see Figure IV-5.


The latter comprises five professors, appointed by the government to the council, who draft an independent, annual assessment of the German economy, which is released in November.


See Chapter IV of IMF Staff Country Report 021240, and Chapter III in this volume.


Compliance with the golden rule could still be monitored with such data, with public investment defined accordingly. For the general government this would have revealed frequent violations of the golden rule during the 1990s (Table IV-2).


Preliminary data through 2003 suggest that the Bund missed its expenditure target, largely because of the Bund’s contributions to the social security system, while the Länder achieved it. The deficits targets of the Bund and Länder (combined) were appreciably overshot.


Similar suggestions have been made by a commission of experts, appointed by the Ministry of Finance.


Even upon fully considering municipal revenue, some calculations suggest that each Land reaches at least 89.5 percent of average per capita revenue; under the new, 2005 system the figure will be 91.2 percent (Lenk, 2003). Matters differ somewhat for the city states because of the population weights.


Providing such incentives with tax rates might be more effective than doing so with subsidies.


The Länder can vary the Gewerbesteuer (local business tax) but its tax base likely is fairly mobile, unlike the tax base of personal income tax rates, for example.


Public sector employees in the new Länder receive a lower wage than those in the old Länder although the difference is set to be phased out.


This is not to deny that government transfers to the new Länder likely have been helpful in fostering their rapid catch-up during the first half of the 1990s.

Germany: Selected Issues
Author: International Monetary Fund