Although there has been much discussion concerning the future of the welfare state, researchers have only very recently considered broader aspects of the `welfare society'. A CEPR joint conference with the Institut für Höhere Studien on `Social Protection and the Enterprise in Transitional Economies', held in Vienna on 25/26 March, focused on the role of enterprise welfare provision in the particular circumstances facing the transforming economies of Central and Eastern Europe. The conference addressed the theoretical underpinnings of enterprises' social role, the empirical importance and composition of such social benefits, the possible marketization of their traditional social functions as a result of privatization, and the effects of this social role in helping or hindering economic transition. It was organized by Martin Rein, Professor of Urban Studies and Planning at the Massachusetts Institute of Technology, and Andreas Wörgötter, Professor of Economics at the Institut für Höhere Studien and Research Fellow in CEPR's Human Resources programme, Generous support from the Institut für Höhere Studien and from Directorate-General II of the Commission of the European Communities (Economic and Social Affairs) is gratefully acknowledged.

In the first paper, `Provision of Social Benefits in State-Owned, Privatized, and Private Firms in Poland', written with Saul Estrin, Mark Schaffer (LSE) reported preliminary results from their 1993 survey of 200 Polish manufacturing firms of four different ownership types (emerging private firms, privatized state firms, commercialized state-owned firms and unincorporated state-owned firms). Though their analysis of these data was still at an early stage, some patterns in levels and changes in enterprise social benefits were apparent; in particular, large and state-owned enterprises appeared to offer the greatest benefits but had also seen the greatest shedding of benefits (and workers) over the previous two years. Emerging private firms offered fewer social services and had very low levels of unionization, which may also be a function of their small size. These were the only enterprises to experience consistently strong employment growth, but establishing consistent links between the provision of benefits and either performance or the extent of unionization was more difficult for the other enterprise types.

Gérard Roland (European Centre for Advanced Research in Economics (ECARE), Université Libre de Bruxelles, and CEPR) argued that econometric analysis and improved measurement of the dependent variable were needed to draw strong inferences from this study. Jan Svejnar (Center for Economic Research and Graduate Education (CERGE), Prague) noted that the provision of benefits varied substantially within these groups, in particular among the private firms, while changes in their provision of such benefits may also indicate contracting out by workers rather than firms' phasing out of such services.

In `Social Protection and its Implications for Enterprise Restructuring', Paul Hare (Heriot-Watt University, Edinburgh) examined the effects of enterprises' obligations to supply social benefits on their ability to restructure and also the broader welfare consequences of their rapid restructuring, focusing on the coal industry in Russia and industry in general in Hungary. Russian firms have traditionally been more involved in providing social protection and are today much less competitive than their Western counterparts, so restructuring requires drastic reductions in the numbers of workers employed and/or the level of benefits provided and a potentially catastrophic collapse of employment and public services. A formal model and evidence from the Russian coal region of Vorkuta suggested that there is a trade-off between the urgency of socializing benefit provision (to encourage firm restructuring) and the risk that over-rapid restructuring will strain public finances and social services, especially in single-firm or single-industry regions. A generous system of state benefits in Hungary had produced greater short-run unemployment but also created a better long-run competitive position.

Georg Winckler (Universität Wien) suggested several improvements to highlight the incentive effects of firms' budget constraints, incomplete markets and existing social security institutions. Douglas Lippoldt (OECD) noted a conflict in Hare's analysis between the apparent burden of social benefits on firms' restructuring and the potential need at a macroeconomic level to finance their losses to avert the consequences of losing these services. Mitchell Orenstein (Institute for East-West Studies, Prague) suggested that restructuring could be achieved through wage-productivity deals rather than mass lay-offs.

In `Providing Social Benefits in Russia: Redefining the Roles of Firms and Government', with Simon Commander, Richard Jackman (LSE) examined the division of social functions between firms and local governments, looking particularly at enterprises in the Moscow area. Where firms bear a disproportionate burden of the provision of social services, workers are less mobile and firms less likely to employ `high-risk' workers, such as women of child-bearing age; moreover, in `company towns', economic shocks may threaten employment and public services. Market pressures have led firms to try to divest many social functions that are not directly linked with production and offer no specific tax advantages (such as subsidized clothing and catering). Local authorities' fiscal weakness and their difficulties in disposing of social assets suggest that Russian enterprises will nevertheless continue to play a central role in such services' provision, even when this would be better undertaken by local governments. In particular, the combination of inflation and rent control implies that workers' housing may come to represent a major liability for firms.
While agreeing that local authorities should take on more social services and that the federal government should guarantee them a minimum revenue base, Hans-Georg Heinrich (Universität Wien) identified a number of barriers to this rationalization in Russia. These include regional disparities, an inadequate and arbitrary system of collecting revenues, the role of semi-legal autonomous funds at all levels of government and the strain placed on local government budgets by price subsidies and the current influx of ethnic Russians from elsewhere in the former Soviet Union. All point to gradual transition at best and the need to maintain current services to avoid social catastrophe. Andreas Wörgötter emphasized that only a clear constitutional settlement in the current Russian Federation could settle the division of social responsibilities between enterprises and the various levels of government.

