4 class conflict, corporatism and comparison: a japanese...

4 Class Conflict, Corporatism andComparison: A Japanese Enigma!

M. Shalev

The 'Japanese enigma' which concerns this chapter is this: ina society in which the labour movement has been almost continuously excluded from political power and where labourunions are weak, politicized and competitively divided, thepolitical economy nevertheless generates outcomes characteristicof Sweden, Norway and Austria, countries in which class conflictis managed by 'corporatist' arrangements which emerged againsta backdrop of political dominance by strong labour movements.The mix of outcomes quintessentially associated with modernsocial democracy includes longterm full employment and relatively equal distribution of income; and also, in the periodfollowing the economic shocks of the early 1970s, successfulattempts at stabilization on the basis of organized labour'sconsent to the sacrifice of immediate wage gains and its use of thestrike weapon, for the sake of preserving jobs and revivinginvestment and growth. Japan shares precisely these features,although, curiously, it lacks one other hallmark of the socialdemocratic model: a big welfare state which plays a major role inthe (re)distribution of income and the patterning of ordinarypeople's life chances and living standards.

The dissonance, from this comparative perspective, betweenthe characteristics of the state and state/society relations inJapan, and Japan's favourable employment and growth recordand relatively high level of economic consensus, has yet to beresolved in comparative studies of the political economy of theadvanced capitalist democracies. This essay will review thepuzzle, and argue that several earlier attempts to resolve it including notions of some kind of a 'Japanese-style' equivalent toWestern European corporatism - have been ill-conceived.

60

Class conflict, corporatism and comparison

Instead, I propose a perspective which integrates analysis oflabour markets and the state arena, and thereby renders theJapanese case comprehensible in terms which are also applicableto other nations, yet need not devalue the power of the analyticaltools developed by modern comparativists. The lessons suggestedby this exercise may be enlightening both to students of Japanand to scholars for whom Japan is merely a perplexing case whichit is necessary to accommodate in models of a far broaderresearch universe.

Comparative Political Economy

An important contemporary stream of the comparative study ofpolitical economy emerged in response to the puzzle of why thedemocratic capitalist societies of the West diverged so markedlyin their adjustment to the global economic 'shocks' of the 1970s.The concerns of this literature have not only been confined toissues of economic adjustment and stabilization, but also includea quest for the origins of political stability and social peace.Indeed, insistence on linking distributional conflict and consensuswith economic performance in a single causal argument is thevery foundation of the political-economic approach. One particularly influential model of the political underpinnings of socioeconomic stability has been the resurrection and adaptation (tomodern democratic conditions) of the concept of corporatism.However, this contemporary revival has been motivated by atleast two different intellectual agenda. For analysts of interestgroup politics, it originates in a longstanding unease with thepluralist model, which tended to portray the state as a neutralarbiter among competitive pressure groups (Schmitter, 1974).Yet as experience in Continental Europe and Scandinaviaseemed to suggest, such a view may obscure powerful interlocksand interdependencies between the state and interest organizations, and it fails to account for the continuing importance ofcorporate (encompassing, monopolistic and internally hierarchical) forms of 'associability' in many modern liberal democracies.

Alongside this concern to revise the political analysis of state/society linkages in the Western nations, there also emerged a

61

Michael

Text Box

Shalev, Michael. 1990. “Class Conflict, Corporatism and Comparison: A Japanese Enigma.” Pp. 60-93 in Japanese Models of Conflict Resolution, edited by S. N. Eisenstadt and E. Ben-Ari. London: Kegan Paul International.

