Cohesion and Community public finance

Horst Reichenbach & Marc Vanheukelen*


No union can exist without solidarity. In the European Union the pursuit of solidarity between Member States and regions is guided by Title XIV of the EC Treaty on "economic and social cohesion". The objectives are to promote the overalll harmonious development of the Community and to reduce disparities between the levels of development of the various regions and the backwardness of the less-favoured regions, including rural areas.

The main grant instruments for this purpose are the structural funds, consisting of the European Regional Development Fund, the European Agricultural Guidance Fund and the European Social Fund. A fourth structural fund was created in 1993, with an envelope of 419 mecus in 1994, and thus very small in relative terms, the Financial Instrument for Fisheries Guidance seeks to expedite the adjustment process in the fisheries' sector. The structural funds were supplemented in 1993 by the Cohesion Fund, as foreseen in the Maastricht Treaty, in favour of countries with a per capita income level below 90% of the EU average.

The structural Funds as well as the Cohesion Fund are financed out of the Union's general budget. On the other hand there are the Union's lending instruments for cohesion, primarily the EIB and the ECSC loans, which are funded by way of EIB or Community borrowing on international capital markets.

The growth of these various financial instruments is illustrated in Table 1. Until the first enlargement, in 1973, the volume of grants and loans was very modest. Since then the financial means for structural actions have expanded steadily, in particular upon the 1986 accession of Spain and Portugal. Between 1988 and 1992 the structural funds' envelope was doubled in real terms. Following the Edinburgh agreement of December 1992, the budget for structural actions is to be the object of another considerable rise until the end of this decade (see Table 2), especially with respect to the four poorest Member States, which are to receive in 1999 twice as much in constant prices as in 1992.

Table 1
The evolution of Cohesion grants and loans 1961 - 1994
(Millions of current ECU, payment appropriations)
ESF (1)
Guidance (2)
Fund (4)
EIB loans
ECSC loans
1961 8,6 - - - 0,6 21 23
1968 24,5 34,0 - - 58,5 213 108
1975 148,6 76,7 150,0 - 375,3 814 731
1985 1407,4 685,5 1610,0 - 3702,9 5699 1265
1988 2298,6 1140,9 2979,8 - 6419,3 7666 880
1990 3212,0 1825,3 4554,1 - 9591,4 10966 1086
1992 4303,7 2847,4 8553,8 - 18298,3 12974 1474
1993 5395,6 3026,7 2548,4 795,0 20663,0 16779,4 918,3
1994 (5) 5819,0 3301,0 7702,0 1679,0 21528,8 n.a. n.a.
(1) operational since 1961
(3) operational since 1975
(5) as forecast in 1994 budget                
(2) operational since 1968
(4) operational since 1993

Table 2
The evolution until 1999 of cohesion grants according to the 'financial perspectives', as modified following enlargement to 15 Member States
(million of ECUs in 1995 prices; commitment appropriations)
Structural Funds Cohesion Fund EES Financial Mechanism (1) Total Structural Actions

24 069

2 152


26 329


25 206

2 396


27 710


26 604

2 663


29 375


28 340

2 716


31 164


30 187

2 769


32 955

(1) Regional transfers previously made by Austria, Sweden and Finland in the framework of the European Economic Space Accord, which are now taken over by the EU budget. Figures are expressed in current prices.

In the next part a brief survey will be given on how these developments have come about Part 3 will provide some guidance for cohesion policy from the fiscal federalism literature. In Part 4 cohesion expenditure is analysed with respect to its redistribution and allocative role. The paper concludes with a brief outlook on the possible consequences of EMU and enlargement.


The harmonious development and the diminution of regional disparities have ever since the creation of the Community been an important element in the collaboration of the Member States. To this end, the Treaty of Rome foresaw two chief instruments: The European Investment Bank and the European Social Fund. The former's task was, and still is, to contribute to a steady development of the internal market and especially to facilitate the economic expansion of the lagging Communities regions (art. 198 e EEC). The objective of the Social Fund is to promote the employment opportunities for workers through vocational training as well as to further their spatial and occupationial mobility (art. 123 EEC). The Guidance sector of the European Agricultural Fund came into operation in 1968 in order to help farmers modernize agriculture and, in a later development of its activities, to bolster the economic viability of rural areas.

