What's new
		New Volume The promise and perils of participatory policy making by Lucio Baccaro, Konstantinos Papadakis Discussion Paper 190 Executive compensation: Trends and policy issues by Franz Christian Ebert Raymond Torres Konstantinos Papadakis Discussion Paper 192 Labour, Globalization and Inequality: Are Trade Unions Still Redistributive? by Lucio Baccaro Discussion Paper 193 Impact of changing work patterns on income inequality, by Uma Rani Subscribe to our mailing list and we will keep you informed about our new publications

Agenda List of Participants Opening speech by Ms. Guigou Speech by Mr. Somavia Background Document Annecy Conference Papers ILO Press Release

Fritz W. Scharpf
Director
Max Planck Institute for the Study of Societies
Paulstrasse 3
D50676 Koeln
Germany

1 What Are the New Challenges?

The new challenges for advanced capitalist welfare states are well described in the program of this conference. I will begin by focusing on those that are being caused by the nation state's loss of control over its own economic boundaries as a consequence of European and international economic integration.^{(Endnote 1)} They include intensified international competition in product markets; uncontrolled international capital mobility; and the creation of the European Monetary Union.

These challenges increase the difficulties of maintaining the international viability and stability of national economies, and they have negative impacts on employment in the exposed sectors of the economy and on the financial viability of the welfare state. In the following sections, I will briefly describe the nature of these challenges, and I will try to show that economically effective policy options are, in principle, available at the national level that would permit countries to combine high levels of employment and high levels of social security with international economic viability. In conclusion, I will turn to the main problem in all countries, namely the political feasibility of economically effective policy responses.

2 Challenges to Employment

Over the last few decades, world trade in goods and services has grown much more rapidly than world GDP, and that the same is true for OECD countries and even more so for the member states of the European Union. The reasons are significant advances in trade liberalization achieved in successive rounds of GATT negotiations, and above all the successful completion on the Single-Market program in the European Union which has not only removed the remaining non-tariff barriers to trade but extended international competition into areas of service provision and infrastructure facilities which in the past had been exempted from market competition even within the national economies of member states.

If trade grows more rapidly than production, the consequence is more intense international competition -- which forces firms in all countries to use available technical and organizational opportunities for reducing production costs and increasing labor productivity in those sectors which are exposed to international competition. By now, these include not only agriculture and industry, but also construction and energy supply as well as transport, communications, financial and business services (i.e., all activities in the ISIC classes 1-5, 7 and 8). As a consequence of increasing competition and productivity, average employment rates in the exposed sectors of advanced OECD countries have steadily declined over the last three decades (Figure 1). If total employment remained remarkably stable on average, the effect is due to the rise of employment rates in the sheltered sectors of the economy -- i.e., in those services which are locally produced and locally consumed (ISIC 6 and 9).

2.1 Challenges in the Exposed Sectors

The diagram reflects changes over time in average OECD-18 employment rates, rather than the considerable differences among countries. It also does not reflect the shifts in the composition of exposed-sector employment which had the effect that large job losses in agriculture and industry have been partly compensated by employment gains in financial and business services. Nevertheless, total employment in the exposed sectors has been at best stagnant, but mostly declining over the last thirty years in the advanced industrial economies. This is not meant to suggest that these sectors are losing their economic importance. The wealth of countries depends on their ability to maintain or improve their competitiveness in internationalized markets for goods and services. Moreover, though employment rates in industry are declining everywhere, the speed of decline varies among countries, and the same is true of the rise of production-related services in ISIC 7 and 8. In other words, national structures and policies continue to matter. On average, in any case, about half of all jobs in OECD-18 countries are still provided in the exposed sectors. By and large, these are highly productive and hence well-paying jobs, and countries have every reason to defend them against further erosion.

But this defense has become more difficult. Protective measures and subsidies are constrained by the legal commitments of the GATT and WTO, and they are effectively ruled out among EU member states by the strict enforcement of European competition law. Among the member states of the European Monetary Union, moreover, the adjustment of exchange rates is also no longer available to correct for losses of international competitiveness. As a consequence, above-average increases of unit labor costs -- regardless of whether they are the result of high wage settlements, of increases in payroll taxes or of more stringent employment regulations -- will directly reduce the competitive position of national producers in the European markets. By the same token, competitiveness will be improved by all measures reducing the costs of production or increasing productivity -- regardless of whether they are achieved through wage restraint, through more flexible work organization, through outsourcing and greater specialization, through public support for technology transfer and innovation, or through training and retraining the workforce.

