Isabelle Joumard1 Problems of Equalisation in Federal Systems – The Concept of Equalisation (Work Session 10) 1. Introduction The need for fiscal consolidation and the search for higher public sector efficiency have led many countries to reconsider the assignment of spending responsibilities and revenue-raising powers across government levels, as well as their inter-governmental grant system. Revenue-raising powers of subnational governments have recently been raised in a number of countries, including Spain, while efforts to reduce the overlapping of spending responsibilities are currently under consideration in Switzerland. One key objective is to reach a better congruence between them, thus allowing a more consistent implementation of one of the basic principles of fiscal federalism – “(s)he who decides, spends and pays”. Such moves would probably promote sub-national governments’ accountability and contribute to reducing the size of inter-governmental transfers. However, they leave open the question of how poor jurisdictions can finance high quality public services with tiny tax bases. Another salient feature of recent reforms in fiscal relations across government levels is the move from earmarked to block grants, as earmarked grants in many cases have failed to promote public sector efficiency. Canada and the Nordic countries have replaced most earmarked grants by block grants over the past two decades while in the United States, the trend towards block grants initiated in the 1970s has continued in the 1990s. However, in many countries earmarked grants often contain a re-distributive component. In this context (less reliance on earmarked grants and a better match between spending and revenue-raising powers), discussions on the role and need for redistribution across jurisdictions are brought to the fore. Whether or not fiscal equalisation schemes entail a conflict between efficiency and equity objectives is an important aspect to take into account. 2. Reassessing the assignment of spending responsibilities and revenue-raising powers across government levels Many participants heralded a de facto and progressive erosion of effective powers of sub-national governments as central governments tend to introduce binding minimum standards for the domains that are formally the responsibility of sub-national governments. As an illustration, the German federal government has a strong influence on the design of the tertiary education system, which used to be the sole responsibility of the Länder. More recently, the poor educational outcomes, as measured by the Programme for International Student Assessment (PISA) international comparisons, have led the central government to plan the introduction of national standards in primary and secondary education (OECD, 2001a). The Canadian federal government has also had a history of influencing the design of the healthcare system despite the fact that the provinces have the constitutional responsibility for healthcare (OECD, 2001b). The system of cost sharing tied to specific expenditure was abandoned in the 1990s. However, the central government has made federal funding to the provinces conditional on respect for comprehensiveness and universality in the healthcare system. Swiss cantons were also reported to have lost effective control over some of their spending programmes, and to be increasingly acting as the operating agents of the central government (described as a “centralisation process by the back door”). A similar development occurs within the European Union (EU) area, where decisions taken at an EU level cover an increasing range of domains (e.g. water treatment), while the costs are being borne by the individual countries, in many cases at a local or regional level. The overlapping of spending responsibilities across government levels is often accompanied by an extensive reliance on inter-governmental transfers. These two features of fiscal federal relations have raised serious efficiency concerns: they reduce the autonomy and accountability of sub-national governments and they dull incentives to contain costs and excessive demands by generating “fiscal illusion” (as sub-national governments and/or local citizens do not bear the full cost of local public services). Many participants suggested that realigning spending responsibilities and revenue-raising power of subnational government levels would contribute to an easing of these tensions. In fact, new taxing powers have recently been granted to sub-national government levels in a number of countries. As an illustration, Spanish regions can now set tax rates on 30% of the personal income tax base and introduce tax credits. Some participants argued in favour of granting more taxing powers to sub-national governments in Germany and Poland, for example by allowing them to raise surcharges on federal taxes. 