In the wake of the first enlargement, the European Regional Development Fund was set up in 1975, providing infrastructure and later on also investment aid to lagging regions. Its relative importance among the structural funds has grown over time, accounting for about 60% of total structural outlays in the period 1989-1993.

The integration process underwent a profound deepening and widening in 1985 and 1986 with the launching of the completion of the internal market programme and the third enlargement. Upon the accession of Spain and Portugal, the share of the European population living in backward areas went up by 30%, the number of unemployed by 25%, and the number of people active in agriculture by 35%.

As to the effect of the completion of the internal market, it was generally accepted that there was a risk that regional disparities might be aggravated, even though real income of the Community as a whole would receive a considerable boost. The status of the redistribution question was therefore enhanced in the 1987 Single European Act which inserted a specific Title in the Treaty on 'Economic and Social Cohesion' laying down in articles 130 A to 130 E the objectives and instruments of cohesion policy and announcing a comprehensive reform of the structural funds.

The budgetary consequences of the explicit adoption of the cohesion objective at the 'constitutional' level emerged soon afterwards with the so-called 'Delors I' package in 1988 on the EC budget to double the size of the structural funds in real terms, reaching in 1992 almost 30% of the total budget and about 0,3% of EC GDP.

The cohesion objective was given renewed emphasis in the Maastricht Treaty. Pursuant to the latter's cohesion protocol, the cohesion fund came on stream in 1993. In line with this emphasis, the 'Delors II' package agreed at Edinburgh in December 1992 regarding the future of EU public finance up to 2000 again raised expenditure for structural operations very considerably, as indicated in Table 2, climbing to 0,46% of EU GDP.

The Edinburgh agreement maintained the privileged treatment of objective 1 regions and in particular of the four poorest Member States, whose real transfer receipts from structural actions will have risen almost fourfold in real terms between 1988 and 1999. In sum, the overriding characteristic of the development of the EC budget since the end of the eighties has been the major reinforcement of the redistributive dimension in pursuit of cohesion.


What light does economics shed on the role of a supranational layer of government in redistribution? The fiscal federalism literature, dealing with the division of economic and budgetary powers between different levels of government, points out that if it is deemed necessary that a certain level of homogeneity or fairness across the Union be reached, transfers need to be decided on and organized at the 'highest' level of government. Another possible rationale is when individuals display a high-level of cross-border mobility, such that relatively poor individuals would migrate to countries with a generous social security system whilst the well-off would move out. This would create serious externality problems calling for internalising measures, possibly at 'federal' level. Given the limited degree of cross-border migration within the Union in the foreseeable future, this second motivation for a European redistributive function is hardly relevant.

However, theory has little to offer on the question how high this aforementioned level of homogeneity should be, since the latter is a function of political preferences, reflecting the strength of the feeling of togetherness.

Empirical evidence on mature federations in the developed world does not present a clear picture either. Federations function with widely-ranging degrees of expenditure and revenue centralisation, also in the redistribution field. Much depends on the political process. The single stylized fact that stands out in this regard is that there exists an inverse relationship between a constituent state's public finance autonomy and redistribution between the states of the federation. Put differently, the more states' ability to spend, tax and borrow is restricted, the more interregional income disparities are equalized by way of federal taxes and transfers. These federal public finance flows close the interregional income gap by 30 to 50%.

As far as expenditure assignment on equity grounds is concerned, a distinction needs to be made between interpersonal redistribution on the one hand and interrregional transfers on the other.

- The conventional conclusion from fiscal federalism that interpersonal redistribution should be carried out at the highest level of government is based on the assumption of near-perfect labour mobility, giving rise to the problems we outlined earlier. As this assumption does not apply to the Community, neither the 1977 MacDougall Report nor any subsequent work on the subject has seriously advocated that interpersonal redistribution should primarily be a Community responsibility; on the contrary, recent literature has further challenged the traditional view, pointing to the importance of geographical proximity for such redistribution.