The importance of relative production costs is questioned by a literature which emphasizes the comparative advantages of different production regimes, asserting that high-cost-high-skill and low-cost-low-skill economies may be equally successful in different international product markets. (Soskice 1999). Fair enough. Nevertheless, once countries have joined the Monetary Union, any later changes in production costs that are not matched by corresponding changes in productivity will have a direct effect on market shares and employment. In other words, the existence of a common currency increases both, the penalties on excessive wage and tax increases and the temptations to adopt beggar-my-neighbor strategies. But since these risks are well understood, unions and governments should generally be able to deal with them.

It is less clear that they are equally prepared to deal with another risk arising from the Monetary Union. After the demise of Keynesianism in the early 1980s, the primary responsibility for managing national economies had shifted to central banks -- whose "non-accommodating" policies defined the narrow monetary corridors within which governments and unions had to operate. But Euroland is not an "optimal currency area", and business cycles in its member economies are not perfectly synchronized. Hence the monetary corridor defined by the ECB, which can only respond to average conditions in the Union as a whole, may be either too wide or too tight for some countries. Where it is too wide (as is presently true for Ireland), nominal Euro interest rates will turn into very low or even negative real interest rates. The likely outcomes are an overheating economy, escalating rates of local inflation, a loss of international market shares, and ultimately massive job losses in the exposed sectors. If the ECB monetary corridor is tighter than is justified by the state of a national economy, real interest rates will rise and local consumer and investment demand will be reduced, even though the economy is not overheating and unemployment may be high.

Under either of these conditions, disaster can only be avoided if national governments and unions reassume a more active role in macro-economic management than they had played in the "monetarist" 1980s and 1990s. If monetary policy is too loose, governments must be ready to impose tax increases and to cut expenditures in order to reduce aggregate demand; and unions should at least prevent unit labor costs from rising faster than the rate of inflation. If the monetary corridor should be too tight for a country, fiscal expansion would be called for -- which, under the constraints of the EMU Stability Pact, presupposes that governments had built up fiscal surpluses, or at least avoided deficits, in normal periods. Moreover, to avoid massive job losses under conditions of excessive monetary constraint, wage settlements would need to reduce real unit labor costs in order to allow real aggregate demand to rise as a consequence of falling prices.

In other words: The potential mismatches between an EMU-wide monetary policy and national economic conditions increases the danger of policy-induced job losses in individual member states. If these are to be avoided, there is a need for strengthening the remaining national capabilities for macro-economic management. Since these must be exercised in response to differing economic conditions, and in the face of differing institutional and political constraints, they should and could not be preempted by the directives of an EMU-wide "European economic government" which, again, could only respond to average conditions in the EMU area.

2.2 Sheltered Sectors

These difficulties of macro-economic management in the member states of the Monetary Union primarily affect employment in the exposed sectors of the economy. While they will also be felt in the sheltered sectors, these are more affected by domestic conditions -- which, however, seem to differ greatly between countries. It remains true that employment in those services that are locally supplied and locally consumed has increased everywhere, and that only countries with high levels of sheltered-sector employment have been able to maintain or achieve high rates of total employment (Table 1, columns 1 and 3). But these services differ in the way they are provided and financed, and so do the conditions which allow them to expand.

The main theoretically salient difference is between public and private financing. While the consumer-oriented services in "wholesale and retail trade, restaurants and hotels" (ISIC 6) or in house repair and maintenance are typically produced and consumed in the private sector of the economy, that is not generally true of the "community, social, and personal services" (ISIC 9), which include the large blocks of education and health care which in most countries are primarily financed from public sources, but may or may not be provided by the state itself.