3. Could such a reassessment be a substitute for equalisation transfers? A regular reassessment of revenue-raising powers and spending responsibilities, and their better alignment, would probably help to lower reliance on extensive inter-governmental transfers. As an example, the Swiss authorities have put forward an ambitious program to revamp spending assignments across government levels. The project aims at reducing existing overlapping in cantonal and federal spending responsibilities by applying the subsidiarity and fiscal equivalence principle more consistently. If consistently implemented, it would allow the cantons to tailor the supply of pure “local public goods” more closely to their residents’ needs and willingness to pay. However, the reassessment of revenue-raising powers and spending responsibilities is often a lengthy process. In the Swiss case, a wide-ranging discussion and negotiating process was launched in the early 1990s and the project, if approved by the cantons and the Swiss population, is expected to enter into force in 2006. Large fiscal disparities associated with local taxation and provision of public goods across jurisdictions and externalities are likely to persist, however, together with vertical and horizontal imbalances. In this context, reconsidering the assignment of spending and revenue-raising responsibilities across government levels could not be a substitute for fiscal equalisation. For instance, the proposed reform in Switzerland, if adopted, would be accompanied by a strengthened emphasis on equalisation, in the form of block grants. International comparisons indeed suggest that the vertical fiscal gap – measured by the share of sub-national governments in total public spending compared with their tax revenue (and the corresponding intergovernmental transfers) – tends to be greater in countries where sub-national governments are responsible for a large share of public spending. In particular, the role of fiscal equalisation schemes is likely to increase when sub-national governments are responsible for the provision of social and redistributive policies and when securing roughly equal living conditions and/or public services is a principle embedded in the constitution (e.g. in Canada, Germany, Spain and Switzerland) and/or supported by the population. While formal fiscal equalisation schemes mobilise a significant amount of public resources in many countries, redistribution also stems from a number of other mechanisms. In fact, formal equalisation schemes no longer exist in the United States: they were abolished for states in 1981, and for cities in 1986 (Dafflon and Vaillancourt, 2002). However, some spending programmes managed by the states are largely financed through conditional grants which contain an element of fiscal equalisation (i.e. the federal contribution rate in state spending varies from one state to another to reflect their income per capita). Medicaid is the most prominent example, but roughly the same principle applies for the State Children Health Insurance Programme introduced in 1997.2 Conditional grants with equalisation supplements have also been used widely in Switzerland. Progressive federal income taxes also introduce some form of redistribution by taking more from jurisdictions with wealthier inhabitants. In addition, national social insurance schemes – and in particular unemployment insurance – do not set differentiated risk premium across regions. In Canada, the national unemployment insurance program further discriminates in favour of high-unemployment regions (concentrated in the Atlantic provinces and Quebec) by granting more generous benefits for the same contribution rates (OECD, 2001b). 4. If a reassessment of spending and revenue-raising responsibilities cannot be a substitute for equalisation transfers, why should the central government or rich regions be willing to support poorer ones? Participants widely acknowledged that most politicians and a large part of the population do agree on the need to ensure that citizens have access to reasonably comparable levels of public services at a similar level of taxes – i.e. the basic solidarity and equity objectives of equalisation transfers. As an illustration, a 2001 survey revealed that the Canadian equalisation program is strongly supported by 45% of Canadians, and moderately supported by another 38% (Environmics – CROP – CRIC). Broadly similar results were obtained for the rich provinces of Ontario and British Colombia, which are both net contributors to the program. An important factor influencing the view of citizens is the (perceived) probability that they (or their children) could move to a poorer province in the future. Thus they desire some guarantees on the provision of public goods of a comparable quality throughout the country. Furthermore, rich Canadian provinces often ally with the poor ones in their negotiations with the central government to increase the scope of fiscal equalisation. Indeed, they prefer these unconditional transfers to other forms of the central government intervention, in particular conditional grants. Although, some participants noted that vertical fiscal equalisation – such as the scheme in place in Canada where, at least formally, the central government pays for the costs of fiscal equalisation – was likely to gain more political support than horizontal equalisation schemes. This is of particular concern in Germany where tax revenues are redistributed both vertically