Shalev

fascination with the corporatist model as a form of conflictresolution and, in particular, as a metaphor for a distinctive formof political management of class conflict (Crouch, 1977; Panitch,1981). Since about the mid-1970s, many analysts, with their eyesquite clearly fixed on the small 'successful' states of WesternEurope and Scandinavia, have posited a close relationshipbetween the class representation of labour, political bargainingbetween state and associational elites, peaceful industrial relations, and effective management of the macro-economy. In someversions, it was pointed out that the strong labour movements ofNorway, Sweden and Austria had constructed a 'virtuous circle'(Castles, 1978) based on longterm cooperation between unifiedand powerful peak union organizations and Social Democraticgovernments, in the context of which workers were willing toexercise self-restraint in the labour market and throw theirelectoral weight behind the left in return for public policiessustaining the full-employment welfare state (Korpi and Shalev,1980). In other versions, it was suggested that corporate forms oflabour representation and quiescent industrial relations prevailnot only under Social Democratic regimes but also in theconsociational polities of Switzerland and the Netherlands.Despite labour's relative political and organizational weakness inthese societies, its representatives are integrated into policymaking networks, and a high degree of political consensus isevident on economic issues (Katzenstein, 1985).

In cross-national mappings of the outcomes which have been ofcentral interest to the new political economists, the case for thecorporatist paradigm has been empirically quite striking (forexample, Goldthorpe (ed.), 1984; Lehmbruch and Schmitter,1982). The five European countries named above as 'corporatist'stand out in their comparatively low levels of labour disputes,successful maintenance of full employment, big welfare states,and relative success in avoiding economic 'misery' in the form ofstagflation. The fit between outcomes and expectations is,however, rather less impressive for the consociational cases thanfor the Social Democratic ones. Switzerland's welfare state isgenerally evaluated as the smallest in the West, while by the1980s the Dutch commitment to full employment had proved tobe painfully fragile. While this slippage might be regarded asmerely a vindication of the distinction between the two subtypesof corporatism, the empirical credentials of the theory are much

62


more seriously called into question by a potentially crucial case,which in many ways exemplifies the favourable outcomesattributed to corporatism, while diverging quite markedly fromits political and institutional traits. This case is, of course, Japan,in which a weakly mobilized and internally divided working classhas been largely excluded from both political power-holding andparticipation in policy networks - and yet unemployment is low,economic growth is the highest in the GECD bloc (albeit farbelow the spectacular levels of the 1960s) and, since the late1970s, both inflation and overt industrial conflict have been wellnigh eradicated.

The Treatment of Japan in Empirical Studies

The notion of corporatism, as it is by now widely known, hasbeen conceptualized in diverse ways, with correspondinglyeclectic implications for empirical measurement - even when thefocus is confined, as it is here, to the implications of corporatismfor labour market phenomena. In some versions, corporatism isseen as virtually a corollary of the 'social democratic model'(Shalev, 1983a). A recent exemplar is Czada's (1987) study, inwhich corporatism is treated as the pinnacle of a hierarchy ofattributes encompassing almost all of the distinctively SocialDemocratic features of working class mobilization (broadly-basedand highly centralized trade unions as well as a dominant party ofthe left). In an earlier contribution by Cameron (1984), on theother hand, it was argued that Social Democracy ought to beviewed as no more than a special case of labour movementsenjoying a high degree of 'organizational power' (in effect,corporatism by another name). But the contrast between thesetwo studies at the theoretical level is invisible in their empiricalwork. Both Czada and Cameron generated almost identicalscaling of the 16-18 GECD countries of Europe, North Americaand the Pacific, which are the common research universe of thestudies under review.i Not surprisingly, they both classifiedJapan as minimally corporatist.

In a seminal although unpublished attempt at quantifyingcorporatism, Crouch (1980) anticipated Cameron's approach of

63

Shalev

evaluating corporatism in terms of the institutions of tradeunionism and collective bargaining, rather than the politicalpower of the left. Again, Japan appears at the negative pole ofthe continuum. Crouch, however, also pointed out that labourrepresentation and cooperative bargaining - among the hallmarksof corporatist industrial relations - are in some settings to befound at the enterprise level as well as, or instead of, in thenational political arena. In an application of the Crouch scale byBruno and Sachs (1985) which incorporated this additional factor,Japan was shifted somewhat away from the non-corporatistextreme on account of the regime of enterprise-level relationsprevailing in its big private firms.