From a comparative perspective, it is important to note that those countries which have comparatively high rates of employment in the sheltered sectors are characterized by contrasting public-private-sector profiles: If we follow the classification introduced by Esping-Andersen (1990),^{(Endnote 2)} the Scandinavian welfare states tend to have very high levels of government employment and relatively few jobs in private-sector services,^{(Endnote 3)} while this pattern is reversed in the Anglo-Saxon countries and Switzerland. Continental countries, however, seem to conform to the Anglo-Saxon pattern in having very few public-sector jobs, while they conform to the Scandinavian model in having relatively low rates of private-service employment (Table 1, columns 4 and 5). It is no surprise therefore to see that they not only have the lowest rates of employment in the sheltered sector but also the lowest rates of total employment.

The explanations of these different patterns of sheltered-sector employment are directly related to the functions performed by different types of welfare states, and to the ways in which they are being financed. Countries have widely differing dividing lines between the functions the welfare state is expected to perform and those that are left to private provision, either in the family or by the market (Figure 2).

All three types of welfare states provide means tested social assistance that assures basic incomes and social services for the poor. Beyond that, Scandinavian and Continental welfare states have also assumed responsibility for providing earnings-related social insurance that is meant to secure the standard of living of average-income families in the case of unemployment, sickness, disability and in old age. In Britain and other Anglo-Saxon welfare states, by contrast, workers with average or higher incomes have to rely on private provisions for these eventualities. Moreover, only the Scandinavian welfare states are providing universal and high-quality social services that have alleviated the family duties of wives and mothers while at the same time providing the public-sector jobs that have raised female participation in the labor market to record levels (Table 2, column 1). In Continental countries, such services are to a large extent still provided in the family, while in Anglo-Saxon countries the market plays a bigger role (Scharpf and Schmidt 2000).

The conclusion, then, seems to be that high levels of employment can only be achieved in countries which are willing and able to transform a large part of informal household and family work into paid employment. The size and the specific patterns of welfare state functions and finance will encourage or discourage this transformation, and they will also determine whether the additional employment opportunities will evolve in the public or in the private sector of the economy. The Anglo-Saxon and Scandinavian societies have made their choice. They have accepted or promoted the growing participation of women in the labor market, but they have opted for different solutions: Scandinavia has expanded publicly financed social services, whereas the Anglo-Saxon welfare states have created conditions that allow the expansion of affordable private services.

Under the influence of catholic social theory, most Continental countries have so far avoided this choice. While they no longer discriminate explicitly against women in the labor market, they continue to emphasize, and subsidize, the traditional family roles of married women (Daly 2000). Hence they have done little to help women to combine employment and family responsibilities through publicly financed child care facilities, all-day schools, and home-care services. At the same time, their systems of taxation and social protection tend to reduce the attractiveness of part-time work for married women, while prevailing patterns of welfare-state finance and labor-market regulations tend to increase the cost of private services for employers and to reduce their attractiveness for potential customers.

But even though the policy patterns of the Continental welfare state may continue to have political support, their socio-economic viability is rapidly eroding. Forced to choose between having children or having a career, better-trained women have increasingly opted for the latter. Thus, the countries with the lowest rates of female employment also tend to be the countries with the lowest birth rates -- with dire consequences for the age distribution of their societies and for the financial viability of their comparatively generous public pension systems.

But whereas pension reform is now on the political agenda of all Continental welfare states, few countries (with the Netherlands as a noteworthy exception) have begun to change the underlying structural conditions that impede the creation of employment opportunities in the sheltered services. In part the inertia may be due to a continuing ideological bias (and to the under-representation of women's interests) in government and union decisions. Yet even if policy makers had the best of intentions, effective employment strategies would still confront a double jeopardy: They would presuppose a basic choice between public- and private-sector solutions which, for practical purposes are mutually exclusive, and whose pros and cons are bound to divide supporters. Moreover, either one of these solutions could only succeed against massive political opposition.

2.2.1 Public Services

In the late 1990s, government employment amounted to 22.6 percent of the working-age population in Denmark, but only to 8 percent in Italy (Table 1, column 4). By the same token, female participation rates are also very high in Sweden and Denmark (Table 2, column 1). These data reflect the strong political commitment of Scandinavian welfare states to providing high-quality and universal services to families with small children, to the sick, the handicapped, and the old (Table 2, column 2). Lacking this commitment, public expenditures on social services and public-sector employment are equally low in Continental and Anglo-Saxon welfare states. Nevertheless, female participation rates are considerably higher in Anglo-Saxon countries than they are on the European Continent -- which reflects conditions favoring or impeding private-sector service employment that I will discuss below.