In `Employee Benefits and Labour Market Behaviour in the Transition Conceptual Issues and Evidence from the Czech Republic and Romania', John Earle (Central European University Foundation, Prague) explored the trade-off between wages and benefits and its effects on the prevalence of work-sharing versus lay-offs in enterprise restructuring. Workers might prefer in-kind benefits to cash wages on account of tax advantages and the convenience of workplace facilities, or if these take the form of `shortage goods' in an underdeveloped market. Firms will prefer to provide benefits if this reduces the overall wage bill (as a result of tax advantages or scale economies) or improves productivity (through reduced shirking and increased turnover) or if sunk costs are embodied in social assets in which the secondary market is limited. Recent data on Romania and the Czech Republic offer little support for this presumed trade-off between wages and benefits, however, since they are positively correlated across sectors; nor does a simple regression analysis support the view that high social costs per employee encourage firms to shed workers. In any case, such benefits amount to less than 5% of total compensation, which suggests that explanations of labour market behaviour and enterprise performance should be sought elsewhere.

Peter Rosner (Universität Wien) questioned the value of Earle's empirical results since recent data on transforming economies is unlikely to reflect an equilibrium outcome. Viktor Steiner (Zentrum für Europäische Wirtschaftsforschung, Mannheim) and Jan Svejnar added that other determinants of labour demand, in particular variable profitability across sectors, may influence both wages and benefits, which should therefore not be treated as exogenous. Douglas Lippoldt suggested that intangible benefits such as job security may play a key role in shaping workers' incentives.

In their joint paper, `Social Protection in the Large Enterprise: The Case of Slovakia', Andreas Wörgötter and Jan Planovsky (CERGE, Prague) examined the social role of three large Slovak firms (state-owned, joint-stock and privatized). Contrary to a priori expectations that privatization would lead to a substantial reduction in firms' provision of social benefits, the results of a detailed questionnaire designed to assess labour costs according to Eurostat's criteria indicated that levels of social provision were in fact comparable across all three ownership structures. While economic performance was a significant factor and the social role of the firm was generally declining, the slow pace of this shift and continued high compensation in the state sector was hampering Slovak firms' restructuring and their competition for workers. The inertia exhibited in the social role of the enterprise strengthened the argument for decisive transition policies to redistribute social responsibilities and promote restructuring. Even in Austria, the social contribution of nationalized industries had proved far too costly to enable them to operate efficiently over the long run.

Jan Mladek (CERGE, Prague) suggested comparing these findings with the experience of the Czech Republic, which had encountered adverse selection problems in the state/firm division of social welfare responsibility (as enterprises pushed high-risk individuals into public insurance schemes) while the strength of trade unionism had determined firms' ability to divest their social functions. Mark Schaffer cited Polish experience to question whether new private firms faced difficulties in competing for workers.

In `Fringe Benefits in Transition in Hungary', Gaspar Fajth (UNICEF) and Judit Lakatos (Central Statistical Office, Hungary) presented the results of two recent extensive surveys of labour costs and enterprise benefits. Although firm-level developments were quite complex, their unique data sets showed conflicting patterns in the transformation of the formerly extensive social functions of enterprises: while firms were scaling back many services (especially child care, cultural and recreational facilities), the main effects of tax considerations and new compulsory benefits such as severance and sick pay were felt in a shift of the benefit mix. Fajth and Lakatos questioned more broadly whether these discretionary non-wage labour costs should in fact be considered as social protection, since many simply reflected fringe benefits for managers or the rerouting of wages for tax purposes.

