has published widely on the welfare state, labour markets, and contemporary class structures. His most recent book is Relations of Ruling (with Wallace Clements, McGill-Queens University Press, 1994). ■ Ito Peng, born in Taiwan, was educated in Japan and Canada and obtained her PhD from the London School of Economics. At the time of writing, she was Post-Doctoral Fellow associated with the Suntory and Toyota Inter-national Centre for Economics and Related Disciplines (STICERD) at LSE, and was undertaking further research on East Asian welfare regimes at the Institute of Social Science, University of Tokyo. She is now a Lecturer at Hokusei Gakuen University in Sapporo, Japan. She has worked at policy level with the Ministry of Health and Welfare in Japan and with the Ontario Provincial Government in Canada. Guy Standing is Director of Labour Market Polities in the International Labour Organization. From 1992 to 1994, he was Director of the ILO's Central and Eastern European Team, based in Budapest, responsible for the ILO's technical and advisory work. He has a doctorate in economics from the University of Cambridge, and among his most recent publications are Reviving Dead Souls: Enterprise Restructuring and Mass Unemployment in Russia (Macmillan, 1996) and Minimum Wages in Central and Eastern Europe: From Protection to Destitution (edited with Daniel Vaughan-Whitehead, Central European University Press, 1995). He is co-chairman of the Basic Income European Network. John D. Stephens is Professor of Political Science and Sociology at the University of North Carolina at Chapel Hill. He is the author of The Transition from Capitalism to Socialism (Macmillan, 1979), and the coauthor of Democratic Socialism in Jamaica (with Evelyne Huber Stephens, Princeton University Press, 1986) and of Capitalist Development and Democracy (with Dietrich Rueschemeyer and Evelyne Huber Stephens, Polity Press and University of Chicago Press, 1992). He is currently working on a comparative historical and quantitative study of the social origins and outcomes of welfare states, and on the current impasse of social democracy. 1 After the Golden Age? Welfare State Dilemmas in a Global Economy G0sta Esping-Ander sen According to T.H. Marshall (1950)^ modern citizenship is the fruition of a democratization that spans three centuries. M thlTeighteenth century the foundations were laid with the principle of legal-civil rights; political rights emerged in the nineteenth century; and, as a preliminary culmination of the democratic ideal, we see the consolidation of social citizenship in the twentieth century. On the threshold of yet another century, legal and political rights appear firmly entrenched in most parts of the advanced, industrialized world. The same, however, cannot be said for social rights. Many believe that the welfare state has become incompatible with other cherished goals, such as economic development, full employment, and even personal liberties - that it is at odds with the fabric of advanced postindustrial capitalism. The case for the inevitability of a third historical stage of social citizenship also seems circumspect when we broaden our analysis beyond the old, mature democracies. Despite what ^jdQľ^úignJhegxy believed some decades ago, the new emerging industrial democracies do not appear set to converge along the Western welfare state path. Was T.H. Marshall, then, wrong to assume that modern civilization is cumulative and irreversible? Or, put differently, what kind of welfare state is likely to emerge in the future? The modern welfare slate ..became an intrinsic part of capitalism's postwar 'GoldenL Age', an era in which prosperity, equality, and full employment seemed in perfect harmony. It cannot be for lack of prosperity that welfare states are in crisis. The dizzying levels of postwar economic growth are long gone, but nevertheless real gross national product in the rich OECD countries has increased by a respectable 45 per cent since the oil crisis in the mid 1970s. Public (and private) social outlays, of course, grew even faster but this trend was generally arrested in the 1980s. It is in the equality/full-employment nexus that the essence of the crisis must be found. „_----------— There seem to be as many diagnoses of the1 welfare state crisis as there are experts. Most can, nonetheless, be conveniently subsumed under three main headings. There is, firstly, the 'market-distortion' view which argues ' that the welfare state stifles the market and erodes/incentives to work, save , , and invest. A second popular diagnosis focuses on the cataclysmic long-( term effects of population ageing. And a third group of arguments focuses, on the consequences of the new global economy, which mercilessly punishes ^/profligate governments and uncompetitive economies. Our study will not reject theie~~arguments. We basically agree that a —new^-nd-quit^fiindamenta^ and employment; that global competition does narrow the field of domestic policy choice; and that ageing is a problem. At the same time, we feel that these standard accounts are exaggerated and risk being misleading. In part, the diversity of welfare state types speaks against too much generalization. In part, we must be very careful to distinguish what are chiefly exogenous and en^graojjs^^oj^e^oObe^crisis. On the one hand, many of The difficulties that welfare states today face are caused by /market failure: that is, badly functioning labour markets produce an overload on existing social programmes. Some, of course, insist that this is the^/fault of the welfare state itself. Thus, on the other hand, there is possibTý"also welfarestcite failure: that is, the edifice of social protection in many countries~ls"ifrozen'Jn a^a^^o^p^ecoMini^ojcder^hat no longer obtains, rendering it incapable of responding adequately to new risks and needs. -*==—- The malaise that now afflicts the advanced welfare states influences also strategic thinking on social security development within the emerging industrial democracies. Most pointedly, there no longer seems to be a Swedish 'middle way'. The neo-liberals suggest that the road to growth and prosperity is paved with flexibjhTjLiyaddeiegulation. Their recommendation for Latin America and East-Central Europe is therefore to emulate Chilean -privatization rather than Swedish welfare statism. Critics hold that such a choice causes too much polarization and needless impoverishment, and that it may prove counter-productive for modernization. Comprehensive social security, they hold, is necessary because traditional familial, communal^ or private market welfare arrangements are wholly inadequate. It is also necessary because stable democracy demands a level of social integration^ that only genuine social citizenship can inculcate. Indeed, these were the very same issues that dominated in postwar Europe. Then, welfare state construction implied much more than a mere upgrading of existing social policies. In economic terms, the extension of income and employment security as a citizen's right meant a deliberate departure from the orthodoxies of the pure market. In moral terms, the welfare^tate^promised a more universal, classless justice andTolidarity of 'the people'; it was presented as a ray of hope to those who were asked to sacrifice for the common good in the war effort. The welfare state was therefore also a political project of nation-building: the affirmation of - liberal democracy againstthe twrn^perils of fascism and bolshevism. Many countries became self-proclaimed welfare states, not so much to give a label to their social policies as to foster national social integration. Such issues are of pressing concern in contemporary Asia, South America, and East Europe precisely because economic modernization tears apart the old institutions of social integration. Yet, policy makers in these nations also fear that such moral and political aims might jeopardize their comparative economic advantage (cheaper labour), traditional elite privileges (non-taxation of the rich in Latin America), or social culture --------(Cönfm^ramsnHrr~East~Asia-):----------------------------------------------------------- The advanced Western nations' welfare states were built to cater to an economy^jißmmatejd by industrial mass production. In the era of the ■'"""Xéynesian consensusjL-Čhere was no perceived trade-off between social securi^naiid economic growth, between equality and efficiency. This consensus has disappeared because the underlying assumptions no longer obtain. Non-inflationary demand-led growth within one country appears impossible; f fillip employmenTToday/ must be attained via services, given -industrial deciine; the conventional [male breadwinner family is_erodingJ \ fertiljty^sjEalling, and thp life course is increasingly 'non-standard') Such structural shifts challenge traditional social policy thinking. In many respects the symptoms of crisis are similar across all nations. In others, there is notable divergence. Europe's single largest problem is chronically high unemployment, while in North America it is rising inequaHty and poverty. Both symptomize what many believe is a basic trade-off between employment growth and generous egalitarian social protection. Heavy social contributions and taxes, high and rigid wages, and extensive job rights make the hiring of additional workers prohibitively costly and the labour market too inflexible. The case in favour of deregulation seems validated in the North American 'job miracle' of the 1980s even if this occurred against the backdrop of greater inequalities. Critics^ insist that the associated social costs of the American route-are~^ toO'-Mgh^in^enris^Qf polarization and poverty. They suggest ^^ demographic trends (especially population ageing), both of which threaten the future viability of present welfare state commitments. The demographic and economic problems have received most attention. The former are caused by the combination of low fertility and longer life expectancy which will engender burdensome dependency ratios and, without strong economic growth, severe fiscal strain. In the EEC, the age-dependency ratio will increase by 50 per cent between now and 2020; with existing rules and benefits, this will absorb an estimated additional 5-7 per cent of GDP (European Community, 1993a: 24). OECD (1988) projections until 2040 indicate that ageing alone will double or triple health and pension expenditures, especially in countries, like Japan, which experience unusually rapid ageing. Still, population ageing does not automatically imply crisis. In part, the cost of ageing depends on long-run productivity growth. The OECD (1988: 70) estimates that real earnings growth at an annual average rate of 0.5-1.2 per cent (depending on nation) will suffice to finance the additional pension expenditures.6 More to the point, the demographic burden is subject to political management. Many countries are today reversing" a decades-long policy of lowering retirement age. Also, maximizing employment will automatically lower dependency rates. It makes a huge difference when, as in Scandinavia today, the overall activity rate is 10 or even 15 per cent higher than in continental Europe. Here, again, it is decisive whether social policy encouragesjow femaleemployment and early retirement~((aš~in the EEC nations), or maximum participation (as in ScandinaviaJ.7 It is