In the Scandinavian welfare states, record rates of public-sector employment are matched by record levels of social spending (Table 2, column 3) and of total taxation (Table 2, column 4), whereas social spending and tax burdens are quite low in Anglo-Saxon countries. Continental countries are at an intermediate level in both regards because, in contrast to the Anglo-Saxon pattern, they share the Scandinavian commitment to state-provided social insurance. In other words, in order to expand publicly financed social services, Continental countries would either have to raise taxes and/or social security contributions to Scandinavian levels, or they would have to reduce social-insurance spending to Anglo-Saxon levels. But neither of these options appears politically feasible.

Continental governments have for some time struggled with the need to reduce the generosity of public pensions in order to cope with the demographic challenges of rapidly aging populations without raising social insurance contributions to ever higher levels. Since citizens had based their life plans on the trustworthiness of existing pension systems, however, political opposition against such cuts is very strong, and retrenchment, if it is achieved at all, must be gradual and quite limited. Under these conditions, it seems out of the question that benefits could be reduced further, by about three or four percentage points of GDP (Table 2, column 2), in order to increase publicly financed social services to the levels realized in Sweden and Denmark.

At the same time, tax increases of similar proportions are also not a politically realistic proposition. Within the European Union, capital and business investments have become highly mobile. As a consequence, member states are in competition over internationally mobile tax bases, which creates downward pressures on the taxation of capital incomes and business profits. Since efforts at European, let alone international, tax harmonization have not been successful, these pressures are likely to continue. Yet if taxes on business and capital incomes are reduced for economic reasons, political demands for a general reduction of personal income taxes also become stronger, and governments could hardly dare to propose increases instead (Ganghof 2000).

In the 1990s, it is true, some governments have increased taxes on immobile factors of production -- primarily, social insurance contributions -- and in Germany, for instance, some of these revenues are indeed being used to finance additional employment in long-term home care for the elderly (Manow and Seils 2000). In the meantime, however, the member states of the European Monetary Union have come to realize that in the absence of exchange-rate adjustments all increases in the costs of production will reduce international competitiveness and hence employment in the exposed sectors. As a consequence, governments are now struggling to reduce non-wage labor costs or at least to stop the further rise of social contributions. That leaves value-added taxes whose increase will not affect competitiveness as long as the EU does not switch from "country of destination" to "country of origin" rules. But the distributional impact of consumption taxes is thought to be regressive; moreover, VAT rates are already quite high in many countries so that further increases would not only run into strong political opposition, but also increase incentives for tax evasion.

Conceivably, many of these obstacles could be overcome by well-designed revenue-raising strategies -- if there were strong political support for a major expansion of publicly financed social services. But that does not exist. The reasons are not only ideological. The expansion of social services in Scandinavia began in the high-growth period of the late 1950s and 1960s, when opportunity costs were relatively low. It continued, it is true, in the crisis ridden 1970s and early 1980s, but by that time the virtuous cycle was already in operation through which the availability of professional caring services -- allowing wives and mothers to seek employment and creating a large number of public-sector jobs which women could take -- was generating the political support for further expansion (Benner and Vad 2000. In Continental countries and in the present period, however, these jobs and the political support which they could create do not yet exist, and any expansion of public services would not only run counter to the Zeitgeist of the lean state, but would have to be achieved at very high opportunity costs in terms of tax increases or expenditure cutbacks in other areas.

In short: The Scandinavian solution does not seem to provide a politically viable solution to the present employment problems of Continental welfare states. But what about Anglo-Saxon solutions?

2.2.2 Private Services

The Anglo-Saxon countries have achieved high levels of sheltered-sector employment through the expansion of privately financed services. They include an increasing number of highly productive and well paid jobs in education, further education and training, health care and other forms of therapy and counseling, entertainment, and the media. These also exist in Scandinavian and Continental countries. There, however, the expansion of high-skill employment is held back by the fact that public financing plays a comparatively larger role in health care and education, and that tightening fiscal constraints have called for retrenchment rather than growth in public services during recent decades.