Fabrizio Coricelli (Università degli Studi di Siena) questioned the general case for devolving benefits to the enterprise level, noting that the combination of high mandatory social contributions and generous unemployment benefits in Hungary increased unemployment. Viktor Steiner argued that a variety of efficiency concerns might lead firms and workers to prefer social benefits over cash compensation. Robert Holzmann (Universität Saarland) reported evidence that making the provision of benefits such as sick pay mandatory at the firm level in Hungary had been a response to incentives for workers and employers to abuse the state-provided system.
In `Enterprise Social Benefits and the Economic Transition in Hungary', Martin Rein and Barry Friedman (Brandeis University) argued that Hungary's economic transition had led not to the collapse of the social role of the enterprise but rather to its recasting and even expansion. Statistical and survey data on the changing scale and nature of benefits packages among Hungarian firms suggested that non-wage compensation had remained quite significant, even where its composition had changed. Detailed analysis of the components of labour costs indicated that there was a continuum rather than a clear-cut division between cash and in-kind benefits. They concluded that tax advantages and the mandating of benefits had ensured that Hungarian firms will continue to play a social role, even though its contribution to economic efficiency was unclear; their future work would hinge on more precise definitions of benefits, subsidies and social protection itself.

Robert Holzmann questioned whether strong inferences could be drawn from this study given the statistical anomalies, the absence of new private firms from the sample and the lack of a market equilibrium as an implicit bench-mark. He stressed the importance of the transition from state to market incentives, where new private enterprises without sunk costs in infrastructure might offer very few social services. János Gács (International Institute for Applied Systems Analysis, Laxenburg) and Fabrizio Coricelli added that firms did not decide their social role in a vacuum: trade unions in particular could be influential in determining benefits.

In his analysis of `Early Exit from the Labor Force: the Case of East Germany', Martin Kohli (Freie Universität Berlin) examined the extent to which enterprises in the former GDR had used the social protection system to shed older workers disproportionately during the transition. Whereas social protection had been heavily tied to the workplace and many had continued to work after 55 under the communist regime, there was a striking decline in these workers' labour force participation after 1989. Firms had employed a wide array of social protection measures, in particular early retirement benefits, to shed these workers. Enterprise restructuring not only created distributional problems but also resulted in the scaling back of many services, except where these were customary in the West or governments had successfully pushed the burden of social provision onto firms.

Winfried Schmähl (Universität Bremen) cautioned against generalizing from the ex-GDR's transition to the rest of Central and Eastern Europe, given the transfer of resources and an administrative system from the West and the absence of effective resistance to reform. The effective currency appreciation of 300% after 1989 was also unique among former communist states and will continue to restrict enterprises' provision of social benefits.
In `Social Protection, Labor Market Rigidity, and Enterprise Restructuring in China', Zuliu Hu (IMF) drew lessons from 15 years of economic reform in China for current attempts to encourage labour market flexibility. Up to the mid-1970s, large state-owned enterprises (SOEs) had accounted for four-fifths of urban employment and had been saddled with an array of social responsibilities including lifetime job security, housing and pensions (financed out of current revenue). Faced with a demographic influx into the job market, Chinese reformers opted to encourage employment in collective (and to some extent private) enterprises while pursuing a cautious liberalization of the state sector. Local and regional governments had made some headway in improving the availability of housing and in socializing firm pensions and unemployment insurance; at the same time, the success of collective enterprises (which had created 15 million new jobs since 1978) had placed competitive pressures on SOEs, thus encouraging job mobility (e.g. through fixed-term job contracts) and linking compensation to performance through bonuses.

Barry Friedman emphasized that the outstanding feature of Chinese reform its resistance to privatizing SOEs had slowed overall progress, adding that the ability of state enterprises to lay off redundant workers had remained limited even after the 1986 overhaul of unemployment insurance.

Richard Portes (CEPR and Birkbeck College, London) launched the general discussion by emphasizing the need to consider the extent to which the social role of the enterprise represents a burden to a successful transition and whether the West European model of the welfare state is even feasible for these economies. Jan Svejnar suggested that search, turnover and monitoring costs in labour markets would generate some level of enterprise benefits and worker immobility even without state intervention; `adding the state' would generally increase social protection by promoting trade unionism, mandatory provision of public goods such as worker safety and training, and distributional concerns. Economic transition had led to new capacities and constraints for firms, workers and the state, and it would therefore require new negotiated outcomes (e.g. fewer benefits in the newer, smaller private firms). Fabrizio Coricelli called for a closer focus on the transition process itself to investigate the effects of the enterprise's social role on firm restructuring, worker behaviour and support for reform; who determines this role; and how much of the social burden governments could bear. Martin Kohli identified three dimensions of social protection: providers (state/firm/individual/civil society), types of provision (social services versus insurance) and recipients (in terms of distribution and social stratification). Robert Holzmann identified the key public policy issue as whether firm benefits and contributions to state provision should be compulsory, Andreas Wörgötter framed this question in terms of a trade-off between employment opportunities and social benefits. Richard Jackman cautioned that in-kind benefits typically did not represent true social protection (in terms of uninsurable risks and `shortage goods'), while Douglas Lippoldt added that these firm-provided benefits were small in relation to mandatory contributions to the welfare state.