Social Europe
One for All?

To what extent should national policies of social protection and labour market regulation be harmonized in a deepening and widening EU? This question, which lies at the heart of the debate about the Social Chapter of the Maastricht Treaty, is the focus of CEPR's eighth annual Monitoring European Integration report, entitled Social Europe: One For All? Compiled by Charles Bean (LSE and CEPR), Samuel Bentolila (CEMFI, Madrid, and CEPR), Giuseppe Bertola (University of Turin, European University Institute, Florence, and CEPR) and Juan Dolado (University Carlos III, Madrid, and CEPR), the report provides a detailed analysis of European labour markets, explores ideas for the Social Chapter, and sets out specific recommendations for the design and implementation of social policies within the EU.

The authors note that the 'pervasive' welfare states of Western Europe, and the frequent references to a 'social dimension' in EU documents and treaties, testify eloquently to the desire of both people and governments to ameliorate the undesirable social consequences of economic life. Measures to address these concerns – 'social policy' for short – can take a variety of forms. These range from workplace regulation, such as constraints on worker dismissals or health and safety regulations, to income transfers, in the form of pensions and unemployment and other social benefits. For the most part, social policy seeks either to remedy market failures – in which case it will have beneficial consequences for economic efficiency – or to protect disadvantaged members of society from the consequences of their economic weakness. In the latter case, the primary purpose is redistributional, and hence has a less desirable impact on efficiency. Thus the composition of social policies involves an inescapable trade-off between economic efficiency and social protection, i.e. between the average level of welfare and its distribution across society. But social policy can also have perverse consequences, because sometimes it can protect better-off groups within society. Moreoever, those who benefit from regulation will have a vested interest in retaining it and, hence, will inhibit the political feasibility of reform, even when the arguments for reform are compelling in other respects.

Of particular significance, from the EU perspective, is that there are interactions between social policy, broadly interpreted, and economic integration. The impact of economic integration – which also encompasses a variety of forms, ranging from trade liberalization to enhanced labour and capital mobility – on social policy can be summarized thus: closer integration is generally desirable on overall efficiency grounds, because it offers mutually beneficial gains from trade; but integration usually also has adverse consequences for relatively inefficient producers. As a general rule, closer integration between a richer and a poorer country can be expected to increase inequality in the richer country (by adversely affecting the incomes of its poorer citizens in relative and, possibly, absolute terms). By the same token, inequality in the poorer country is likely to be reduced. Thus it is unsurprising that, as the EU has expanded over time, with the richer original core of countries being augmented by poorer peripheral countries, the demand for greater social protection in the richer countries has grown. This is confirmed by wider empirical evidence showing that social spending is not only higher in rich countries – which can better afford the luxury of such protection – but also in countries that are more open to international influences.

But integration also affects the supply of social protection, in that redistribution can become more difficult to accomplish when labour and capital are both more mobile. In these circumstances, in order for national social policies to be effective, more coordination at EU level may be necessary. Otherwise, governments could be tempted to use social policies strategically to benefit their own citizens at the expense of foreigners. For example, a government may seek to encourage greater inflows of capital, and so drive up domestic wages, by offering lower taxes or less regulation and lower social protection than in other EU countries. With a given and limited total supply of capital, the effect of this – other things equal – would be to reduce the tax base in the other countries, thereby pushing up the tax burden on their immobile factors, or forcing a reduction in their public spending (or both). Such 'social dumping', however, eventually will generate retaliatory or defensive responses from the other governments which, if unchecked, will result in lower levels of social protection all round as the countries engage in a 'race to the bottom'.

The authors do note that, if existing national social policies are ill-designed or fail to protect the most disadvantaged members of society, a 'race to the bottom' outcome may not necessarily be a bad thing. In general, however, it is clear that lack of coordination can lead to suboptimal levels of intervention or regulation. A canonical example of the importance of international coordination of social policy is afforded by the temptation for shipowners to escape the imposition of safety regulations or work rules by registering their vessels in more 'lenient' jurisdictions. As integration in the EU becomes deeper, with the consolidation of the single market and the introduction of EMU, and as the Union expands to embrace some of the Central and East European countries (CEECs), with their significantly lower wage costs and very different institutional structures, the question of coordination of social policy issues will come increasingly to the fore.