A final cluster of comparative studies sets out from ratherdifferent premises about the nature of corporatism, proposingcriteria based on processes and outcomes of relations betweenlabour capital and the state, rather than either the structure ofworking-class representation or the balance of class politicalpower. For Manfred Schmidt (1982), the management of classconflict may be described as corporatist in those contexts whereorganized labour exercises restraint as a consequence not ofcoercion but rather out of commitment to a 'social partnership'ideology. And while this kind of outlook is common in countrieswith highly developed tripartite bargaining among union,employer and state elites, Schmidt argued that it could also beobserved in the 'paternalist' societies of Switzerland and Japan.The economist Ezio Tarantelli (1986) also came to the conclusionthat Japan should be evaluated as a paradigm case of neocorporatism. He proposed that the Japanese system of industrialrelations not only exhibited a high degree of harmony betweenlabour and management, but in several respects was closer to theScandinavian and Austrian corporatist model than was generallybelieved. Tarantelli argued (citing Shimada, 1983) that theannual 'spring wage offensive' functions in effect as a system ofcentralized wage bargaining, complemented in the 1970s by theco-optation of organized labour into a tripartite, if informal,system of national policy-making. He thereby stood on its headPempel and Tsunekawa's (1979) widely cited characterization ofJapan as a case of corporatism without labour.

It is clear from this brief survey that the Japanese case exposesmore than any other the contingent character of the notion ofcorporatism. We may go even further than this and suggest that

64


Japan is the veritable achilles heel of empirical operationalizations of corporatism. For, in spite of the substantial conceptualand methodological differences among the studies reviewed here,they approach a quite pronounced consensus on the ranking ofeveryone of the core OECD countries except Japan. The pointmay be demonstrated statistically by a factor analysis of sevendifferent measures (those of the six authors cited here, plus astudy by Lehner (1987). Most of the variance between scores isaccounted for by a single factor on which everyone of thecorporatism scales 'loads' strongly. However, if Japan is includedin the analysis (and only if it is included), then a secondsignificant factor is also present - one on which the loading ofeach author's scale can be perfectly predicted by its treatment ofthe Japanese case!

Japanese Puzzles

What substantive difference does it make whether Japan isranked as strongly or weakly corporatist in comparative studies ofpolitical-economic outcomes? That of course depends on theoutcome which is of interest and how it, in turn, is defined andmeasured. For instance, strike activity in Japan has experiencedquite dramatic shrinkage since 1975. But before then, theJapanese pattern of industrial conflict resembled the French andItalian profile of a high level of worker participation in broad butbrief disputes (Korpi and Shalev, 1980). From this perspective,Japan would not have made sense as anything but the antithesisof corporatism. Similarly, an argument commonly voiced in theseventies was that inflationary bubbles had been prevented by'social contract bargaining' in the corporatist democracies, andwere most likely to develop where labour was alienated fromparliamentary politics and accustomed to pursuing wage demandsby work stoppages. The Japanese case, circa 1974, provided itwas again viewed as non-corporatist, offered tangible supportngevidence for this hypothesis (as in Crouch, 1980).

Not only did Japan experience a surge of rising prices in theshort-term aftermath of the first oil shock; it also suffered a verysharp decline in its previous growth rate. In these terms, Japan in

65

Shalev

the mid-1970s appeared to belong to that unenviable group ofnations suffering the double-headed misfortune of stagflation;and here too, a non-corporatist Japan provided the best fit withobserved outcomes. This was empirically demonstrated in awidely read volume by Bruno and Sachs (1985; see especiallyp. 228). Here Japan appeared to vindicate the view that only realwage moderation could ward off the disastrous economic effectsof an external shock; but that in most instances, such moderationwould only be forthcoming under conditions of highly centralizedpay determination and a high degree of social consensus (that is,under corporatism). If on the other hand, a desirable stabilizationpolicy was defined as one which permitted the economy tocontinue to grow and prevented rising unemployment - thenJapanese developments, even in the early years of the crisis, hadto be described as a success story. In this instance, a convincingcross-national argument required that Japan appear in the guiseof a strongly corporatist nation - which indeed it did in severalrelevant studies (such as those of Tarantelli and Schmidt).