Nevertheless, suggestions for a more limited supranational involvement have been advanced, such as the idea of a Community unemployment fund. This originated from the 1975 Marjolin Report, was revived in the MacDougall Report and resurfaced in the context of stabilisation policy under economic and monetary union. Politically, this idea has never been taken seriously, and we think rightly so, because the case for centralisation has not been convincingly made.

- Inter-regional transfers fall into two broad categories: general-purpose equalising grants and specific-purpose grants. Whether the Community will wish to embark on unconditional block grants for fiscal equalisation purposes largely depends on the feelings of solidarity between member states and their mutual confidence that funds are being wisely spent. The time clearly does not seem right now and the early stages of economic and monetary union will add very little. In our view, further steps toward political union will need to be taken before this matter could become the subject of serious negotiation between the member states.

Specific-purpose grants serve two goals at once. On the one hand they give rise to interregional transfers. In existing federations, they represent on average about 2.5% of national GDP and reduce regional income differentials by about 6%. On the other hand, they serve allocative purposes influencing the provision of public services at lower levels of government in a wide spectrum of fields covering welfare, health,education, transport, vocational training, research and regional policy. They are thus complex policy instruments on which an extensive theoretical and empirical literature exists. Community cohesion expenditure falls into this category of specific purpose grants.


Displaying a redistributive as well as an allocative dimension, specific-purpose grants will, for some time to come, continue to be the most appropriate instrument for effecting interrregional transfers at Community level in pursuit of economic and social cohesion. As always, however, when one instrument has to contribute to two different objectives, it cannot be expected to be optimal with respect to both objectives taken separately. Part of the criticism of the Community's cohesion expenditure is due to this unavoidable conflict.

4.1. Redistributive aspects

The redistributive role of cohesion expenditure is determined by its level and distribution between member states and regions. For structural operations to exhibit a positive redistributive effect, resources must flow, in net terms, from richer regions of the Community to poorer ones. The extent of regional transfers is a matter of political choice, but this choice is not made in a vacuum or arbitrarily. It will be guided by a number of qualitative and quantitative factors, such as objective indicators of regional disparities; an appreciation of the distribution of the overall costs and benefits of integration; the potential scale and effects of integration; the degree of homogeneity in terms of citizenship, culture and language; and the economic effectiveness of the transfers.

- As for the level of cohesion expenditure, we do not believe that there is an objective method that can be considered reliable from an economic point of view or acceptable from a political point of view. Nevertheless, there are three main methods that have been used and will continue to be used in order to provide some guidance for the necessary political decision on the level of transfers. The first consists of mechanical calculations of the number of people who need training, the kilometers of roads to be built, etc. The second, the investment gap method, draws on a very simple growth model and considers how much additional investment is required to generate a desired level of extra growth. The third method relies on extrapolations from past experiences be they the Marshall fund, previous EC structural fund efforts, or intergovernmental transfers in existing federations.

In the run-up to the 'Delors-II' package the Commission made an attempt to quantify "catch-up" needs, in particular the additional investment requirements in Objective 1 regions with respect to different categories of infrastructure. This was certainly a useful exercise for a rough quantification of these regions'absorption capacities. Beyond this, however, the definition of needs itself is such a controversial subject that it could not facilitate the political negotiations on the scale of the structural operations.

The Marshall Plan for the reconstruction of post-war Europe is often mentioned as an example of a timely and generous programme of aid. During 1948-51 the USA granted about 1% of its GNP to Europe every year; this represented on average about 2% of the recipient countries' annual GNP (with considerable differences between countries; see Eichengreen and Uzan). Compared to this 1%, the Community effort of 0.46% in 1999 look modest, but the latter should be seen, of course, as a much longer-term commitment.