also decisive whether, as fi~Southerrr Europe and Latin America, the incidence of informal, black market employment is high and growing. The spread of irregular work in countries like Italy is very much part of an inbuilt negative spiral: the heavy social contribution burdens incurred by overloaded income maintenance programmes stimulate informal employment which, in turn, further erodes the tax base. The ageing problem depends mainly on births. It is often feared that female employment will jeopardize Térfilíťý, and thus aggravate the ageing crisis. The facts, however, tell a different story. High fertility may accompany low female employment (as in Ireland), but then it may not (Italy _ and Spain have, today, Europe's lowest fertility levels). Female employ-n^eiatänŕ/fca^ welfare state makes aäečlsive difference because female employment with fertility^ possible if social services and liberal provisions for leave are available. They are in Sweden, but not in most of continental Europe. To the extent that S|3l|tS; '?rovaen'* economic independence is a defining element of postindustrial value-added production: hence, the East Asian governments' phenomenal stress on education. What, then, are the prospects for the welfare state as we step into the twenty-first century? Will the advanced nations be forced to sacrifice some, or even most, of their welfare state principles? Will the new industrializing nations opt for a model without a welfare state or, alternatively, adopt some varianrof~Western sťyle~welfäTe™3tates? Overall trends, alas, give little comfort to those who adhere to the ideals of the welfare state, at least as it was traditionally conceived. The new conflict between equality__a.nd employment that the advanced nations confronris^incrěašmglý difficult to harmonize. The conditions that made the welfare state an essential part of economic development in the postwar Western nations may not apply to, say, contemporary Argentina, Poland, or South Korea. The causes of such pessimism are to be found in both international and domestic change. The changing international environment The harmonious coexistence of full employment and income equalization that defined the postwar epoch appears no longer possible. Many believe that North America's positive employment performance could only be achieved by deregulation and freed markets which, in turn, reward the winners and punish the" losers: hence, rising wage and household income Inequalities, growing poverty rates, and maybe even the re-emergence of an 'underclass' (Gottschalk, 1993; OECD, 1993; Jencks and Peterson, 1991; Room, 1990). Western Europe, with its much more comprehensive industrial relations systems, welfare states, and also powerful trade unions, has maintained equality and avoided growing poverty, but at the price of heavy (especially youth and long-term) unemployment, and swelling armies of welfare dependants, the combination of which overburdens social security finances. Demand-led, reflationary strategies are no longer an option, partly because unemployment is not merely cyclical, and partly because income growth leaks out of the economy to purchase imported goods.1 The case for convergence: global integration Integration in the world today almost automatically implies open economies. Sweden, Australia and New Zealand, Chile, and the ex-communist countries in Europe, are all shedding the protectionist measures that once upheld their respective welfare state arrangements. Openness is said to sharply restrict nations' capacity to autonomously design~Their own political economy. Both Australia and Sweden illustrate the erosion of national options. As Castles shows in Chapter 4, Australia could pursue what he calls the 'wage earners' welfare state' model of job security, full employment and high wages only as long as it adhered to j-Lj ici me \jvuA.tii /ige; h protectionist measures. The price that Australia paid was lagging growth. Sweden, as Stephens shows in Chapter 2, could balance (over-) full employment with the world's most generous and egalitarian welfare state only as long as governments could control domestic credit and investments, and as long as the labour market partners could guarantee wage moderation consensually. Following liberalization in the early_I9_8-Qs^the Swedish ěconomjrsufferediieavy. capitäHMkage_abmad7Thus undercutting domestic" investment and job generation. At the same time, Sweden's tradition of centralized national social pacts eroded. Enhanced openness in both countries has compelled governments (both left and right) to cut back social expenditure. Is it, then, the case that openness inexorably drives welfare states towards a lowest common welfare denominator? Much of Latin America and East-CentraLEurope is presently undergoing harsh liberal adjustment strategies. In the short ran this tends to cause heavy unemployment, an often dramatic fall in incomes, and more inequalities. In the longer run - as the case of. Chile since the mid 1980s suggests - it can improve nations' competitiveness, growth, and thus employment.2 The problem with radical liberalization is that its costs are üneqTIälly^j^ provoke organized-resistance. The Chiíeanlcašéls'illustrative. Huber shows in Chapter 6 that Chile's poverty rate rose from 17 per cent in 1970 to 38 per cent in 1986. In 1983, the unemploymenťräte reached a third of thelabour force.3 In authoritarian Chile, organized resistance was effectively crushed. In liberal democracies, policy makers will have to rely on either persuasion or compensatory social guarantees. Persuasion assumes broad consensus, while compensation may strain governments' already fragile finances. In Latin America, as in East and Central Europe, the gap between social need and financial means is deepened by rising 'informalization' of employment. Employers and workers exit from the formal employment relationship to dodge taxation and job regulations. If global wage competition is a major source; _ of .we.]^e_state_crisis inthe advanced7~nations,,.-convergence may paradoxically emerge from two t)ppoirte^responses. Lowering wage costs in Europe and America may, at least in the interim, safeguard otherwise uncompetitive domestic firms. The offshoot, of course, is an implicit sanctioning of poor productivity performance. The other source of convergence would come from rising labour costs amongjhe jnain global competitors, such as Japan, Korea or Taiwan. Their relative labour costs have been rising, and will do so even more if, as our study believes, these countries are hard put to stall major social security reforms in coming years. p The case for divergence: the role of institutions There are additional reasons why we should not exaggerate the degree s^ to which global forces overdetermine the fate of national welfare states. After the Golden Age? society, the contemporary famityjieedj_the welfare state in order^to harmonize work and family objectives; likewise, the welfare state needs childrenT " ~~ ~~ ~~ ^ The economic problems that confront the Western welfare states are typicMy^lHeĽÍmedľ'ín"lems_jf_Äejmemjp|oyment problem. Thertiombi-nation of high wage_costs (due to mandatory sociB^óntrjbutions) and rigidities (such as job~tenure, costly termination payments^or generous social benefits) is widely regarded as a main impediment to jgb_growth. Generous social benefits are also held to reduce the work incentive. There is evidence that high marginal labour costs and stringent job rights prohibit job growth. However, privatization of social security^may not offer a real solution. Firstly, as we knów-ďřom the iJňíted States and, more rečenťlýrChile, private plans deper^cm^favourable Jaxjconcesjions, tha.ys on public Jubsidization, Secondly, experience from the United. States shows that denned^b^nefittype occupational welfare plans may incur the exact same kinds of rigidities and cost burdens as social insurance. They tend to -t- inhibit labour mobility because workers fear to lose benefits, and because of vesting" requirements (the norm in the US is a five-year minimum); like social security, private plans also impose high fixed labour costs.8 Hence, public sector efforts to trim social security are paralleled in the private sector. In the United States, coverage under occupational plans has declined dramatically in the past decade: medical care coverage by 14 per cent, defined-benefit retirement coverage by 25 per cent. In its place have grown individual contribution plans. Postindustrial employment trends are also potentially problematic. Since they favour professionaränd^EUeďôccupations, demand for unqualified labour will depend mainly on low wages. • They also seem to foster ^atypical', precarious jobs, such as in cctóingentjwork, involuntary part-timewOTk7hôm£WOT the consequence may be greater poMSätioF"between € core and a p^pher^ ČoTSnunlíy, 1993b; OECD, 1993). The Umted^Šteteiecgoys'comparatively low unemployment, but a disturbing rise in jobs that pay belowrpoyerty wages. The level of many social benefits has followed suit, producing unprecedented levels of poverty. Indeed, as we see in the United States, wage decline may easily produce a vicious downward spiral of social benefits too, since adequate social transfers in a low-wage environment are likely to nurture poverty traps. Hence, both unemployment insurance and social welfare have eroded ŕ noticeably^ Poverty and polarization, in turn, may threaten the social^order i and thus burden the public sector on alternative expenditure accounts^ The V^ American male prison population is aJ5ove_j)ne million (and is rising), ^ pushing up spending on prisons, law and order. Security guards and law enforcement personnel are among the fastest growing occupations; the annual per-inmate cost of incarceration is almost twice that of tuition costs at Harvard University.9 The 'endogenous' problems of the welfare state He in the growing 1 After the Golden Age? 9 discrepancy between existing programme design and social demands. The contemporary welfare state addresses a past social order; its ideals of uni-versalism and equality emerged with reference to a relatively homogeneous industrial working class. The much greater occupational and fife cycle differentiation that characterizes 'postindustrial' society implies also more heterogeneous needs and expectations. With greater career uncertainty, demands for more flexible adjustment, and changing family arrangements, not to forget female employment, citizens also face more diverse risks. Also the welfare state's erstwhile 'm^deljamily' is no longer pre-eminent. On the one hand, we see the rise of ťnej^o^rner, double-career unit; on the other, thejise of divorced, single-person, and single-parent households. The former are often privileged, but it is also clear that wives' labour supply is becoming the only means by which lower-income households can escape poverty or maintain accustomed living standards today. This is evident in the American case (Mishel and Bernstein, 1993). 