Thus, whether by luck or design, most researchers succeededin verifying the associations which they hypothesized, despitetheir confusion as to whether Japan was or was not corporatist,was or was not a case of successful economic stabilization. It istrue that scholars had to contend with a very fluid reality, inwhich Japan's economic position improved in the 1980s. Nevertheless, this is not the whole of the story. If Japan is (as weargued at the beginning of this chapter) an anomaly, then little isto be gained by disguising the problem, whether by more'appropriate' coding or by hiding Japan behind measures ofaverage effects for many countries. On the contrary, it isnecessary - and, moreover, can be extremely fruitful for those insearch of wider generalizations - to confront what are apparentlydeviant cases head on. The result, in at least some recentcomparative research, has been to recognize what was always anobvious logical possibility - namely, that identical outcomes indifferent countries may well have divergent causal antecedents. Agood example is Goran Therborn's study of the determinants ofthe contemporary divergence across OECD nations in theseverity of unemployment. In Therborn's (1986, pp. 23-4)summary, during the post-war period:

Full employment was institutionalized for two major, quite

66


different reasons. One was an assertion of working-classinterests. It owes its success to a politically dominant labourmovement [as in Sweden and Norway] .... The second reason... was a conservative concern with order and stability as beingof equal importance to capital accumulation. Full employmentin Japan and Switzerland has this background.... Austria fallssomewhere in between these two poles.

What is important here is not so much the validity ofTherborn's interpretation of the Japanese case (which may haveover-emphasized the non-economic motives of Japanese elites),but rather the fact that alternative causalities, firmly grounded inthe diversities of national context and history, are admitted to theterrain of legitimate explanation.

The other central issue which comparative political economyhas to confront in taking Japan seriously, is whether there is infact some theoretically persuasive notion of corporatism whichpossesses empirical plausibility in the Japanese case. If so,corporatist theory might have an important role to play inexposing the underpinnings of what must, from the perspective ofthe late 1980s, be regarded as truly an economic miracle. Werefer of course to Japan's exceptionally strong showing onvirtually every dimension of macroeconomic performance, including growth, inflation, investment, and trade. And clearly, it is notmerely coincidental that these achievements have been wonagainst the backdrop of self-restraint by labour - although this isa form of discipline with no obvious links either to strategicpolitical exchange from a position of working-class strength(Pizzorno, 1978) or the state's coercive use of its authority tosteer the economy towards mass unemployment (as in Kalecki,1943).

The comparative literature, we have discovered, contains twodifferent applications of corporatism to Japan. One of theserefers to class relations at the micro level (labour-managementcollaboration in the enterprise). The other is the more conventional idea of union elites undertaking to rein in worker militancyin return for negotiated policy concessions from the state. Anintriguing question is whether the Japanese case is 'functionallyequivalent' to the corporatist political economies in one or both

67

Shalev

of these senses. To address this issue requires some discussion ofthe peculiarities of the evolution of labour markets, industrialrelations and social policy in Japan, and how these are related tothe distinctively Japanese pattern of labour quiescence andcooperation in the enterprise. Attempts to label this pattern'corporatist' will be criticized as distortions of both the concept ofcorporatism and the dynamics of the Japanese cas~. I s~all thenturn to Japan's success in combatting wage mflation andproductivity decline. The critical question is wheth~r labou:'scooperative behaviour should be attributed to a t~rn in the m~d1970s towards corporatist participation of labour m. the n~~otiation and coordination of national economic and SOCIal policies (Iargue that it cannot). My conclusion will be that u~derlying theJapanese enigma is a constellation of forces with P?werfulhistorical and logical coherence, and that aspects of thl~ sameconstellation may also be discerned in some other countnes.