At the same time, however, the expansion of private services in the Anglo-Saxon countries has also created large numbers of less productive jobs with relatively low skill requirements and correspondingly low market wages. In the employment statistics, these are found in ISIC 9 as well as in ISIC 6, but as we cannot distinguish between public and private financing in ISIC 9, we take ISIC-6 data as a proxy, since employment in hotels, restaurants and trade tends to be in the private sector, and much of it is characterized by relatively low skill requirements, low labor productivity, and low wages (Table 1, column 5).

In the Anglo-Saxon countries, the expansion of such services is facilitated by a relatively low tax burden and by largely deregulated labor markets with weak unions, high wage dispersion and low minimum wages. However, the market wages earned in low-skill jobs are often insufficient to provide for the needs of families with children. In countries like the United States or New Zealand, where welfare benefits are low, the resulting problem is "working poverty"; in the United Kingdom and other countries where social assistance benefits are more adequate, the consequence are "workless families". In both cases, the solutions that have been adopted are in the nature of "in-work benefits" -- the Earned Income Tax Credit" in the United States, or the Working Families Tax Credit in Britain -- which, according to the logic of the negative income tax, complement the wages of low-paid workers with social transfer payments that are only gradually withdrawn as income from work increases (Howard 1997; Rhodes 2000; Schwartz 2000). Where these transfers are sufficiently generous, employment opportunities for low-skilled workers can expand in the private services without entailing a dramatic rise in poverty and inequality.

In Scandinavian and Continental countries, by contrast, structural conditions are generally hostile to the expansion of low-paid private service employment. Tax burdens are high or very high, labor markets are highly regulated and inflexible in most countries (with Denmark as a notable exception), unions are often very strong, and minimum wages (defined by collective bargaining agreements or by statute) are relatively high. As a consequence, many of the less productive services that are provided by private firms in the Anglo-Saxon countries are priced out of the market. For the Scandinavian countries that is less of a problem, because an important part of these services are publicly provided. For Continental countries, however, the result are low levels of total employment, low female participation rates and, increasingly, the denial of working opportunities for low-skilled groups. Even if the rise of poverty is held in check by adequate unemployment and social assistance benefits, the structural exclusion from the labor market of a substantial part of the working-age population must be considered a threat to social cohesion.

Better education and training must surely be part of any solution, especially for second- and third-generation immigrants whose skill potential is vastly underdeveloped and underutilized in most European countries. But not everybody can be trained to be a computer specialist; and in any case, supply-side measures alone cannot suffice where the demand for labor is structurally inadequate. In other words, Continental countries cannot expect to overcome persistent underemployment unless they are able to create structural conditions that are more favorable to an expansion of private service markets. The strategies that would allow them to do so are, in principle, well known (Hemerijck and Schludi 2000): They would need to reduce the negative employment effects of high tax burdens; they would need to transform generous benefits for the unemployed into in-work benefits for low-paid workers; and they would need to increase the flexibility of labor markets.

In theory, the first of these requirements could be met most easily. In our project, we have shown that private-service employment is hardly affected at all by the level of income and corporation taxes, but is highly sensitive to social security contributions and consumption taxes (Figures 3 and 4). The reason is straightforward: Income taxes are progressive, and they are not imposed on incomes below a basic exemption. Hence they hardly affect low-wage and part-time employment. Social security contributions, by contrast, are generally proportional (and in fact regressive, since they are capped at higher wage levels) and they are usually collected without an exemption. In quite a few countries, therefore, wages from low-paid and part-time jobs are taxed at a rate that exceeds the rate of income taxes levied on millionaires. If these contributions are borne by the worker, net incomes will often be reduced below the level of social assistance benefits. In order to find willing workers, therefore, employers would need to bear most of the burden -- with the consequence that the costs of low-skilled labor, and hence the price of less productive services will rise to levels where potential customers are likely to turn to self-service (Gershuny 1978) or to "black-market" suppliers. The same is true if the price of services is raised by high VAT rates.