'Atypical' families constitute a rapidly growing high-risk poverty clientele.10 Welfare regime challenges in/ other regions / The ageing problem is, with the notable exception of Japan, less acute in other regions. However, an equally severe demographic problem is massive migration into urban industrial centres, a process which undermines traditional forms of social protection. In East Asia, this poses a dilemma between welfare state construction (in Japan and South Korea combined with corporate plans) and the Confucian tradition of familialism with its care obligations.11 The main economic problems of the 'non-welfare states' depend on their pôšTEiôn"' 5rThr"wbŕld" economy. In Eastern" Europe""the old communist welfare regime was characterized by three basic pillars: full and quasi-obligatory employment; broad and universalistic social insurance; and a highly developed, typically company-based, system of services and fringe benefits. In fact, very much as in Scandinavia, its employment-maximization strategy was the sine qua non of system equiHbrium since it assured minimal social dependencies. The post-democracy reforms have eroded the first and third of these pillars. Instead of full employment has emerged mass^ unemployment; the collapsing (or privatized) state Éllfe'r^prises are increasinglycapable of furnishing accustomed services. As / j:^^l.r ^e viability of both is destroyed, existing income maintenance programmes ť^yř;::,fače under-financing and over-buMemngľ Ťhe~consequence, as Standing (/ Igg^ shows in Chapter 8, is an akrming rise in poverty and mortality. %f^§jßyy: Where countries define their competitive, edge in terms of favourable |gj||p|^our costs, they will be wary of major welfare state advances. This is, Ipjjjl^owever, only partially the case. Following Japan, East Asia in general, lllpfegnd South Korea in particular, see their economic future in terms of an &Hg|to&ted workforce - very much like Sweden did with her 'productivistic' 10 After the Golden Age? welfare state design. This obviously implies growing commitments to education, health, and social services.12 A strong income maintenance system will probably be difficult to avoid in this scenario to th^lÄtraflhät (1) an increasingly educated, urbane, and professionalized labour force is likely to distance itself from the traditionalist principles of the Confucian culture; -^jid^Q)---occupational company schemes are highly uneven in coverage, ■ - - • ■" T_____ .C______ Yli being rarely present or even viable in small or medium firms. In contrast, Latin American development is to a much greater extent based on natural resources. As these countries abandon protectionist, import-substition policies they clearly face the labour cost problem more acutely. It is in this light that Chile's vanguard attempt to shift social security from state to market must be understood. Welfare state adaptation in the past decade Simmering symptoms of crisis became increasingly evident in the past decade. Popular perceptions notwithstanding, the degree of welfare state roll-back, let alone significant change, has so far been modest. This is clear from the essentially stable levels of social expenditure (see Table 1.1-1.3). Most nations, with the notable exception of Britain and New Zealand, have llimited-_interyention to marginal adjustments,) such as delayed^benefit indexation, diminished income rěplačěmeWfřates and, most recently, a return Jo contribution-based (rather than eärhmgs-bišed) pension benefit cjalculationľf Still, marginal cuts today may have long-term rc^CTEtive effects of avquite radical nature.) If social benefits gradually fall behind earnings, those who can seek compensation in private insurance will do so, thus weakening broad support for the welfare state. Among the 'new . nations' the signs of system change are more evident: on the one hand, active privatization in Latin America and East-Central Europe; on the other hand, embryonic welfare state construction in East Asia. Since the early 1970s, we can identify three distinct welfare state responses to economic and social change. [Scandinavia followed until recently a strategy of welfare state employment expansion. The Anglo-Saxon countries, in particular North America, New Zealand, and Britain, have favoured a strategy of deregulating wages and the labour market, combined with a certain degree of welfare state erosion. And the continental European nations, like Germany, France and Italy, have induced labour supply reduction while basically maintaining existing social security standards! All three strategies were intimately related to the nature of their welfare states. The Scandinavian route \j By the late 1960s, the Scandinavian welfare states had largely achieved1 their aims as far as income maintenance programmes are concerned. Albeit more comprehensive, universalistic, and generous, at this point the 'social democratic model' was not radically different from, say, the Dutch ot After the Golden Age? 11 Table 1.1 Public social security and health expenditures as a percentage of gross domestic product in selected countries, 1980-1990 OECD countries1 Canada2 Denmark France -Germany---------- Netherlands Norway Sweden United Kingdom United States Other countries3 Czechoslovakia Hungary Ukraine USSR Australia New Zealand Japan Argentina Brazil Chile Costa Rica 1980 1990 "17'.3 18.8 26.0 27.8 23.9 26.5 25.4 23.5 27.2 28.8 21.4 28.7 32.4 33.1 21.3 22.3 14.1 14.6 1975 1986 17.2 21.5 14.9 16.2 13.8 17.3 13.6 15.5 10.3 9.2 14.5 17.9 8.9 12.2 6.8 6.1 5.2 5.0 11.0 .13.1 5.1 7.3 1 These figures are based on OECD definitions which are not comparable with the ILO's. 2 Data for Canada refer to 1982 and 1990. 3 ILO-based data. For the ex-communist nations, spending is calculated in terms of net material product. Sources: OECD, Employment Outlook, Paris, 1994, Table 4.7; ILO, The Cost of Social Security, Geneva, 1991 ©'German. A truly distinct Nordic - and especially Swedish - model came ||ii.to being with the shift towards active labour market policies, social ^s|^ice_expansion, and gender equalization in the 1970s and 1980s. The ffipVe heralded an explicit second stage to consolidate both equality and a ||?ductivistic social policy by maximizing employment and equalizing the feus of women. It was, however, also motivated by growing employment ""^lems. Äth- a steady decline in manufacturing employment, and given Scan- "iSla's unusually egalitarian wage policies, it was from the start clear s^us'tained full employment, let alone the rise in women's employment, j|ishaye to rely on public sector service jobs. Indeed, until the mid Äp^en its expansion came to aThalt, this sector accounted for roughly *Ä|ént of total net job growth in Denmark and Sweden (with Norway p||ftd), and public employment now constitutes about 30 per cent of plf^10131 ^e Pomt °f VVSSN °f women's economic emancipation, the 1l|feeded- With public day care covering about 50 per cent of small After the Golden Age? Table 1.2 Social investment policies:1 (a) percentage of labour force involved in public training and employment measures, averaged 1990-1993; (b) percentage of 18-year-olds attending full-time education and training, 1990-1991 (a) (b) Denmark Sweden France Germany Italy Netherlands Australia Canada United Kingdom United States Japan Czech Republic Hungary (1992-3) Poland (1992-3) 12.8 69 6.3 56 9.9 78 4.9 81 4.8 na 3.0 74 4.9 52 3.3 58 2.0 25 2.6 55 0.1 na 1,7 na 3.0 na 3.6 na 1 These figures exclude general education and private training programmes Source: OECD, Employment Outlook, Paris, 1994, Table 1.18 and Table 1.B.3 Table 1.3 Population share of the elderly (aged 60-plus) in 1990 Share of aged (%) Europe Czech Republic 16.6 France 18.9 Germany 20.9 Hungary 19.0 Italy 19.9 Norway 21.2 Poland 14.8 Russia 15.3 Sweden 23.4 The Americas Argentina 13.1 Brazil 7.1 Canada 15.7 Chile 8.9 Costa Rica 6.4 United States 16.9 Asia and Pacific Australia 15.3 Japan 17.2 Korea (South) 7.4 New Zealand 15.1 Source: United Nations, Demographic Yearbook, New York, 1993 After the Golden Age? 13 children, and with generous provisions for paid maternity and parental leave, women's participation rates (also with small children) in Denmark and Sweden, at around 80 per cent, are higher than those of prime-aged males in the rest of Europe. Tne cgnsBqiJgnces of this strategy, intended or not, are both positive and negativeTOffTfif^ careers and' fertility. It has helped absorb unsMle^Pw^k^lri^eTrpaid employment. "And, it has also generated equality: the difference in men's and women's earnings and life cycle behaviour is rapidly eroding; the two-earner, double-career household is now the norm; and in comparison with everywhere else, the poverty rate among female-headed families is insignificant. And, with maximum employment levels, the welfare state is assured of higher tax revenue and lower dependency levels.13 On the negative side, the most dramatic result is an extremely high_ degree of_gender segregation, with women concentrated in (typically pärt-tim^"p^^Sörj^it and males in the private sector. Although this may partly reflect women's preference for the more flexible conditions of public employment, the high social costs, absenteeism rates, and disruptions to production that are connected with women's employment lead private employers to prefer male workers. Work absenteeism rates in Sweden are in fagtextremelyhjgh;14 Another, less nolič^a^ěTOTseqtience is'thevěrpiigh pro^rtLon^f Ä-skifled Mbeit well-paid) jobs that a^ sodal-seryices^led strategy_ producesrin facf^the "ov^řajTlharě^olTEsl^ service ]ob¥ is higher hiJDenmark and Sweden than in the notörit>us%äse of tEe~Ühited States (Esping-Ändérsen, Í993);)TEis; again, suggests an unpleasant tradeoff between either mass joblessness or mass suboptimal employment in services, be they driven by the private sector (as in America) or the public sector. Of course, it makes a huge difference from a welfare point of view that Scandinavian public employment offers good pay and security, but here we also arrive at the increasingly evident Achilles heel of the system: th^grqwjng ta^ With high rates of productivity growth the system can be sustained; when productivity or private investments are sluggish, severe cost problems emerge. This is exactly the situation that especially Sweden faces today: declining fiscal capacity combined with rising pressures on public job creation and/or income maintenance. Only Norway, with her oil revenues, has so far avoided the problem. Swedish policy makers and unionists can no longer avoid wage flexibility and major social benefit cuts.15 Still, Nordic social policy trends hardly point in an 'American' direction. True, wage differentials have grown and adjustments of the marginal tax rate and social entitlements have aimed to reduce negative work incentives and high absenteeism. Thus waiting days for sickness benefits have been reintroduced, replacement rates for sickness, parental leave, and unemployment benefits have been trimmed and, in Sweden, the second-tier pension programme (the second-tier earnings-related pensions) has been radically overhauled: pension contribution years have been extended and, more 14 After the Golden Age? importantly, benefits are now more closely related to contributions. This marks a move away from Sweden's tradition of allocating benefits as a matter of rights relatively independent of contributions. There is also a visibly stronger accent on 'workfare' (despite rising unemployment). Thus, work and training requirements have been strength- "~ěně^™^uMtanttailyŤr Swedish unemployment insurance, and Denmark introduced ajob guarantee for young workers with one year's unemployment. Also, to combat informal employment and negative work incentives, the marginal tax rates have been drastically