Is the Japanese Employment System 'Corporatist'?

The distinctive features of labour organization and industrialrelations in Japan are sufficiently well known to warrant only thebriefest rehearsal here. In Dore's authoritative account, thesefeatures - 'factory and company based trade union andbargaining structures, enterprise welfare and .security, greaterstability of employment . . . and a cooperative or corporateideology' - add up to a system of welfare corporatism (19:3,p. 370). More recently, Lincoln and ~alleberg h.ave also apphedthe concept of corporatism to the rmcro level m Japanese andother settings characterized by 'organizational structures andmanagement practices aimed at fostering corporate ,loyalty,commitment, and dependence on the part of labour (1985,p. 740). The task before us in this section is to evaluate whethersuch a conceptualization makes sense. . .

Japanese trade unionism did not emerge on a significant s~a~euntil after World War II, although it retained a core charactensticof pre-war and wartime precedents, namely organization ~n afirm-by-firm basis (outside the public services~. Th~ potentIallyclass-based character of unionism is compromised m Japan not

68


only by its decentralized structure, but equally by low and unevencoverage and keen internal divisions. With only about a third ofthe wage-earners organized (even less in the eighties) unionpenetration is modest by international standards. Membership isalso unusually concentrated in the public sector (accounting forabout a third of the total) and in large private enterprises. WhileJapan's smaller (under 100 employees) privately-owned firmsgenerate more than half of GNP, they have never accounted formore than one tenth of the organized work-force. Moreover,while it is customary for big firms to employ large numbers of'temporary' workers and indirectly to engage the services (bothoff-site and on) of workers who are on the payroll ofsubcontractors, as a rule only permanently employed staff areadmitted to unions.

Although enterprise unions are affiliated to national associations, the latter are divided into federations with competingpol!tical loyalties. There is a component of collective bargainingWhICh takes place on a supra-enterprise and multi-confederalbasis. Nevertheless, the main instrument for this - the annual'spring wage offensive' (shunto) - sets only a basic floor on whichenterprise unions then build. (Another major form of multi-firmnegotiation has emerged in association with the InternationalTrade Union Secretariats, but it has paradoxically further dividedthe workers by involving only 'non-political' coordinating bodies,which negotiate with employers within specific industries.) Whilethe 'offensive' and collective bargaining generally may be accompanied by quite dramatic forms of worker hostility towardsmanagement, stoppages of work have a primarily demonstrativecast. Most involve large numbers of workers participating in briefpre-scheduled actions in support of shunto demands. As a result,until the contemporary withering away of open disputation,Japan's comparative strike volume ranked a little below themedian for the main OEeD countries.

A historical perspective is essential for comprehending themeaning of this pattern of collective action by workers for theJapanese enigma of presumptively corporatist labour market andmacroeconomic outcomes. Strong elements of continuity areevident in the strategic action of Japanese elites in politics, thebureaucracy and big business, who have sought since thebeg~nnings of industrialization to protect the authority of state,family and employer against the bacillus of working-class

69

Shalev

mobilization and solidarity (Gordon, 1985; Pempel andTsunekawa, 1979). In the pre-war period, the state intervened toprevent coercively the formatio~ of uni.ons a~d .the emergence ofworker suffrage and labour parties, while assisting and encour,aging employers to develop labour po~icies with ~ strong enterpnselevel orientation. Employers expenmented with personnel practices which subsequently became associated with the post-warsystem of labour management - including division of the ~orkforce into 'lifetime' and temporary employees, and the mternalization (to the enterprise) of the supply of skilled labour. Agood deal of modern scholarship (see ~lso Do~e, 1973; Jacoby,1979; Taira, 1970) is agreed that such mnovatlOns need not beunderstood as a form of cultural particularism, but rather aseconomically rational employer responses to specific problems ofrecruiting and controlling labour. ,

