The OECD Observer
October/November 1998, No. 214

 

Germany—Reforming federal fiscal relations
By Eckhard Wurzel

 

In Germany fiscal relations between the various levels of government have come to the fore of the policy debate. The fiscal system can encourage overspending in some areas, while it appears to have had little success in achieving economic convergence between the Länder. Reform is therefore needed. (OECD Economic Surveys: Germany, OECD Publications, Paris, 1998 (http://www.oecd.org/eco/surv/esu-ger.htm))

The high fiscal burden arising from the integration of the new eastern states into Germany has intensified discussion of whether or not the system of inter-governmental transfers discourages the federal states (Länder) from improving their finances. Furthermore, there are questions about how to spread the burden of fiscal consolidation implied by the Maastricht treaty between the different levels of government. And there is a third question of whether a leaner and more efficient public sector would be achieved by more closely matching local and regional public spending with state and local tax responsibilities.

In assigning tasks to the different levels of government and stipulating the budgetary independence of the federal government (Bund) and the Länder, the German constitution broadly follows the subsidiarity principle. But in practice the functional segregation of responsibilities is not a strict one. Co-operation and policy co-ordination are an important feature of fiscal relations. The Länder have an important input into federal decision-making through the upper house of parliament, the Bundesrat, while policy co-ordination is institutionalised in a number of inter-governmental councils. Some co-financing of Länder tasks by the Bund—covering jointly planned projects, investment aid and regional and social transfers—reflects the weight Germany attaches to attaining broadly similar living standards across the federation. Furthermore, a fiscal equalisation system—or Finanzansgleich—aims at providing fairer revenue levels across states, or for the communities within a state. Taxing powers are largely centralised and the degree of sharing of tax revenue between the different layers of government has increased over recent decades.

 

The fiscal equalisation system

Under the fiscal equalisation system, financially weak Länder receive transfers from wealthier states and from the federal government. This yields a substantial redistribution of income in favour of the poorer Länder. Yet, it appears to have had little success in achieving economic convergence as measured by the Länder’s GDP per capita, with the variation between the old Länder of real GDP per capita having declined little. In statistical terms, the standard deviation remained at around one fourth of the mean over the last 27 years (Figure).

The reason for this disappointing outcome appears to be one of disincentives within the system. Consider what happens under the rules of tax sharing and subsequent redistribution of revenues via the equalisation system to additional tax revenues generated by an individual state. An additional DM1 million in income tax receipts generates only between DM80,000 and DM290,000 in extra income for that state. The remainder is allocated to the Bund and other Länder. So for wealthy states most of the additional revenues are transferred to poorer states, while for less wealthy states additional revenues reduce their eligibility for extra transfers. Thus the system reduces the incentive for an individual state, new or old, to build the conditions required to attract corporate investment and increase its GDP and its tax base. Moreover, even tax auditing by an individual Land may appear uneconomic, since it gets only part of any extra tax generated, but must bear the full cost of the audit itself.

A system which moderated this reduction in Länder revenues would improve public sector efficiency without necessarily compromising the basic aim of the fiscal equalisation process. One such reform could be a two-stage system which allocates lump-sum payments, fixed over a multi-annual period, to poorer states in stage one, and in stage two conditions the current redistribution of taxes between states on their financial capacity as it prevails after stage one. Introducing such lump-sum payments would allow lowering the rates of tax redistribution in comparison to the present system.

 

Implementing the stability and growth pact

With Germany obliged to observe the Maastricht treaty’s budgetary criterion, it has been argued that the Länder could run excessive fiscal deficits, while the Bund would face any consequent sanction. The excessive deficits procedure was agreed by EU leaders in Dublin in 1996, with the objective of preventing excessive general government deficits among those EU governments participating in the single currency and, if they occur, to ensure their prompt correction, probably by imposing stiff financial penalties. Germany’s federal government has therefore proposed budget caps for the Bund and the states, which would apply in the event of an excessive budget deficit. Evidence from elsewhere suggests that such caps help strengthen budgetary consolidation.

While poorer Länder would argue for the caps to be set in line with their higher financial need as evidenced by large deficits, states with low deficits argue that this would be another disincentive to use resources efficiently. For them, a ‘per capita’ allocation of the deficit ceilings would be preferable on efficiency grounds, but the deficits of the new Länder would then have to fall significantly; in 1997 the eastern states accounted for 21*% of the German population but for 35% of the overall Länder deficit. Hence, to avoid cuts in investment outlays in the new Länder, seen as necessary to catch up with the west, a transitional arrangement would be required to allow the new Länder to move gradually from present deficit positions to per capita-based deficit allocations.

 

The case for more state autonomy

Economic efficiency would justify co-funding of fiscal projects if externalities existed between governments. Leaving financing exclusively to the government which provides the goods could then lead to under-provision. Bund participation in funding the construction of a university, for example, is justifiable as long as the university in question conveys positive externalities to other states in terms of the education provided. However, even in this example the mere existence of externalities does not automatically demand the involvement of the federal government, since part of the university’s construction costs could be transferred to the states that benefit from it by introducing fees for the services. After all, in some fields services can be provided more efficiently if governments co-operate or are combined into larger administrative units. But if the externalities are minimal, federal grants end up encouraging overspending by the recipient Länder, since the full costs of providing goods and services are not then reflected in their budgets. Effectively, federal grants can allow the Länder governments to provide services and benefits and to pass on part of the associated costs to taxpayers in other states. Thus, in Germany the Länder are responsible for investment in hospitals, but not for financing their running costs or covering their deficits. That gives a strong incentive to expand hospital capacity, to create local employment, for example. Overall, co-financing in Germany appears excessive and should be reduced.

Similarly, responsibility for financing the social benefits could be given entirely to either the Länder (and communities) or to the Bund, depending on whether they are judged to be supplementary at the regional level, or whether complete standardisation of provision is considered to be desirable. If some regional discretion is desired over benefits which should be provided across the federation, the Bund could fully finance basic provision, which the Länder or communities could then top up.

In addition, more regional tax autonomy would typically engender pressure for tax or expenditure savings. Regional tax autonomy could be increased by allowing for a Länder-specific or community-specific surcharge on the income tax. However, in such a system the underlying income tax schedule would have to be lowered. While compliance costs could be higher than in a completely harmonised system, restricting regional tax autonomy to variations in the surcharge would require virtually no additional expertise on the side of the tax payer and would keep additional administrative costs for the tax authorities to a minimum.

Similarly, incentives to reduce the tax burden and to balance the costs and benefits of publicly-provided goods would be strengthened by a shift to a system which relied more on charging fees for services. This would match public service provision more to local demand, because it would introduce incentives for customers to ask for services that were tailored to their preferences.