reduced - as with the Reagan ~ r^oírnT^aTfičulSriSTso' fór Tngher-income earners. Finally, there is a drift towards decentralization and privatization of service delivery, particularly in SwedeňTTt would, however, be a mistake to see this as a neo-liberal strategy of marketization. All providers remain subject to centrally defined, stringent norms and the move appears much more motivated in terms of efficiency criteria and in terms of allowing services to vary more in accordance with differentiated client demands. Here we see an example of how the more heterogeneous 'postindustriaľ need structure compels social democracy to depart from its traditional universalism. Regardless, as Stephens argues in Chapter 2, the drift of these reforms is marginal adjustment, not a paradigmatic shift away from the basic principles of the welfare state. ....._____ Perhaps the mostjgmarkable trend in Scand^yian so_cial_policy is the shift of priorities in favour of the young and adults - groups that in the traditional full ejnployMenrsetting'were assumedJorequire only marginal welfare_stateJntervention. In a sense what is emerging is a new fife jcycle definition of social policy with the recognition^ that^contempörary^Jamily and employment transformation poses new risks and needs over the active, adulrphase~of people's lifeiaurses^TEsls reflected in the surge of adult rej^mn£l^IÍGÍes "and lifelong learning, in the schemes to facilitate geographical and job mobility, and nTThe joint parental leave provisions. It is also reflected in the attempts to secure the economic well-being of new family types, such as single-parent households. Scandinavia, indeed, is the only group of European countries in which social expenditure trends favour the young over the old. Here, then, is one manifestation of an emergmg 'social investxnent' ^approach. However, its longer-term viability is doubly uncertain. There is, / firstly, the conflict between the principle of universalist egalitarianism and the growing heterogeneity of the population structured There are indications that the more privileged social strata are exiting from the welfare-state, be it in terms of private (mainly individual) pension plans or services. Thus, failures to constantly upgrade welfare programmes may, in the long run, provoke an exodus of the elites which, in turn, will undermine the solidity of the welfare state foundations. The dilemma, of course, is that the fiscal capacity to effect such an upgrading does not exist. A second, and more serious, threat comes from the cpllapse__o£_full emnlovment. The limits to public employment growth have beenreached, After the Golden Age? 15 which means that any employmentjtratejyjuustjrelyjnainly on services in the private sector. This, in turn, poses the question of investménTifičěnoves aríd^äge^dlfferentials. A low-wage strategy of the American type might generate more jobs but would, in effect, seriously weaken the welfare state edifice. ^-~<"~~— . ___Regardl€SS,™presently-jffiry^hi:ghlunemployment rates-seem-t-o-contr-adiet-the validity of an active 'sociaL investment' approach. In other words, does the presently severe crisis of the..-Swecush~.mo.del_ affirm the neo-liberal position that large welfare states in whatever guise are to be dismantled? The answer will ultimately depend on one's diagnosis of the present crisis. Some, like the Swedish economist Assar Lindbeck (1994), diagnose it as a oM^wayjcausal. street: a crisis induced primarily by the welfare state's negative eľfcxts'blFwork;.savings, aľnd^mWstmeítThis analysis is," hówever7¥otTy"coht¥steär The main negative effect seems to derive from the very egalitarian wage structure and marginal taxes which give disincentives to work more hours and to invest in additional skills. Otherwise, evidence of any. i^í3IL-Wprk-cMrwatóy^ effect is_srarce (Atkinson and MogěnšenT 1993); moreover, Swedish long-term pjroductiyity performance has not been appreciably inferior to the European or OECD average (Korpi, 1992). In fact, the slowdown in growth, productivity and employment over the past 5-8 years may easily be attributed to transitory factors (especially the sudden haemorrhage of capital to the EEC in anticipation of the single market in 1992) or cyclical factors (the recent recession). As Sweden is now a full EEC member, investors' fears of being left out should subside. In the final analysis, the viability of the Swedish model will depend a lot on whether Sweden's once so celebrated consensus-building infrastructure is capable of overcoming its present fragmentation. The real issue may then have more to do with rebuilding institutions rather than dismantling the welfare state. The neo-liberal route A group of nations deliberately adopted deregulatory, market-driven strategies during the 1980s, notably the United States, Britain, and New Zealand; to a lesser degree, also Canada and Australia. Since Britain and New Zealand were once pioneer welfare states with strong full employment commitments, this exemplifies a radical regime shift. This cannot be said for the United States. The policy shift has been far from uniform. It accompanied the curtailment of protectionism in New Zealand and Australia; it meant a noticeable weakening of trade unions in the US and UK, while in contrast Australia's liberalization policies were actually implemented with trade union cooperation. In either case the gist of the policy was to manage economic decline and domestic unemployment with greater labour market and_wa^ej^cibjhty. This has involved social policies, primarily in terms of reducing the socialwage and legislated or de facto minimum wages. Except 16 After the Golden Age? for New Zealand's active programme dismantling, the most favoured approach combines a move towards greater selectivity, gradual erosion of benefits and/or coverage through failure to adjust social programmes in line with economic change, and 'workfare'. This style of more 'passive' aftftratirm wi113 as Myles argues in Chapter 5. have only marginal effects in the immediate term, but possibly far-reaching consequences in the longer run. The failure^adjust approach typifies American social policy. The minimum^wagF^oppoTto^only 38 per cent óf^veragTěarňmgs, and the value of social assistance benefits (aid to families with dependent children: AFDC) to 24 per cent by 1989 (Moffitt, 1990: 210). Similarly, the percentage of the unemployed receiving insurance benefits declined steadily from about 70 per cent in the mid 1970s to 33 per cent in 1989. Thus, with the principal exception of pensions, the already quite weak American social safety net was allowed to further erode. f K basic assumption in the American model is that themarket should supplementTEe^basic "public safety "netIn^ the postwar era this meant prhSHly" negotiated occupa^tionaTpJlansľ1! In this sense, however, welfare state decay is accompame3"Sy market decay: r^riy^tf-eoverage inJiealth and pensions has declined steadily during the 1980s, particularly among young and low-wagě~wórkers. The reasons are quite clear: on the one hand, employers seek to cut down on high (and growing) fixed labour costs; on the other hand, an increasing share of the labour force is employed in firms and sectors with traditionally low coverage. Yet, while conventional occupational plans decline, there is a noticeable growth in more individualized employee financed (and tax advantaged) programmes, such as the 401K plans. A common feature in the neo-liberal route is risjng_Jnequality and • poverty. During the 1980s, the lowest-decile earners lost ground, relative to "tlíernedian, by 11 per cent in the US, 14 per cent in the UK, 9 per cent in Canada, and 5 per cent in Australia (OECD, 1993). In contrast, most European countries exhibit essentially stable earnings differentials, and a very__modest rise in poverty. r^,^ <^The undej^yjn^ause is wagejiere|ulation> The low-wage' phenomenon in"thěse"countries is espe^lly^lcufr'among unskilled, non-unionized workers, and among young entering cohorts. However, as we have seen there are substantial national variations in the incidence of poverty and income polarization. Both Castles's (Chapter 4) and Myles's (Chapter 5) studies suggest that this can be explained by welfare state differences. Compared with the United States, Canada's unemployment coverage did not erode. Both Australian and Canadian welfare policies have become less universalistic and much more targeted. The methods of targeting, however, differ appreciably from classical means-testing: eligibility is decided on the basis of income or tax returns, and the principles of selectivity are meant to preclude the rich rather than to include only the demonstrably poor. Take-„~ ~~+~n o^-von,- v.;rrVi onri thp anrvrriíir.ív spurns miite effective in Drotecting After the Golden Age? 17 high-risk groups. Indeed, Australia actually raised benefits for especially vulnerable groups, such as families with many children. It is primarily this which accounts for Australia's substantially lower incidence of poverty. There is some evidence in favour of the positive employment effect of wage flexibilization.CEmployment growth in the 1980s has been two to three tunes higher inJhe^e_c^Sries than in the rest of the OECDpThis is also quite consistent with the ^Bjímoj_cos^jsjeasďJhesis, since much of the job growth appears related to loweTwages^ in services (see Baumöl, 1967; Blackburn et al, 1990: 72ff). The question, however, is whether the employment outcome is desirable. Low-end jobs may be unattractive, but they do provide a large pool of easily accessible first-entry jobs. This helps integrate youth, women, and immigrants into the labour market. On this count, the American scenario contrasts very favourably with the European. The burning issue, of course, is whether these jobs become dead-end traps; that is, whether the low-wage strategy fosters a new kind of chronically impoverished postindustrial proletariat. Research on this issue is still fairly rudimentary, but much suggests that mobility chances are substantial, conditional upon adequate skills (Esping-Andersen, 1993). Unskilled workers nave a high riskof^^rerna^mng^rajiped. Hence, an active social investment strategy seems to be paramount if we wish to avoid the emergence of a proletariat of the working poor. The low-wage strategy nurtures employment growth in low-productivity 'lousy jobs' where even full-time, all-year employment results m below-poverty income (Burtless, 1990). Hence, a low-wage labour market entails a double-jeopardy: it necessitates higher income maintenance transfers (such as social assistance) and, at the same time, produces poverty traps (since low wages create a disincentive to work). The wage flexibility scenario brings with it additional problematic consequences. There is a worrying erosion of the traditional fringe benefit packages of corporate welfare. Disappearing jobs tend to be in industries with developed welfare plans; a large part of the new jobs is concentrated in companies with little or no occupational benefits. Despite the fact that the United States spends almost 13 per cent of GDP on health care, the number of persons without adequate protection is very high (an estimated 30-40 million) and growing. In other words, an American style lean welfare state that assumes company provided supplements is likely to face a growing crisis of adequate social protection. ( The, gap