Japan's military defeat in World War II and the radl~~l reformsinitially instituted by the American occupation authorities at firstconstituted a sharp break with the prior political supremacy of aconservative ruling coalition and the economic domination of thegiant zaibatsu combines. In the space of only a few.y~ars, s~pportfor left-wing parties mushroomed (they even p~rtl~lp~ted in thegovernment for a brief period in 1947-8) and unionization spreadto more than half of the work-force. Labour chose to organize forcollective action in the first instance on an enterprise basis. Thischoice accorded with prior experience and also reflected anassumption that only by reviving their enterprises' economiccapacity could workers hope to transcend their desp~ra~eeconomic plight. However, after only a few years of ent~uslaStlC'democratization' the United States began to throw Its considerable weight'behind employer and state ,offensives, w~ichlargely restored the traditional political

3exclu~lOn and organ.lza

tional weakness of the working class. Inspired by a revisedconception of global US interests, the Occupation mov~~ sharplyaway from its initial objective of crushing Japan's mlht~ry ~ndeconomic might, adopting in its stead a Cold War l~s~lreddetermination to create an economically robust and politicallysympathetic Japan able to 'support herself ~nd assum~ t,he :ole ofa protective barrier against possible unrest m East ASia (cited by

Sumiya, 1974, p. 67).A variety of interventions during the late. 1940s h~d the e~fect

of deterring labour militancy and of deepenmg left-nght schisms

70


in the labour movement. The Occupation effectively banned orbroke a number of pivotal strikes, 'red purges' were carried outin the radical unions, and the legal framework was in significantways de-liberalized. Particularly important was the offensiveagainst labour during 1949, when state policies (mass lay-offs andthe rescinding of union rights in the public sector, coupled withslashes in subsidies and unemployment pay) created conditions inwhich the major private employers were expected to reasserttheir authority. Management indeed succeeded in winning aseries of showdowns with militant enterprise unions. At the sametime, the big firms accepted union demands for job security andlinking wages to worker need (age). In a context of economicrevival, these could be awarded at relatively low cost, whileyielding pronounced benefits - the loyalty of older workers (whoif necessary were willing to support the formation of docile'second unions') and acceptance of the highly constrained termson which the large enterprises were willing to accommodate tocollective bargaining: no negotiations with 'outsiders', and nounionization of temporary or subcontracting workers."The emergence of rapid economic growth and the transition fromlabour surplus to labour shortage might have shifted the balanceof power yet again in favour of labour and had the effect oflowering workers' attachment to the enterprise and stimulating amore class-oriented form of organization. That this did not comeabout was by now (circa the mid 1950s onwards) the result mainlyof employer initiative, rather than the role of the state. Firstly,extensive 'rationalization' (capital-intensification) was carried outwith an eye to reinforcing managerial control in the workplace.Secondly, the unions' attempt at multi-employer wage bargaining- in the form of shunto - was accepted as a means of partiallymitigating competition among the big firms, provided that morewage uniformity could be achieved without squeezing profits. 5

Thirdly, the big enterprises have structured their 'internal' labourmarkets in ways that heighten the dependence and loyalty ofthose workers who enjoy the privileges of job security and unionmembership. The relevant incentives include formidable barriersto upward mobility across firms," the deferment of wagesthrough such devices as seniority-based pay and large retirementgrants, and the benefits provided by company welfare servicesand profit-sharing bonuses.