and horizontally, and where political debates are mostly focused on the horizontal component.3 Specific historical and sociological factors were considered to make fiscal equalisation legitimate. The separation of Germany after the Second World War significantly affected the growth prospects of some eastern Länder in western Germany. Equalisation transfers partly compensated the most affected regions. Later, the fall of the wall in Germany could have led to mass migration from eastern and western Länder, which was widely considered as an undesirable development. Generous financial transfers to eastern Länder helped to prevent this from happening. More generally, participants recognised the role of fiscal equalisation transfers in preserving regional/linguistic identities (e.g. by maintaining the presence of Frenchspeaking people in Quebec), and thus contributing to political stability. Redistributing income from rich to poor jurisdictions also entails economic efficiency considerations. Some participants argued that spill-over effects associated with the provision of local public goods could warrant equalisation transfers. As an illustration, city centres often bear the significant costs of developing infrastructure networks (e.g. for public transport), which also benefit residents from suburban jurisdictions. For these jurisdictions, the “user-payer principle” is often difficult to apply, in particular when the personal income tax is based on the residence criteria, as is the case in Germany.4 To compensate city centres for this special burden, fiscal equalisation transfers in Germany partly account for the differentiated costs of providing public services according to the size of the municipalities. The underlying assumption is that the cost per inhabitant follows a U-shaped curve according to the population density. Some participants further noted that contractual solutions could be envisaged to internalise spill-over effects. As an example, cooperation agreements between cantons have been developed in Switzerland – e.g. for the financing of universities.5 These agreements make it possible to reap economies of scale and to internalise spillovers by concentrating the supply of these services in some cantons and by requiring other beneficiaries to pay for the benefits. Some participants suggested that strengthening equalisation transfers could provide economic benefits for the whole country. Inter-regional transfers could contribute to the avoidance of self-perpetuating regional disparities, especially if low income levels are due to a lack of infrastructure or education. Strengthening the demand and the growth potential of the poorer jurisdictions could bring indirect benefits for the richer ones through trade and investment linkages. In addition, fiscal equalisation transfers can smooth temporary regional revenue shortfalls. Acting as an insurance against asymmetric shocks, they would avoid temporary shocks being translated into long-lasting consequences. However, the issue of whether fiscal equalisation helps to boost the growth potential of receiving jurisdictions or conversely creates “poverty traps”, was a subject of debate. Some participants suggested that high outflow rates – such as those embodied in the German fiscal equalisation schemes – for any tax revenue generated by developing a jurisdiction’s tax base were discouraging these jurisdictions from introducing growth- promoting policies (see also Baretti et al., 2000). However, the new fiscal equalisation scheme across Länder to be introduced in 2005 is expected to improve economic incentives.6 The economic literature suggests that similar disincentive effects were at work in other countries, including Austria and Canada.7 Some studies have shown that equalisation transfers have had a negative incidence on economic growth in Australia. However, most participants argued that this “disincentive effect” was muted by the fact that the main objective of policy makers was to create employment. The fact that Canadian receiving provinces have grown faster than others over the past two decades, was presented as evidence against the “poverty trap” hypothesis. 5. Does fiscal equalisation promote fiscal competition? Fiscal equalisation levels the playing field between poor and rich jurisdictions and could therefore in principle be considered a pre-requisite for effective fiscal competition. Equalisation transfers reduce ex-ante fiscal disparities across regions (i.e. their ability to provide a given basket of public goods and services at a comparable level of taxation). Thus, differences in the quality of public services and in levels of taxation across jurisdictions should mainly reflect differences in local citizens’ preferences as well as their ability to reap efficiency gains. As an example, in the presence of complete fiscal equalisation, rich sub-national governments deriving substantial revenues from natural resources (e.g. in Canada or Norway) could, in principle, only lower their tax rates or improve the quality of their public services compared to other jurisdictions if they were able to improve the cost-efficiency of their public spending