in social protection is most acute Jnjrounger child fami|es)This is due to the low earnings capacity of single mHheršWmbmed with a real decline in social benefits. But also married women's ability to supplement the incomes of low-earner husbands is often impaired by the lack of affordable child care. In both types of cases, we see an alarming rise of child poverty in Canada, the UK and in the USA.16 The poverty problem associated with the low-wage' strategy is clearly concentrated among particularly vulnerable clienteles such as the unskilled and single-parent households. In the short run the risk can be reduced by 18 After the Golden Age? upholding the standards of income maintenance programmes, but if low wages remain the only option to welfare dependency, this clearly nurtures poverty traps. Political conflict in today's America is heavily flavoured by two opposed solutions to this: the right's strategy of essentially abolishing welfare; and the Clinton administration's active social investment strategy ^hfclrlavcws^subsMizEd^am potentially much less harmful were it systematically connected to an active training programme (Lynch, 1993).17 The labour reduction route The jobless growth,scenario-is especially acute in EEC Europe. If we go back to the late 1960s, overall employment ratios were basically identical in Scandinavia, North America, and contintental Europe (an average of 65 per cent of the working-age population). Since then, the employed-population rate has risen to 76 per cent for males and 60 per cent for women in the United States, and to 83 per cent for males and 76 per cent for women in Sweden; meanwhile, the EEC average has fallen to 57 per cent. Besides unemployment, the main difference lies in married women's and older males' activity rates.18 While the Scandinavians have managed the surplus of 'deindustrialized', largely unsMlled, masses with retraining and job creation, and the Americans with wage erosion, the continental European nations have opted to subsidize their exit, „especially through early retirement This has arguably produced an 'insider-outsider' divide, with a small, predominantly male, imsider'.) workforce enjoying high wages, expensive social rights, and strong job security, combined with a swelling population of 'outsiders' depending either on the male breadwinner's pay or on welfare state transfers. The roots of this strategy lie in the continental European welfare states' combination of highly (if not overly) developed social insurance (inordinately biased towards pensions) and underdeveloped social services.19 Social insurance means that entitlements are related to one's employment record, implying the necessity of a long unbroken career. The underlying assumption is that family members can depend on the full-time male breadwinner, and that wives are generally responsible for social care within the household. Hence, tax policies typically punish working wives, and the welfare state is extremely underdeveloped in terms of social services to fammes./Public child care coverage in Germany, the Netherlands, and s Italy is below 5 per cent. Similarly, the percentage of elderly living with ' their children is about 40 per cent in Italy and Spain, but under 10 per cent in Scandinavia and 15 per cent in the US (OĚCD, 1994a: Table 13).^)The continental European welfare state is thus essentially a familialistic transfer state. This helps explain its preference for early retirement (or disability pensions) as the principal policy for managing 'deindustrialization'. As a derived consequence, it also explains the high labour cost problem, After the Golden Age? 19 employment inflexibilities, and the catastrophic levels of long-term youth unemployment. The productivity gains that may come from labour reductions are easily outweighed by the associated costs. Social insurance finances are increasingly in deficit because contributions fall short of benefit payments. This jproblem is augmented by fragmented insurance funds: deficitsiäre^n^times-l^ (sučh^ásmfiefš^or general workers' insurance), while the funds for growing occupations tend to be financially healthy. The rising financial requirements that come from mass retirement and mass unemployment mean growing social contributions and thus fixed labour costs. This is especially true in Italy and France where labour supply reduction has been most intense. The indirect effect is that employers will prefer to regulate their labour needs via an adjustment of hours rather than assume extra workers; it also means that the marginal cost of part-time workers tends to be prohibitively high, thus additionally disfavouring female employment. And, high labour costs with rigidity in the context of mass unemployment mean that both employers and job seekers have a strong incentive to exit from formal employment relationships. This can be seen in the very large black economy, and in the rise of self-employment, neither of which of course broaden the welfare state's tax base. This kind of system has an inbuilt tendency to augment.labour market rigidities. If we consider that most families depend on the male earner's pay and social rights, and when we add to this the declining number of active years due to later entry and early exit, the result is that the typical worker can ill afford any risks or employment breaks across his active career. The consequence is that voters and trade unions will defend the existing rights of the 'insiders' as forcefully as possible. There is an implicit conspiracy to safeguard the prime-age male worker even when this harms his wife's, sons' and daughters' employment prospects. The problem has obviously not gone unnoticed, but owing to trade union, employee and even employer resistance, major efforts at flexibiliz-ation are easily blocked or neutralized. Italy's liberalization of part-time employment in the 1980s has had ^virtually-no, .practical effect. Many countries have implemented temporary hiring provisions, but except for Spain and, to a lesser degree, France there has been no visible rise in temporary workers - and,in these two countries the reform has augmented the temporary worker share but not overall net job growth. Paradoxically, there is strong evidence that a growing workforce of flexible temporaries only enhances the rigidities and privileges of the insiders (Bentolila and Dolado, 1994). There are two equally plausible explanations for why flexibility fails: (1) it is still too early to fully see the effects of flexibilization; (2) employers may avoid shifting to new labour practices in the interest of maintaining harmonious industrial relations.21 A case in point is that, despite weakened worker dismissal provisions in Belgium, France and Germany, employers' lay-off behaviour has hardly changed (Blank, 1993: 166). 20 After the Golden Age? Most agree that these countries need to decrease labour market rigidities. The problem is twofold. The first aspect is that the welfare of individuals and families depends on precisely those elements that cause rigidities in the first place: job security, high wages and expensive social contributions. The second is that informal, atypical and often black market activity becomes the chief compensatory strategy for those seeking work. These have in common that they dodge burdensome social contributions and rigid dismissal regulations. It is symptomatic that self-employment is the only source of real job growth in economies like the Italian or Spanish. From this perspective, it seems clear that the transfer-induced labour reduction strategy must be drastically reversed. In fact, on this there is widespread agreement. There is now a uniform move to raise retirement age, lengthen contribution requirements, and also diminish the burden of mandated social contributions. One strategy is to encourage the growth of employer plans, and a certain trend in this direction is visible. Still, it is unlikely to be a panacea precisely because - as in North America - this does not solve employers' labour cost problems. Privatization, then, more likely implies individual insurance plans and, thus, very uneven coverage. More generally, to reduce rigidities it is clearly necessary to diminish families' dependence on the single male earner. The key, then, is to augment the supply of, and demand for, women workers. Thus, it is difficult to see how the continental European model can avoid breaking with its traditional familialist, income transfer bias. It is in fact on this issue that much of contemporary political conflict focuses: the left typically advocating a 'Scandinavian' social service expansion; the right (especially Christian democracy) proposing a 'welfare society' approach that would reinforce the family - for example by introducing a housewife salary - and local community voluntarism. Considering the fiscal strains of the present social insurance systems, neither strategy seems particularly viable.22 The emergence of new welfare statesf Are the nations of East-Central Europe, East Asia, or Latin America in the process of emulating the Western model, or are they following qualitatively new trajectories? If by 'new' we mean models that deviate markedly from existing welfare states, the answer is essentially no. Our survey suggests the makings of distinct trajectories that do not necessarily correspond to regional clusters. .—, ~~~X One, comprising East-Central Europe, Chile, and Argentina/follows broadly a -liberal strategy^ based on privatization of social insurance, a reduced public social safety net, a shift towards targeted means-tested assistance, and a free-market bias in labour market regulation. The market-driven strategy in Latin America must be seen against the backdrop of a highly status-segmented, quite clientelistic, and seriously underfunded social insurance, tradition After the Golden Age? 21 r K second group of countries, exemplified by Brazil and Costa Rica, has so~far~shunned'neo-liberalism and has in fact taken some steps towards strengthening their public social safety nets, in both cases adopting a fairly universalist-ic approach in terms of population coverage. -The"tfiifd,-East_Asian, group is paradoxically both globally. unique_áncLa_ hybrid"of existing welfare state characteristics. It shares with the continental European model an emphasis on familialism and an aversion to public sociál services. Its embryonic social insurance schemes tend to follow the European tradition of occupationally segmented plans, favouring in particular privileged groups such as the civil service, teachers or the military. In these countries, social security is far from comprehensive, nor does it aim to furnish income maintenance. By default more than design, the vacuum of sociaLprotection has spurred the rise of c^^^y^sp^üred-' occupational welfare, especially in Japan. Äs a consequence, a certain degree of 'Americanization' has evolved: the modesty. of public welfare rests on the assumption that primary sector male workers will be covered under private plans. When we evaluate the paths taken in these regions, we should first of all remember the stark contrast between the crisis-ridden economies of Latin America (and recently also East-Central Europe), and the amazingly dynamic economies of East Asia. Indeed, the general economic climate of trie former two regions in the 1980s was much more similar than most believe: declining