Under these circumstances, it is understandable that private

71

Shalev

sector trade unions have taken up a very limited adversarial roleand have generally been cooperative towards managerial objectives. Indications of union co-optation by the firm abound including company-provided union facilities and salaries, occupancy of union positions by supervisory employees, and thecareer line which runs from union office into management(Hanami, 1979). Nevertheless, collaborative unionism is not theresult of co-optation per se but rather of the terms of the'settlement' which ended labour's post-war struggles, and thecompelling economic logic institutionalized in the Japaneseemployment system. It is clearly materially worthwhile for thepermanent/unionized work-force to join efforts to improve thecompetitive position of the enterprise - especially since theseefforts are supported by ramified forms of social control overwould-be free-riders. The benefits for employers have includedlow turnover costs, a high probability of enjoying the fruits ofinvestments in worker training, little or no disruption toproduction or challenge to managerial authority by unions, andthe ability to mitigate the fixed costs of a permanent work-forceby flexible job assignments and the elastic utilization oftemporary and subcontracted labour.

Just as Japan's dual economy organically links the bigoligopolistic firms at the core with their suppliers and subcontractors on the periphery, so too do the workers of the 'primary' and'secondary' segments of the dual labour market perfo~mcomplementary roles - the one exhibiting flexibility in adaptingjob content and level of remuneration to the needs of t~e. firm,and the other accepting insecure employment as a condition ofexistence. That the workers of the secondary labour market haveaccommodated to their fate may be explained, firstly, by the notinconsiderable rewards that could nevertheless be obtained and,secondly, by the powerful social forces which legitimate therationing of 'good jobs'.7 As in other dual labour markets. thelower tier of undesirable jobs is peopled mainly by SOCIallymarginal (that is, relatively powerless) groups. In Japan, theseare not foreign guest-workers but principally women, complemented by older men in second careers, certain ethnic minorities,and rural seasonal workers. Social attitudes regarding genderroles, together with admission and streaming policies in theeducational system and outright discrimination by employers,have combined even more effectively than in the West to support

72


labour market inequalities (Business Week, March 4, 1985;Steven, 1982).

Is Enterprise Welfare Japan's Substitute Welfare State?

Before returning to the larger question of whether the specificities of workplace labour relations in the Japanese context mayusefully be subsumed under the corporatist label, we shouldrecall that in Dore's conceptualization the label of choice waswelfare corporatist. The implication is that just as the corporatistmode of managing class conflict has been transferred in Japan tothe micro level, so too have the functions of the welfare state.And if it is indeed the case that 'welfare corporatism' is a form ofquasi social democracy transplanted to a different institutionaldomain, then the apparent paradox of full employment coexisting with meagre social expenditure and decentralizedindustrial relations would dissolve. The explication of theunderdevelopment of the Japanese welfare state and thedevelopment of comprehensive company welfare in large enterprises is therefore essential for reaching a verdict on theapplicability of the micro-corporatist model.

The size and significance of the state in the Japanese politicaleconomy presents a marked contrast to familiar patterns of themainstream of Western experience, in that the state combines arole of great importance in steering and assisting the economicactivity of (big) business with exceptionally modest interventionin the distribution of national income. Whether considered interms of public provision of social services like health andeducation, or of public transfer payments under minimum incomeand social insurance schemes, Japan's welfare state has alwaysstood out as an extreme 'laggard' (for example, Wilensky, 1975).It would, however, be mistaken to interpret this as no more thana particularistic artifact - either of Japan's cultural tradition or ofthe predilection of its ruling party for small government andbalanced budgets. Instead, labour market dualism must be seenas a key to understanding Japan's conservative social policy. Anelaborate system of employment-linked social protection simplyexcludes the employees of small enterprises, while at the otherpole of the labour market, social security 'has been tied to a large

73

Shalev

extent to the employer's policy of personnel management'(Takahashi, 1974, pp. 482-3). .