Seen from the angle of recipient relatively less developed member states, the situation is very different for the three smaller ones (Ireland, Portugal and Greece) on the one hand and Spain on the other. While the smaller states can be expected to receive very substantial amounts over 1986-1999 (at an annual average of 3.3% of their GDP) the sums for Spain are more modest (0.9% annually).

In this regard, it should be borne in mind that because of differences in initial disparities vis-à-vis the Community average, equivalent transfers as a ratio of local GDP produce a different redistributive effect, i. e. they close the relative income gap to a different degree. The larger the initial gap, the larger needs to be the GDP percentage transfer to achieve a given redistributive effect. The structural interventions as agreed at Edinburgh should provide for a redistributive effect of about 5% for Ireland, Greece and Portugal and about 1,5% for Spain.

- The distribution of funds between Member States is another sensitive political issue. It has important implications for the redistributive effectiveness of the structural operations as well as for their allocative role. With respect to interregional redistribution, a net fund - that is, one which would be exclusively concentrated on the less-prosperous Member States - would of course be the most effective, and during the successive reforms of the Structural Funds this idea has often been discussed. The Cohesion Fund is an example of this idea. It demonstrates how much redistributive considerations have been at the forefront of its creation.

The 1988 distribution of the Structural Funds between member states and regions has been analysed by Gordon (1991). He found that transfers were carefully targeted but objected that otherwise-identical regions were treated differently depending on the country to which they belong. However, this criticism fails to take account of the mixed Community responsibility vis-à-vis Member States on the one hand and regions on the other.

This means that Member States with the same level of average GDP should be treated differently if their internal regional disparities are significant. Similarly, the degree of Community involvement in a specific region must depend on the financial capacity of the country in which the region is situated, since under the principle of subsidiarity the primary responsibility for regional assistance lies with member states as far as they are able to cope with their regional problems.

4.2. Allocative aspects of structural operations

The reduction of regional disparities can be attained only if Community assistance improves growth performance in the recipient economies. A necessary condition for this is that cohesion expenditure should cover those activities which are the most promising for raising the growth potential of the poorer regions. This can only be achieved by measures affecting the supply side, i. e., through improvements in the factor endowments of the economy, or by making existing endowments more productive, or both. The bulk of Community structural operations correspond well to this requirement: the main areas of intervention are infrastructure and vocational training.

Within broadly-defined areas of intervention, flexibility is necessary to identify the specific measures which are likely to be the most effective for the development of the regions in question. This is the core task of programming within a multi-annual planning framework, the present one running from 1994 to 1999, as well as of monitoring, in order to control the implementation of programmes and to modify plans in the light of changing needs or unforeseen execution problems.

As a counterpart to this possibility to tailor structural interventions to local circumstances and given the pluriannual framework which heigthens the importance of correct strategic choices, the revision of the structural fund regulations of 1993 has stressed the need for a rigorous and systematic ex ante appraisal and ex post evaluation of interventions. By forcing local, regional and national authorities to properly articulate intermediate and final objectives and the way in which programmes are supposed to achieve them, as well as by establishing performance indicators to this effect, evaluation will serve to enhance the cost-effectiveness of structural transfers. This will not only benefit the real convergence efforts of recipient regions; by documenting in a transparent fashion the 'value for money' it produces countries which are net contributors to the EC budget will find cohesion expenditure more acceptable.

Community grants can have a positive allocative impact only if they are not offset by reductions in national spending; they must be additional to national expenditure, not replace it. This might seem straight-forward enough, but analytically and empirically the issue is complex, as the grant literature and practical Community experience testify.

In future the interplay between Community structural operations and national budgetary expenditure, whether at central, regional or local level, will become even more important since national budgetary policies are crucial to the nominal convergence process designed to lead to the final stage of EMU. In a number of member states considerable efforts will have to be undertaken, in particular with regard to budgetary discipline as part of their convergence programmes. Compatibility between this discipline and development efforts will henceforth have to be achieved in a new and more demanding context.