per capita GDP, inflationary pressures, huge debt problems, soaring unemployment, and the urgency to reform highly protected monopolistic industries.23 Both regions embarked on more or less rigorous Uberal stabilization and restructuration strategies in the 1980s. A common trartjya. the ex-communist nations' transition is a first attempt to cushion the^shock ffierapy wíťFIšocíäl security. Initially, virtually all introduced generoü¥Tmemplöyme^ redundancies were countered with attrition and early retirement. The dramatic fjalLin revenues, coupled with unexpected levels of unemployment and income loss "treat wages have fallen by 20-35 per cent, and in the CIS by as much as 50 per cent), led in many countries to virtual collapse of the existing social security system and a uniform shift towards targeted means-testing. The region as a whole experienced a net job loss of 6 million (12 per cent of the labour force) between 1989 and 1993. Participation rates have declined while irregular unemployment and under-employment have risen (OECD, 1994b). Unemployment and poverty have been growing everywhere, but there is a clear difference between countries like Hungary and the Czech Republic on the one hand, and Poland and the CIS states on the other. As Burda (1993) and OECD (1994b) suggest, the former countries were more prone to negotiate the transition strategy, the social safety net remained stronger, and there were more active employment policies, particularly for youth and the unskilled.24 As Standing demonstrates in Chapter 8, the shock therapy - combined with social nnlicv — has often added to alreadv existing distortions. The 22 After the Golden Age? policy of taxing wage growth (as a means to stem inflation) gives the stronger firms incentives to shift to non-money wages. Those outside this sector find their living standards dramatically reduced. An often sharp deterioration of the minimum wage has eroded both earnings and most social benefits (the latter, pegged to the mini-mum wage, have eroded to the „pi^ntj^exeJhey_equaĽ2Q^ the labour market, the drift is from protected, full-time jobs towards marginal, often black economy jobs, and the results are unemployment (the effect of which is to exacerbate the tax problem), an across-the-board fall in real wages, and the emergence of 'Third World' poverty rates (at present, 40 per cent in Poland; reputedly 80 per cent in the Ukraine). In a nutshell, what has been_privatized are individual risks rather than If the means to confront them. The kck of functioning financial institutions makes private insurance difficulTtcrestábHsh-^Hence, with the crumbling of the public social security system and the obstacles to a private alternative, the structure of social protection that remains resembles increasingly the kind of poor reliéf."that the advanced nations successfully left behind. Á quite similar scenario is described for those Latin American countries which embarked on a neo-liberal restructuration strategy. Traditional social security in most Latin American nations can best be described as a patchwork of patronage insurance, typically favouring privileged workers. Hyper-inflation and tax avoidance have meant that these face serious fiscal problems, but also that more aggressive reform efforts are difficult to contemplate. For these, as well as for other reasons, the Chilean experiment with privatization holds considerable interest. Huber's study indicates that privatization has so far been a mixed blessing. Chile's shift to a private individual, retirement account system has necessitated huge public subsidies and, hence, the net effect is a de facto subsidization of private welfare. Also, operating costs appear to be prohibitively high. Genuine coverage under the programme appears also quite modest, perhaps because it is purely employee financed. The new private schemes are essentially inoperable for the large mass of low-paid, marginalized, or unemployed workers. In other words, privatization in Chile has largely meant a replication of many of the same faults that characterized public insurance. The principal advantage of the new system is that it is financially solvent, and that its huge savings help capital markets. It is on the labour market front that Chile's liberalization strategy appears more positive, at least over the longer run. The short-run effects, as we have indicated earlier, were devastating in terms of industrial closures, mass unemployment, and immiseration, a trend we also see in Argentina. Clearly, Latin America's over-protected and monopolistic industries were hardly viable and primary sector labour markets, too, were extremely rigid. I The Chilean shock therapy relied on the authoritarian regime's ability to cníšh the trade unions, and it may therefore not exemplify a viable strategy for other nations. In fact, the Argentine approach seems to favour accords with existing interest organizations. After the Golden Age? 23 As Pereira (1993: 60) suggests, the earlier the adjustment programme, the í wer the long-run cost. The indicators for Chile, at any rate, are indeed "osltive. Unemployment levels have fallen from a catastrophic 30 per cent • 1983 to 5 per cent today, and investments, GDP and wages have all grown healthily. But this should be considered against past erosion: per anita incöme~h^^Hen^6^ercenH^ 1982 Real incomes in 1988 were no higher than before Pinochet, but they were much more unequally distributed (Pereira, 1993: 37-9).25 The alternative response, exemplified by CostaJSicarand-Brazil, has been to strengthen social policy in a clearly universaustiG„direction,^ especially in health care (although, as Huber notes in "Chapter 6, universalism in Brazil is questionable owing to heavy political patronage). So far, these countries have experienced neither declining incomes, nor rising unemployment and poverty. But the long-run viability of this route is uncertain. It remained possible in Costa Rica until generous American aid dried up, and considering inflation (especially in Brazil), huge foreign debt, and stagnant or falling GDP, the future prospects for a Latin American style 'social democracy' seem in doubt. Turning finally to the peculiarly hybrid East Asian welfare regimes, the first thing one notes in a comparative framework is that social security development lags behind their economic achievement. One argument has been that Confucian family welfare remains an "effective functional equivalent to welfare statism. Critics hold that the survival of three-generation households is pretty much due to lack of any alternative. Be that as it may, the issue of welfare state construction is now, for several reasons, intensely debated. In Korea and Taiwan, post-democratic / nation-building efforts mean also a need to extend citizens'_jrights. Also, population "'ageing, urban mobility, and modernization are causing a growing crisis.of elderly care. Moreover, the low-wage-based industrial miracle of Korea and Taiwan is rapidly being exhausted, implying the need for sweeping industrial restructuration and, in its wake, the likely emergence of unemployment and a host of new welfare problems. In much more advanced Japan, the system of lifelong employment and, with it, corporate welfare guarantees is weakening. The equüibrium of the Japanese combination of rather modest public benefits, private supplements, and virtual employment security (for the male labour force at any rate) rests not only on familial care, but also on -thejob guarantee. So far, these fast growing economies have suffered labour shortages rather than unemployment, and this has helped minimize the income risks of the male breadwinner and has sustained families' caring capacity. But this is not likeTy to continue indefinitely. In response to the growing strains of the 'Confucian model', these countries have taken a series of cautious steps towards a more comprehensive social policy. However, exemplified by South Korea's reforms in the late 1980s, neither do they approach anything close to universal coverage, nor are benefit levels adequate to bring recipients much beyond subsistence level. Taiwan's recent national health 24 After the Golden Age? care reform (September 1994), initially intended as universal and obligatory, is voluntary and gaps in coverage are thus likely to remain. Policy makers' hesitation to commit themselves to a genuine income maintenance system is partly due to fears of unusually rapid population ageing in the coming decades. This is particularly the case in Japan where, indeed, the conservatives seek to reinvigorate Confucian familialism as a compensatory strategy. This closely parallels the Christian democratic policy in much of Europe, and for basically the same reasons it is unlikely to be effective. Women in Japan and South Korea, as in Germany and Italy, are having far fewer children (fertility rates are now far below replacement), and are increasingly entering the labour market. What is more, population ageing in Japan is, comparatively speaking, extremely skewed towards the very old, meaning those with particularly intense caring needs. The percentage of people aged 80-plus will triple by the year 2020 (OECD, 1994a: Table 15). --, ^Another concern is with the possibly negative impact on savings. The Asian tigers' economic miracle was premised on high savings rather than Keynesianism: families save for lack of adequate social security coverage. A genuine welfare state, it is feared, will undermine this incentive. Since, moreover, these economies are characterized by sustained growth and quite egahtarian income distributions, there is some legitimacy to the assumption that most households have the capacity to save - at least if they are urban and based on a primary sector breadwinner. Nonetheless, as Japan already demonstrates, the suppression of consumption is not possible forever. The East Asian countries place much greater emphasis on education than on income maintenance, in large part in anticipation of structural unemployment and (partial) 'deindustrialization'. They are therefore potential vanguards in terms of stressing a 'social investment strategy'. Conclusions: major trends and policy dilemmas In most countries what we see isl not radical change, but rather a 'frozen' welfare state landscape. Resistente l^change_js to Jbe expected: long-established policies become Institutionalized, and cultivate vested interests in their perpetuation; major interest groups define their interests in terms of how the welfare state works. Thus, social security systems that are backed by powerful interest aggregations are less amenable to radical reform and, when reform is undertaken, it tends to be negotiated and consensual. Continental Europe is the clearest case of impasse, while Australia and Scandinavia represent change via negotiation. At the other extreme, in Chile and the ex-communist nations, whole scale change occurred against the backdrop of the collapse or destruction of the existing organizational structure. In between these poles are countries, like the United States or Britain, in which a more gradual erosion occurred in tandem with weakened trade unionism. After the Golden Age? 25 ismi.oyer the past decade is one of the grjmary reasonsjfor Jhe_ difficulties -that now also beset the famed Swedish mo^eľ.Tťšloň"g-ständing capacity to reconcile ambitious and egahtarian welfare goals with full