programmes. In the absence of equalisation transfers, rich provinces would dominate and competitive pressures would be lower. Fiscal equalisation could thus, in principle, allow poor jurisdictions to enter into fair and effective fiscal competition with the richer ones via the quest for greater efficiency. In practice, some features of fiscal equalisation schemes have reduced subnational governments’ incentives to improve public sector efficiency, thus limiting the gains of effective fiscal competition. In Switzerland, conditional grants received by the cantons – which often contain an “equalisation supplement” – largely reflect actual ex-post costs, instead of ex-ante standardised costs. In this context, high contribution rates from the confederation weaken the incentives for cantons to contain costs. The poorest ones are likely to be the most affected, since conditional transfers account for a larger share of their budget – more than 40% in some cantons, compared with a national average of about 17% in 1999. Fiscal equalisation based on actual, instead of potential, fiscal capacities could also result in “unhealthy” tax competition. German Länder cannot cut their tax rate or narrow their tax base as they have virtually no effective taxing powers. However, some participants argued that a lax enforcement of the tax rules could be considered as one form of tax competition. German Länder could, for instance, limit their effort to improve compliance with the federal tax laws, especially since they bear the cost of the tax collection and tax auditing processes, while they benefit very little from collecting more tax revenues. This would not only reduce the effective tax burden within their jurisdiction – and thus contribute to boosting economic activity – but would also bring down their contribution to the country’s fiscal equalisation since it is largely based on actual tax revenues. However, some participants underlined that there is no empirical evidence to prove this effect. Discretionary features of the grant system could further reduce the incentives for sub-national governments’ to cut taxes, thus preventing effective tax competition. Norway provides an example whereby all the municipalities have kept the marginal personal income tax rate at the maximum level set by the central government since 1979. This partly reflects the fear that any cut in local taxes would result in lower discretionary grants from the central government (OECD, 2002b). 6. Conclusions Reassigning spending and tax responsibilities should contribute to reducing the size of inter-governmental transfers. Many participants drew attention to a gradual erosion of effective spending responsibilities, with the central government – and the European Commission for EU area countries – more frequently dictating norms and standards. At the same time, revenue-raising powers of sub-national governments are still rather limited in many countries, and do not match their spending responsibilities. More effective application of the principle of “(s)he who decides, spends and pays”, by devolving more taxing powers to sub-national governments would promote their autonomy and accountability, and would be likely to promote public sector efficiency. However, participants did not believe that getting the alignment of functions “right” would suffice on its own in order to meet the objectives of fiscal equalisation. Fiscal equalisation has one key objective – ensuring minimum quality standards for key public services across the country. But the accepted degree of variation in the quality of public services between jurisdictions varies significantly from one country to another – and so does the degree of fiscal equalisation. Furthermore, fiscal equalisation schemes may have other objectives. Most participants acknowledged strong political, constitutional and social reasons for fiscal equalisation. In particular, many participants stressed its role in strengthening national and social cohesion, in particular in countries with strong cultural and/or ethnic fractionalisation. Popular support was also described to be high in some countries, as people appreciate being able to get high quality public services everywhere in the country. There was some discussion as to whether there is a strong “economic case” for fiscal equalisation: does it contribute or is it detrimental to economic efficiency? Participants mostly agreed with the idea that fiscal equalisation schemes act to slow migration flows within countries. While the vast majority considered it to be a good thing (reducing congestion and security costs), a few wondered whether this could hinder structural changes within the economy. Participants also expressed mixed views on the impact of fiscal equalisation on the growth potential of receiving jurisdictions. It is often argued that revenue equalisation schemes create adverse incentives: a region could prefer not to foster economic development (and hence its tax revenues) if this results in lower equalisation grants. However, many participants were sceptical about whether this force is really at work. They argued that sub-national governments often prefer to have more