Since the revenue mix of Continental countries is characterized by an extremely high reliance on social security contributions (Table 2, column 5) as well as relatively high consumption taxes, the appropriate solution seems obvious: VAT and social-security contributions should be reduced for low-paid services, and the revenue loss should be made up by increasing the intake from income taxes. Some countries (including France, Belgium and the Netherlands) are slowly moving in this direction, but elsewhere such reforms have been blocked by trade unions that are, in principle, opposed to any expansion of low-wage employment and "bad jobs". By the same logic, unions also tend to oppose the transformation of unemployment benefits into in-work benefits for workers accepting low-paid jobs.

By itself, however, lowering the costs of labor will not have major employment effects unless it is accompanied by other reforms. Increasing the flexibility of employment relations is particularly important in countries where private services are underdeveloped and can only expand if existing firms find new markets and if new firms will develop new types of products in the hope of finding a market for them. Under such conditions, the rigidity of employment protection legislation -- which generally is very high in Continental welfare states (OECD 1999) -- will discourage hiring when firms are uncertain that demand for their services will be stable. Moreover, the expansion of service markets will also be impeded by regulations that protect traditional professions by restricting market access and the creation of startup firms. In both regards, however, attempts at deregulation tend to encounter the fierce opposition of labor unions and professional organizations, while political support remains weak since the firms that would benefit from them do not yet exist and potential customers are not organized.

3 Conclusions

All advanced welfare states are exposed to more intense international competition in product markets and unrestricted capital mobility. On average, the consequences have been declining employment rates in internationally exposed sectors and tightening constraints on domestic policy choices caused by tax competition, regulatory competition and wage competition. Countries that nevertheless were able to maintain or even increase total employment, owe their success to increases in the sheltered sectors, where services are locally provided and locally consumed. But these increases have been achieved in diametrically opposed configurations that are directly related to the functions assumed by different types of welfare states.

In the Scandinavian welfare states, very high total employment rates are achieved in consequence of a political commitment to provide high-quality and universal social services for families with young children, for the sick, for the handicapped and for the aged in the public sector. Since these services are primarily financed by very high income taxes (with relatively low rates for incomes from capital), they are relatively immune to international competition in product markets and to tax competition. If there are problems, they would arise from political tax opposition which, however, remains muted as long as middle-class voters see themselves as beneficiaries of the service-intensive welfare state.

By contrast, public-sector employment is very low in Anglo-Saxon welfare states. If they nevertheless achieve high levels of total employment it is because of the expansion of services in the private sector. It is facilitated by low tax burdens and highly flexible labor markets. If there are problems, they are the consequences of increasing inequality and poverty among low-skilled workers depending on low-wage jobs. The remedy that is preferred by Anglo-Saxon welfare states are in-work benefits that subsidize incomes from low-wage employment. However, the generosity of this solution, ad hence its effectiveness, depends on the political support of middle-class voters who -- in contrast to the politics of the Scandinavian welfare state -- are not themselves among the beneficiaries of the programs for which they are asked to pay.

The Continental welfare states, finally, are in the worst of both worlds. Their total employment rates are lowest since they have as little private-sector employment as Scandinavian countries, and as few jobs in the publicly financed services as Anglo-Saxon countries. Moreover, the traditional social pattern, on which the Continental model was based, is rapidly eroding as women are no longer willing to forego formal employment in order to perform socially necessary caring services at home.

At the same time, however, the theoretically available solutions to the problems of Continental underemployment are confronted by extreme political difficulties. Under present economic conditions, proposals to expand publicly financed social services to the Scandinavian level would have high opportunity costs. Expenditure cuts as well as tax increases that could finance the expansion would run into fierce political opposition. Even more important: There are no organized and politically influential groups (apart from unions engaged in rearguard battles against public-sector retrenchment) to whose popular demands governments could respond. At the same time, however, economically more feasible strategies that would increase employment opportunities in private-sector services are resisted by unions opposed to the creation of a low-wage labor market, no matter how publicly subsidized, and by professional interests opposed to the deregulation and liberalization of protected, if stagnant, service markets.

In short, like the maintenance of financial support for the Scandinavian service state and the success of anti-poverty strategies in Anglo-Saxon countries, the fight against persistent underemployment in Continental welfare states is, at bottom, a political challenge. Economically viable solutions are available, but they cannot be used -- against the veto positions of entrenched interests and the populist temptations of opposition parties -- unless governments are able to develop public discourses (Schmidt 2000) that are able to convince their electorates of the blatant injustice and the societal risks of excluding large parts of the population from participation in the processes of social production, and of the feasibility and normative acceptability of policies that would reduce underemployment.