employment has seriously decayed. There Js_ajeemmgIy_universaL trade-off between equality and employ-___ ment. Its roots may lie primarily in the new global order, but our study Identifies significantly different national responses. Within the group of advanced welfare states, only a few have undertaken radical steps to roll back or deregulate the existing system. All, however, have sought to trim benefits at the margin or to introduce cautious measures of flexibilization. As we have seen, those following a more radical liberalization strategy do better in terms of employment but suffer a high cost in terms of inequality and poverty. In contrast, those resilient to change pay the price of high unemployment - continental Europe in particular. A similar perception of a trade-off between equality and efficiency has always dominated social policy debates. In the postwar era it was widely accepted that the Keynesian welfare state provided a positive-sum solution. Today, there are few that are optimistic with regard to a viable 'third waýj. Still, many of tnFcoTmTrieTv^n¥vé~šurveyed pursue strategies designed to mediate or soften the trade-off. One, represented by Australia and Canada, combines liberalization and a shift towards more selectivity and targeting wTtfTä concomitant rise in benefits to thosemost atrisk. Their approach To selectivity is broad rather than narrow, aiming to guarantee against abject poverty and "stärk" inequalities. Comparative income and poverty data suggest that the strategy is somewhat successful. These countries have enjoyed an employment performance that equals the American without alarming rates of impoverishment. Another strategy, evident in Scandinavia, consists in shifting welfare state resources from passive mcomeToalnfenance to employment and family promotion. The era of pjiM^mplmme^^oynh nas clearly ended and, instead, policy is directed to active labour market measures, such as training and mobility, and wage subsidies. Scandinavia appears now to have accj2ptedjfoa^ to build in guarantees' against ťh^ě^elng~Foncehtrated in any~particular stratum, or becoming permanent across people's life courses. In this regard, the Nordic welfare states may be said to spearhead a 'social investment' strategy. They -have clearly not escaped high unemployment, or the necessity for significant cuts in social benefit levels. Yet, their unemployment record must be gauged against the backdrop of record high activity rates and, contrary to continental Europe, very modest degrees of socialmarginalization, exclusion, and youth unemployment ~~ More generally, if a return to full employment will have to rely on greater earnings inequalities and a profusion of 'lousy' service jobs, active social investment policies should diminish the chance that certain groups become chronic losers. 'Lousy' jobs will constitute only a marginal welfare T)rnnlf»m ionrl motr otk»« Vwa Vwa-rna-fi^íalA if +l-io-ir nra -rvioi-olTr o+nr» n-on r\t* pootr 26 After the Golden Age? first-entry, jobs for school leavers or immigrant workers. They are a major problem if they become life cycle traps. We know that education and skills offer the best odds for people to move on to better jobs. Hence, a low-wage-based employment strategy can be reconciled with equality if there exist guarantees of mobility and improvement. Privatization is one ofthe most commonly^advocated^strategies in the current welfare state crisis. In fact, it is promoted for two distinct reasons: one, to diminish public spending burdens and encourage self-reliance; the other, to respond to the more differentiated and individualistic demands of 'postindustriaľ society. In practice, there have as yet been very few substantial privatization reforms and the case of Chile remains therefore quite unique. However, a process of 'creeping' privatization may be under way in many countries, mostly because of gradual erosion of benefit or service levels. If privatization entails a shift of welfare responsibilities to companies, it is-very unlikely to become a panacea since corporate plans similarly inhibit flexibility and incur heavy fixed labour costs. Indeed, they are being rolled "back in tandem with public programmes. In addition, such plans are hardly viable in a service-dominated employment structure where firms are smaller and the labour force less unionized. The alternative is defined contribution plans or individual insurance schemes (like the Chilean model, or the rapidly growing IRA or 40IK type plans in America). 11 Individual plans do have positive aspects. Besides encouraging savings, I they permit individuals to tailor their welfare package. However, if they are meant to substitute for, rather than merely supplement, public schemes, their capacity to furnish social security in any universal way is highly dubious. Besides, the growth of such schemes has everywhere been nourished by public subsidies^ such as favourable tax treatment. Parallel to privatization is a certain shift away from defined-benefit to I contribution-based entitlements) particularly in pensions. This means essen-\ ' tially t^rwelfäre~stateš"(oT^ômpanies) are withdrawing their commitment u~ to benefit adequacy - one of the major welfare reforms of fheT960s and 1970s. In the Swedish case, this is less likely to generate major inequalities owing to the high levels of income security guaranteed by the basic, universal 'people's pension'. But this is not the case in systems, such as the Chilean, where individual contribution-based plans are the sole source of income maintenance - short of means-tested public assistance. In many welfare states, income transfer programmes were perverted over the past decades, becoming an inducement not to work. In the continental European countries, this strategy has exacerbated rather than eased the underlying employment problem: adding to the burden of labour costs for the shrinking 'insider' labour force and thus raisiřiflhecost of entry for the 'outsi&rs^youth especially. It increases the family's dependence on the sole (usually male) breadwinner's job stability_and pay. It is, then, clear that one of the 'greatest challenges for the future welfare state is how to harmonÍ7R women's emr>1ovmpnt~ with familv formation After the Golden Age? 27 fettfomen demand employment and_greater^ economic mdependence; the family is morTlikeíy to be flexible, and lešslEkelýTo be poor, if it can rely S'' n two earners; and the ageing burden will be lessened if fertility rises. The ^Scandinavian experience demonstrates that these demands can be fc-harmonized with a comprehensive network of public services. However, ^Th fiscal strams-0f-contemporary-we :"-- expansion; high wage costs make it unlikely in the private sector. ľ'-/ To the extent that the trade-off between social security and J ^induced by global wage competition, there is jLjOL^jsm^^^omce^.,of -- Tiositive-sum outcomes "since th"e=mam to the advanced . ec0nomieTare likely to build more comprehensive social protection systems - in the foreseeable future. It would, indeed, be a sad irony if the West engaged in welfare state dismantling in its drive to remain cgmpetitřye4f, at the same time, the mam^omp^ition^ere_to_raiseJis^labour costs. 3 On a final note, we should not forget that the initial impetus behind the postwar welfare state went beyond the narrower social policy concerns. As a mechanism for social integration, the eradication of class differences, and nation-building, the advanced" welfare states have been hugely_successful. . párVof the welfare state crisis today may be simply a question of financial strain and rising unemployment. In part, it is clearly also related to less fángible needs for new modes of social integration, solidarity, and citizenship. The market may indeed be an efficient mechanism of allocation, but /. not of building solidarities. There is little doubt that these more intangible qualities constitute an important element in the embryonic welfare state evolution in the new industrial democracies of Asia, South America and Eastern Europe. The economic effects of the welfare state can certainly not be disregarded. Yet, we should not forget that the only ^credible rationale behind economic efficiency is that it will produce welfares The idea of social citizenship may therefore extend also into the twenty-first century? Notes 1 This argument, while prevalent in current debates, must be accepted with serious caution. To give an example, while the import share from the newly industrialized countries (NICs) has grown substantially, it remains the case that an estimated 80 per cent of total EC member state trade occurs within the EC. 2 A recent study of trade liberalization in Latin America suggests strong, positive effects in terms of productivity performance and growth (Edwards, 1994). 3 Poverty trends in the ex-communist states seem to correlate closely with the extent to which the new regimes apply radical 'shock therapies'. In a recent overview, Cornia (1994) presents figures for the 'ultra-poor', meaning households with income less than 25-35 per cent of the average wage. From 1989 to 1992, their share rose from 8 to 20 per cent in Poland; from 19 to 30 per cent in Romania; and from 3 to 27 per cent in Russia. In contrast, the rise was moderate in Hungary (from 3 to 6 per cent) and in the Czech Republic (from 1.5 to 7 per cent). Similarly, the Czech Republic and Hungary have experienced much lower unemployment rates (OECD, 1994b). 4 The literature on this topic is truly enormous. For a very recent comparative study, see Freeman (1993); for a general review of research, see Esping-Andersen (1994). Streeck (1992) 28 After the Golden Age? has recently argued that these very same conditions also facilitate economies' adaptation to new and more flexible production methods. 5 Two examples will suffice at this point. First, as Castles demonstrates in Chapter 4, the negotiated liberalization strategy that the Australian Labor government pursued with the unions scores more favourably in terms of both equality and growth than New Zealand's, which was pursued in conflict with existingJnterest-associations.-Secondly— decades-of-sonal security institutionalization cultivate vested interests. Thus, it is virtually impossible to amalgamate occupationally exclusive social insurance schemes. 6 An economy's productivity performance is thus vital. The earnings performance of many nations in the past decade suggests that such levels of growth may not be so easily attainable. In the United States, for example, real manufacturing earnings declined by an annual average of 0.2 per cent during the 1980s. In Europe, where labour shedding has been much more dramatic, productivity and thus wages have grown at higher rates (1.7 per cent in France, 0.9 per cent in Italy, and 2.4 per cent in Germany) (Mishel and Bernstein, 1993: Figure 9A). 7 Freeman (1993: 3) shows that the percentage aged 15-64 years working (adjusted for hours worked) was identical in Europe and the United States in 1973. By 1990, Europe's activity rate was about 12 per cent lower than the American. As Freeman concludes, Americans work the equivalent of one month per year more than the Europeans. -8\ In the United States, the typical company pays 11 per cent of wages to legislated social contributions, and another 12 per cent towards fringe welfare benefits (Blank, 1993: 167). Tins compares with the EC average of 24 per cent to the former and 5 per cent to the latter. In heavy social contribution nations, like Italy, the former approximates 47 per cent, the latter 2 per cent (recalculations from European Community 1993a, Table 21). 