Against this background, and drawing on the lessons of past EU integration experiences – notably the enlargement to include the poorer Mediterranean countries and Ireland – the report discusses the implementation of present and prospective national and EU-level social policies. Many of the policy issues raised in the debate surrounding the Social Chapter are addressed. Apart from the general question of the extent to which EU member states can continue to implement their own national social policies, rather than cede responsibility to the EU level, the authors enquire (1) whether 'social dumping' would be inevitable in the absence of a common EU social policy; (2) whether common provisions, if needed, should take the form of agreements about minimum standards or should allow exceptions for individual countries; (3) how the circumstances of potential new EU members from among the CEECs should be taken into account in the design of current EU social policy directives; and (4) to what extent social dumping should be welcomed as a healthy force obliging countries to lighten the sometimes excessive and damaging regulations they impose on their labour markets.

After reviewing the history of EU social policy in relation to the major movements forward in economic integration over more than four decades, the report infers the existence of a causal relationship between closer integration and new attempts to harmonize social policies. Two other trends that are also seen as potentially relevant to development of a pan-European social policy are the increasing heterogeneity in income levels as integration has proceeded, and the increasing unemployment levels in most member states. A quite striking finding, however, is that even though the social-security systems of individual countries have become more highly developed and labour-market regulations much more restrictive, there has not been an associated closing of the gap between national and EU-level social policies. Consequently, progress towards development of EU social policy has been much slower than in trade or monetary policy. This suggests that, to date, EU policy has not seriously constrained national social policies.

The report considers, however, that this situation is likely to change as both deepening and widening of the EU proceed. Thus EU-level social policies can be expected to become more binding in future. The authors conclude that continued integration will accentuate the pressure both for 'flexibility-oriented' reform and for greater harmonization of social policies. From the point of view of reducing the persistently high levels of unemployment in most EU countries, they argue that increased labour-market flexibility would be a welcome development. The 'European model' has involved the maintenance of both a high level of labour-market rigidity and a high level of social protection. In an environment of increasing competition, this model will become ever more costly in terms of unemployment and it risks creating serious sustainability problems for the welfare state. A drive for harmonization, therefore, should not be permitted to stand in the way of reform.

The report further contends that harmonization only makes sense between countries at similar standards of development, and with similar social preferences with regard to the trade-off between efficiency and redistribution towards the poor. Income disparities are already large among the current EU members and will increase dramatically with the next enlargement. There are also large disparities in labour-market regulations and in the organization of social policies. Since these policy differences are present even among countries with similar income levels, they constitute compelling reasons for different national preferences concerning the efficiency versus redistribution trade-off to be respected.

Consequently, the authors argue that policy-makers should strive for minimum standards, but only where these are acceptable to all countries. Examples of desirable and undesirable measures are not hard to find. Most obviously, establishing a single minimum wage, or a single unemployment-benefit level, throughout the EU would not make sense, since any average level would be too low for the richest countries and too high for the poorest ones, and would carry the potential for causing even higher unemployment in the latter. In contrast, fostering a dialogue between the representatives of employers and employees at the European level would be a good idea, because it would help ensure that EU social policy was based on consensus. It would also render more likely the rejection of measures that would harm the ability of firms to compete in increasingly integrated markets. It should be kept in mind, however, that European-level federations of unions, like their national counterparts, will generally represent existing employees rather than the unemployed. Similarly, European-level federations of employers are more likely to represent the interests of large corporations than of small and medium-sized firms. In effect, therefore, the unemployed and owners of small firms will have to rely on politicians to represent them at the European level.

Also desirable would be measures to foster labour mobility, for example, action to suppress any discrimination against migrant workers, or erection of any unwarranted formal or informal barriers to the mutual recognition of diplomas. By the same token, measures that clearly deter mobility for protectionist reasons – such as the 1995 'Posted Workers Directive' – should be avoided. Mobility-enhancing measures would help reduce incentives for 'benefit shopping'.

Measures designed to overcome problems of imperfect or asymmetric information are also potentially desirable, for example, the obligation to furnish workers' representatives with notice of and information about collective redundancies, or to foster exchanges of information among employees of multinationals in different countries. But if harmonization of social policies were to evolve into measures obstructing the shift of production sites across countries, this would harm the competitiveness of firms and would be counterproductive. Problems of asymmetric information may also be overcome by regulations on health and safety, but imposing the same working conditions across countries in a way that runs counter to different national tastes and customs would be undesirable.

The report thus concludes that although EU-level policies are not particularly binding at present, they may become a more significant factor in future. At some point, as the competitive concerns raised by monetary union and the single market make reform more desirable and politically feasible – with the consequence of improving labour-market flexibility and reducing social protection – the Social Chapter may become a binding floor, thereby preventing an unwarranted 'race to the bottom'.

Social Europe; One for All
C Bean, S Bentolila, G Bertola and J Dolado

ISBN Paperback: 1 898128 33 2