This raises the more general issue of the relationship between real convergence (cohesion) and nominal convergence, the prerequisite to participating in the final stage of EMU. More specifically, to qualify for monetary union, Member States need to have a low level of inflation and reduce their general government deficit to the 3% of GDP benchmark of Maastricht. It is sometimes argued that cohesion will be damaged by the pursuit of the Maastricht criteria on account of the alleged deflationary bias of the convergence process. Indeed, the first round effects of deficit cutting and anti-inflationary processus will be to curb aggregate demand. However, other factors such as the reinforced credibility of government policy may trigger changes that more than compensate the growth depressing effects of tight fiscal and monetary policy. 'Expansionary contraction' has occurred in the past, as exemplified by the Irish experience at the end of the eighties. Moreover it is clear that in the long run a precondition for sustained growth is a stable macro-economic framework, of which low inflation and public borrowing requirements are essential components. It is this positive view on the influence of nominal on real convergence that has prompted the macro-economic conditionality provision in the cohesion fund regulation (art. 6). In accordance with the Maastricht protocol on social and economic cohesion it stipulates that cohesion fund transfers will cease if the Member State in question is found by the Council to have an excessive deficit and has taken insufficient measures to eliminate the latter within a specified period.

Structural fund transfers are not subject to any economic conditionality beyond the just mentioned additionality proviso. In broad analogy with the cohesion fund, the question might be asked whether the Funds should not become more performance-related then expenditure-related. For instance, one could envisage, with a view to further boosting the effectiveness of structural transfers, the inclusion in Community Support Frameworks of commitments on the side of the recipient country to concrete regulatory initiatives seeking to improve the supply side performance of the economy such as through measures promoting greater goods and labour market flexibility or tax reforms, in keeping with the White Paper on growth and competitiveness,. Member States delivering on their commitments could be rewarded by supplementary Structural Fund assistance from a tranche that would not be pre-allocated by country.


The course of cohesion policy over the next ten years will be affected by two major prospective events in the integration process: EMU and enlargement to the East.

What influence is the establishment of EMU going to exert on the regional income gap? A precise assessment of net national, let alone regional, benefits is extremely difficult, primarily because chiefly micro-economic gains need to be weighed against the loss of a macro-economic adjustment instrument, the national currency.

However, economic analysis suggests that, whilst risks may be highest for the relatively less developed Member States, overall they stand to gain more than average from EMU (Commission, (1990)). The abandonment of the devaluation instrument is very likely to be more than compensated by a number of advantages, inter alia, the elimination of transaction costs and exchange rate uncertainty, and the reduction of interest rates. Yet, as such predictions are surrounded by a fair amount of uncertainty and anyway apply in general terms only, EMU calls for vigilance on the side of those in charge of cohesion policy.

The enlargement of the Union to the new democracies of Eastern Europe will pose a great challenge to redistributive policy, since the share of the European population living in poor regions will increase very considerably. Four example, the four Visegrad countries, probably constituting the first group of new entrants after the turn of the century, currently have a joint population size similar to that of the four poorest Member States together and a prosperity level well below them.

It is quite impossible to foretell with any degree of accuracy what will be the Eastern countries' income level and economic structure at the time of accession, how Community policies will have evolved by then, and how the acquis communautaire, including cohesion, will be phased in. Therefore, too mechanistic conclusions on the budgetary implication for cohesion policy of enlargement to the East must be avoided. Still, the expression of solidarity between Member States will have to be adjusted to reflect the new reality, which risks being clearly more heterogeneous than that of today.

The achievements, problems and prospects of cohesion policy, which, it should be recalled, does not merely have a budgetary dimension, but should also be taken into account in the formulation and conduct of economic policy at community and national level, will be the object of a major review in the first Cohesion Report, to be presented to the Commission this year in accordance with article 130 b as amended by the Maastricht Treaty. No doubt this report will offer much food for thought on the way cohesion policy should evolve and how this should translate in budgetary terms.



(*) The authors are respectively director and acting head of unit at the Directorate General 'Budgets' of the European Commission. Rue de la Loi, 200 - 1049 Brussels. The usual disclaimer applies.