9 I owe this point to Richard Freeman who, with considerable justification, sees the prison population as the American equivalent to Europe's long-term unemployed (personal communication). 10 Own calculations of Luxembourg Income Study (LIS) data for the mid 1980s show that single-parent (almost all female-headed) households face extraordinarily high poverty risks. Using the standard 50 per cent (adjusted) of median income as the poverty measure, the percentage of these households in poverty is 60 per cent in the US, 57 per cent in Canada, 27 per cent in Germany, and 19 per cent in both France and Italy. In contrast, the Swedish rate is 4.5 per cent. The impact of divorce may also be economically catastrophic, at least for wives. Burkhauser et al. (1991) show a 24 per cent income decline for American wives one year after divorce, and a full 44 per cent drop for German wives. The husbands' income loss is relatively inconsequential: 6 per cent in the US and 7 per cent in Germany. 11 Hashimoto (1992: 38) shows that 65 per cent of the elderly in Japan live with their children (down from 77 per cent in 1970). Choi's (1992: 151) data for South Korea show even higher rates (76 per cent). He also shows that 44 per cent of the aged are economically entirely dependent on their children. According to the official South Korean poverty line definition, more than 20 per cent of the aged are poor; about half have financial difficulties; and more than half of those who actually receive a pension find it difficult to five on it. A major reason cited for poverty among the aged is that their children avoid providing or are unable to provide, support for their parents (1992: 151). 12 The accent on education is already visible. According to Goodman and Peng's data in Chapter 7, the proportion of middle school (junior high school) graduates that continue to secondary-level education (senior high school) is 96 per cent in Japan and around 90 per cent in Taiwan and South Korea. 13 This discussion has focused on the gender angle of the policy, but it should rightly be generalized to the population at large, and to older workers in particular. Thus, combined with active labour market policies of retraining, rehabilitation, and job reinsertion, the strategy has succeeded - so far - in maintaining high employment levels also among youth and aged workers. The activity rate of males aged 60-64 is 64 per cent compared with 54 per cent in the US, 32 per cent in Germany, 25 per cent in France, and only 15 per cent in the Netherlands. 14 In the aggregate, Swedish absenteeism rates are about double those in Germany or the After the Golden Age? 29 Vld aged 0-2 was a whopping 47.5 per cent. Critics argue that the system is too generous and thus encourages abuse. This is partly true, but ignores also the Swedes' tremendous effort to ove the disabled and hard-to-employ from passive income maintenance to active employment in the economy (Bjorklund and Freeman, 1994). 15 The high wage costs and taxes are widely believed to spur negative work incentives and ■h'dden employmenValthough-hard^ nd Mogensen, 1993). Still, it is indicative that self-employment has been the fastest growing form of Swedish job growth in the 1980s. 16 Based on own calculations of LIS data, child poverty in two-parent families doubled in the United States during the 1980s (from 12 to 22 per cent) and tripled in the UK (from 5 to 15 per cent). Canada's rise was more modest (from 11 to 14 per cent). The rise in poverty among single-parent households was even more dramatic, except in the UK which registered a decline. 17 The presence of a more active training policy may, however, in itself not suffice if not coupled to a strong institutional framework. As Soskice (1991) suggests, the lack of this in the UK means that only a tiny proportion of those leaving school at age 16 receive any apprenticeship training. 18 Again, starting similarly in the 1960s, the activity rate of males, aged 60-64, has dropped to 25 per cent in France, 31 per cent in Germany, and 15 per cent in the Netherlands. The comparable rate is 64 per cent in Sweden and 54 per cent in the US. Note, however, that female employment rates, especially among the younger cohorts, have begun to rise since the mid 1980s in Germany and the Netherlands where part-time jobs have become more common. 19 In Italy, 60 per cent of total social expenditure goes to the aged; in Germany, about 45 per cent. This contrasts with 30 per cent in Sweden (which is equally 'aged') and 40 per cent in the US (OECD, 1994a: Chart 1). 20 As in East Asia, there is a clear declining trend. The problem of aged care is doubly acute since the only real alternative to family care is (extremely costly) hospitalization. 21 Several studies give credence to the former interpretation, suggesting that the shift to temporary workers will accelerate throughout the 1990s (Standing, 1993). On the other hand, Buechtemann's (1993) analysis of the German experience suggests that employers use temporary contracts as a screening device but subsequently extend permanent contracts in most cases. 22 In Italy, voluntary associations have grown tremendously over the past decade, particularly in areas šuch as care for the elderly and disabled or drug addicted. It is, however, evident that this has been helped by the availability of a huge pool of non-employed youth and women. 23 State ownership has been widespread in Latin America, accounting for 40 per cent of industrial output (compared with 80-90 per cent in East Europe) (Przeworksi, 1991: 143). 24 The Czech case is interesting since it combines low unemployment rates with a level of employment loss (at 10 percent) equal to others. In part, this is explained by retirement, in part by job creation schemes (250,000 jobs were created in 1992). It also seems that more drastic employment losses have been avoided by the strategy of privatizing prior to rationalizing firms (OECD, 1994b). 25 It is also unclear how Uberal is this liberalization strategy. The encouragement of markets appears to require heavy public subsidies; in part, as we have seen with pensions; in part, subsidies to private enterprises which in the 1980s were around 4.3 per cent of GDP (Pereira, 1993: 37). References Atkinson, A.B. and Mogensen, G.V. (1993) Welfare and Work Incentives. Oxford: Clarendon Press. Atkinson, A.B. and Rein, M. (1993) Age, Work and Social Security. New York: St Martin's 30 After the Golden Age? Baumol, W. (1967) "The macro-economics of unbalanced growth', American Economic Review 57: 415-26. Bentolila, S. and Dolado, J. (1994) 'Spanish labour markets', Economic Policy, April: 55-99 Bjorklund, A. and Freeman, R. (1994) 'Generating equality and eliminating poverty, the Swedish way', National Bureau of Economic Research, Working Paper no. 4945. Blackburn, L., Bloom, D. and Freeman, R. (1990) 'The declining economic position of the less ___skilled American men', in G. Burtless (ed.), A Future of Lousy Jobs? Washington. VtQ- Brookings Institution, pp. 178-204. Blank, R. (1993) 'Does a larger safety net mean less economic flexibility?', in R. Freeman (ed.) Working under Different Rules. New York: Russell Sage. Blank, R. (ed.) (1994) Social Protection Versus Economic Flexibility. Chicago: University of Chicago Press. Buechtemann, C. (1993) Employment Security and Labor Market Behavior. Ithaca, NY: ILR Press. Burda, M. (1993) 'Unemployment, labor markets and structural change in Eastern Europe', Economic Policy, 16. Burkhauser, R., Duncan, G., Häuser, R. and Berntsen, R. (1991) 'Wife or Frau, women do worse', Demography, 28: 353-60. Burtless, G. (1990) A Future of Lousy Jobs? Washington, DC: Brookings Institution. Calmfors, L. and Driffill, J. (1988) 'Bargaining structure, corporatism, and macroeconomic performance', Economic Policy, 6: 14—61. Choi, S. (1992) 'Ageing and social welfare in South Korea', in D. Phillips (ed.), Ageing in East and South East Asia. London: Edward Arnold, pp. 148-66. Cornia, A. (1994) 'Poverty, food consumption and nutrition during the transition to the market economy in Eastern Europe', American Economic Review, May: 297-302. de Neubourg, C. (1995) 'Switching to the policy mode: incentives by the OECD jobs study to change our mindset'. Unpublished paper, Faculty of Economics, University of Maastricht. Edwards, S. (1994) 'Trade policies, exchange rates, and growth', National Bureau of Economic Research, Working Paper no. 4511. Esping-Andersen, G. (1993) Changing Classes. London: Sage. Esping-Andersen, G. (1994) 'Welfare states and the economy', in N. Smelser and R. Swedberg (eds), Handbook of Economic Sociology. Princeton: Princeton University Press. European Community (1993a) Social Protection in Europe. Bruxelles: EC. European Community (1993b) Green Paper on European Social Policy. Bruxelles: DG5 (communication by Mr Flynn). Freeman, R. 1993. 'How labor fares in different countries', in R. Freeman (ed.), Working under Different Rules. New York: Russell Sage. pp. 1-28. Freeman, R. and Katz, L. (eds) (1994) Differences and Changes in Wage Structure. Chicago: University of Chicago Press. Gottschalk, P. (1993) 'Changes in inequality of family income in seven industrialized countries', American Economic Review, 2: 136-42. Hashimoto, A. (1992) 'Ageing in Japan', in D. Phillips (ed.), Ageing in East and South East Asia. London: Edward Arnold, pp. 36-44. Jencks, C. and Peterson, P. (1991) The Urban Underclass. Washington, DC: Brookings Institution. Korpi, W. (1992) Halkar Sverige Efter. Stockholm: Carlssons. Lindbeck, A. (1994) Turning Sweden Around. Cambridge, MA: MIT Press. Lynch, L. (1993) 'Payoffs to alternative training strategies', in R. Freeman (ed.), Working under Different Rules. New York: Russell Sage. pp. 63-96. Marshall, T.H. (1950) Citizenship and Social Class. Oxford: Oxford University Press. Mishel, L. and Bernstein, J. (1993) The State of Working America, 1992-1993. Armon, NY: M.E. Sharpe. Moffitt, R. (1990) 'The distribution of earnings and the welfare state', in G. Burtless (ed.), A Future of Lousy Jobs? Washington, DC: Brookings Institution, pp. 201-35. OECD (1988) Ageing Populations. Paris: OECD. JLjier trie uuiuen Age; WtiCD (1993) Employment Outlook. Paris: OECD. ^ TD (1994a) New Orientations for Social Policy. Paris: OECD. 8^ r-n ri994b) Unemployment in Transition Countries: Transient or Persistent? Paris: OECD. |ÖECL> L /J993) 'Economic reforms and economic growth: efficiency and politics in Latin |jereira, ■ _^ ^ pereiraj j jy[ Maravall and A. Przeworski (eds), Economic Reforms in New f'-n^mocracies. Cambridge: Cambridge University Press. * ski A. (1991) Democracy and the Market. Cambridge: Cambridge University Press. r12*- q 'r^Qyjfe^v—pOVei^^n-the-European-GommunityT-Loľíáoii^Ma.cmĽlaii---------------- i^vT' D (1991) 'Wage determination: the changing role of institutions in advanced rfpŕindustrial countries', Oxford Review of Economic Policy no. 6: 36-61. IgfiP , jjjjg g (1993) 'Labor regulation in an era of fragmented flexibility', in C. Buechtemann tggSU* -TEmployment Security and Labor Market Behavior. Ithaca, NY: ILR Press, pp. 425- «feStreec ■v W. (1992) Social Institutions and Economic Perfoimance. London: Sage.