fiscal autonomy and/or that the impact on employment is a key policy objective. There was, however, wide consensus on the need to avoid fully compensating poor regions through equalisation grants. In order to preserve the sub-national governments’ incentives to improve the cost efficiency of their spending programmes and/or set competitive tax rates, fiscal equalisation schemes should also reflect “standardised” instead of effective, measures of fiscal capacities (i.e. the schemes should aim to equalise standard provision costs of public services, rather than effective costs, and/or potential rather than actual revenues). References Baretti, C., Huber, B. and Lichtblau, K., 2000. A Tax on Tax Revenue. The Incentive Effects of Equalising Transfers: Evidence for Germany. CESifo Working paper No.333, September 2000, ftp://188.8.131.52/CESifo_WP/333.pdf . Joumard, I. and Giorno, C., 2002. Enhancing the Effectiveness of Public Spending in Switzerland. Economics Department Working Paper No. 332, http://appli1.oecd.org/olis/2002doc.nsf/linkto/eco-wkp(2002)18/$FILE/JT00129276.PDF . OECD (2001a), Knowledge and Skills for Life – First results from PISA 2000. OECD (2001b), OECD Economic Surveys, Canada. OECD (2002a), OECD Economic Surveys, Germany, forthcoming. OECD (2002b), OECD Economic Surveys, Norway. Schneider, M., 2002. Local Fiscal Equalisation Based on Fiscal Capacity: The Case of Austria. Fiscal studies, 23 (1), 105-133. Smart, M., 1998. Taxation and Dead-weight Loss in a System of Intergovernmental Transfers. Canadian Journal of Economics, 31, 189-206. Wurzel, E., 1999. Towards More Efficient Government: Reforming Federal Fiscal Relations in Germany. Economics Department Working Paper No. 209. http://www.oecd.org/pdf/M00001000/M00001486.pdf 1 . The author works in the Economics Department of the OECD. The opinions expressed in the paper do not always correspond to the author’s views and do not engage the OECD or its member countries. 2 . The State Children Health Insurance Programme is a partnership between the federal and state governments that helps to provide children with the health coverage they need to grow up healthy and strong. Funds are allotted to each participating state according to their number of uninsured and low-income children, accounting for regional cost differences. This initiative set aside $40 billion over 10 years for states to provide new health coverage for millions of children – the largest children’s healthcare investment since the creation of Medicaid in 1965. Allocations for the fiscal year 2002 ranged from $363,825 for Northern Mariana Islands’ relatively small population to $528 million for California. For more information on this programme: http://www.dhhs.gov/topics/childhealth.html. 3 . The first stage of horizontal tax equalisation amounted to €8.7 billion in 2000. The second stage of horizontal tax equalisation amounted to €8.3 billion. The first stage of the vertical tax equalisation (unconditional grants from the central government to poor Länder) amounted to €3.7 billion. On top of this, the second stage of the vertical equalisation (need-based grants) amounted to €9.6 billion. In addition, co-financing grants with a redistributive element from the federal government to the Länder amounted to €23.6 billion in 2000. For more information on the German fiscal equalisation scheme, see Wurzel (1999). 4 . Horizontal apportioning of personal income tax revenues between states is based on the residence principle in Germany, i.e. revenues accrue to the taxpayers’ state of residence. 5 . The proposed reform of spending assignment and fiscal equalisation should give a further boost to such contractual arrangements (Joumard and Giorno, 2002). 6 . First, transfer rates between Länder with relatively strong and weak financial positions respectively, have been reduced. Second, a “premium” was introduced to reward states whose increase in tax receipts – measured against the year before – exceeds the average revenue increase of all states combined. Moreover, the average transfer rate of a donating state is now limited by a new cap that replaces the previous system of side conditions (Garantieklauseln). See OECD (2002a) for more details. 7 . In Canada, it has been argued (OECD, 2001b) that equalisation can adversely affect economic expansion in the have-not provinces, particularly the lowest-income Atlantic Provinces, and can therefore stand in the way of badly needed regional adjustments that would reduce horizontal imbalances. Negotiations on a major industrial development of mineral deposits in Newfoundland’s Voisey’s Bay were reported to have stalled in part because the province would have seen only a small fraction of the mineral royalties and business and personal income tax revenues from the project since equalisation payments would have declined almost as much as such revenues increased. Smart (1998) also argued that some provinces set tax rates at a high level to reduce their tax bases – the negative effects of higher tax rates on a province’s tax base being partly compensated by higher equalisation transfers from the federal government. On the Austrian case, see Schneider (2002).