References

Benner, Mats and Torben Bundgaard Vad 2000: Sweden and Denmark: Defending the Welfare State. In: Scharpf and Schmidt 2000a: 399-466.

Daly, Mary 2000: A Fine Balance: Women's Labor Market Participation in International Comparison. In: Scharpf and Schmidt 2000a: 467-510.

Esping-Andersen, Gøsta 1990: The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press.

Ferrera, Maurizio, Anton C. Hemerijck and Martin Rhodes 2000: The Future of Social Europe. Oeiras: Celta.

Ganghof, Steffen 2000: Adjusting National Tax Policies to Economic Internationalization: Strategies and Outcomes. In: Scharpf and Schmidt 2000a: 597-646.

Gerschuny, Jonathan I. 1978: After Industrial Society? The Emerging Self-Service Economy. London: Macmillan.

Hemerijck, Anton C. and Martin Schludi 2000: Sequences of Policy Failures and Effective Policy Responses. In: Scharpf and Schmidt 2000: 125-228.

Howard, Christopher 2000: The Hidden Welfare State. Tax Expenditures and Social Policy in the United States. Princeton: Princeton University Press.

Manow, Philip and Eric Seils 2000: Adjusting Badly: The German Welfare State, Structural Change, and the Open Economy. In: Scharpf and Schmidt 2000a: 264-307.

OECD 1999: Employment Protection and Labour Market Performance. In: Employment Outlook June 1999: 47-132.

Rhodes, Martin 2000: Restructuring the British Welfare State: Between Domestic Constraints ad Global Imperatives. In: Scharpf and Schmidt 2000a: 19-68.

Scharpf, Fritz W. 2000: Economic Changes, Vulnerabilities, and Institutional Capabilities. In: Scharpf and Schmidt 2000: 21-124.

Scharpf, Fritz W. and Vivien A. Schmidt., eds., 2000: Welfare and Work in the Open Economy. Vol. 1: From Vulnerability to Competitiveness. Oxford: Oxford University Press.

Scharpf, Fritz W. and Vivien A. Schmidt., eds., 2000a: Welfare and Work in the Open Economy. Vol. 2: Diverse Responses to Common Challenges. Oxford: Oxford University Press.

Schmidt, Vivien A. 2000: Values and Discourse in the Politics of Adjustment. In: Scharpf and Schmidt 2000: 229-309.

Schwartz, Herman 2000: Internationalization and Two Liberal Welfare States: Australia and New Zealand. In: Scharpf and Schmidt 2000a: 69-130.

Soskice, David 1999: Divergent Production Regimes: Coordinated and Uncoordinated Market Economies in the 1980s and 1990s. In: Herbert Kitschelt, Peter Lange, Gary Marks and John D. Stephens, eds., Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge University Press: 101-134.

Table 1: Total and Sectoral Employment as % of the Population 15-64.

Table 2: Selected Indicators of Welfare State Performance 1995-96.

Figure 3: Income Taxes and Employment in ISIC-6 Services (OECD Data 1997)

Figure 4: Social Contributions + Consumption Taxes and ISIC-6 Employment (1997)

Endnote 1:
My remarks are based on the findings of a comparative study of the adjustment of twelve Scandinavian Continental and Anglo-Saxon welfare states to international economic challenges in the period from the early 1970s to the late 1990s (Scharpf and Schmidt 2000; 2000a).

Endnote 2:
A complete classification of EU countries would need to add a group of "Southern" welfare states (including Portugal, Spain and Greece -- and perhaps Italy) that share many of the Continental characteristics, but at a lower level of generosity and with less complete coverage (Ferrera, Hemerijck and Rhodes 2000). But since these countries were not covered by our own project, I will not discuss them here.

Endnote 3:
Since available data do not allow us to distinguish between publicly and privately financed employment in ISIC 9, we take employment in ISIC 6 (where employment in wholesale and retail trade, hotels and restaurants is generally located in the private sector) as a proxy for employment in privately financed services.