Health and pension provisions are the predommant components of public social expenditure in Japan. In both cases, asMaruo (1986, p. 68) comments, 'all Japanese people are cove~e~.That coverage is unequal, though, because of the dualisticstructure of the public and private sectors, and the separation ofbig business, small business, and the self-employed.' Amongwage-earners in the private sector, one third of the labour-forceis employed by enterprises with fewer than ten employees, whereunions and collective bargaining are non-existent and, it may besafely assumed, employer-initiated welfare schemes are rare. It isthese workers who appear to be the principal clientele of thegovernment's scanty health insurance scheme and who are mostreliant on minimum public pensions. On the other hand, firmswith at least a thousand employees are permitted by statute toprovide their own health and pension benefits, under conditionswhich allow for superior benefits at relatively little additional costto employers (Lee, 1987; Pempel, 1982 and 1989). Retirementallowances and other fully autonomous forms of enterprise welfareare even more closely associated with the scale of the enterprise.Not surprisingly then, as Hall (1988; p. 13) has shown, thetotal welfare expenses of employers rise in very similar proportions to wages and the rate of unionization along the gradientof increasing enterprise size. The dose association betweeneconomic dualism and employer-furnished social benefits is alsoevident from the far higher probability of finding specific programmes in very large firms (5,000 or more employees) incomparison with industry as a whole - in 1973, 74 per cent versus8 per cent for medical clinics, and 94 per cent versus 19 per centfor housing loans (Katsumi, 1974, p. 64).

Japan's public welfare programmes experienced unprecedentedexpansion in the 1970s. On the face of it, this might be taken asevidence of erosion of the sharp split between state andenterprise welfare. Yet, if anything, the new reforms actuallyserved to crystallize the dualism of social policy. While the muchvaunted breakthrough to a 'Japanese-style welfare state' washighly consequential for social security expenditure (which roseten times in real terms in the course of the sixties and seventies),this growth was closely tied to a rapid aging of the population andrising unemployment (Maruo, 1986). The responsiveness of the

74


state to increased need may in turn by plausibly interpreted asthe regime's reaction to a dangerous erosion of support for theLiberal Democratic Party (LOP) among the mass public. Subsequent deep cuts in social spending in the 1980s, after the LOPhad more than recovered its electoral losses, reveal that a goodpart of the political commitment to more generous public welfarewas no more durable than the threat to the legitimacy of theruling party (Pempel, 1989). From a more qualitative perspective, although Japan did finally converge on the Western modelin terms of the range of publicly enacted social programmesoffered, to its citizens, it made very little movement towards theSocial Democratic norms of universality, 'decommodification'and generosity so antithetical to its traditional pattern of socialpolicy (cf., Esping-Andersen, 1985). Indeed, enlargement ofstatutory health and pension programmes took the form of anexplicitly two-tiered system of minimal public provision alongside generous company-run and company-tied schemes in thelarger private enterprises and their equivalents for publicemployees.

While the resilience of 'enterprise welfare' in big Japanesefirms thus cannot be doubted, the question remains of whetherprovision for employment-linked social protection in Japan isreally very different from other countries. One obvious differencelies in the widely remarked diffuseness in Japan of the boundarybetween work and private life, which receives concrete expression in the pervasive role played by paternalistic large employers,from arranging marriages to helping to place their employees'children in advantageous schools (Hanami, 1979). However,international comparisons of monetized forms of assistancesuggest that Japan's distinctiveness is to be found not in theoverall magnitude of company-level 'welfare effort' (Hall, 1988;Taira and Levine, 1985) but rather in its implications - for labourrelations on the one hand and for capital formation on the other(cf., Pempel, 1982). The only big ticket items are housingfacilities or subsidies, and retirement schemes. Given Japan'sextremely tight housing market, the former is a compellingincentive for employees to demonstrate loyalty and flexibility(including acceptance of relocations). Contributions to retirementbenefits are a form of deferred pay which, in the case of lumpsum payments, have the added attraction to the firm ofconstituting a fully tax-exempt source of investment funds. Even

75

Shalev

when combined with pensions (as in the largest enterprises), thebenefit payable upon a worker's typically premature retirement islow enough neither to burden the firm nor to free older workersfrom the necessity of returning to work - quite possibly for thesame employer, but at reduced pay. Indeed, the low level ofsocial security for the Japanese elderly has been a powerfulincentive for high rates of savings and participation in insuranceschemes, which have in turn been a major factor in Japan's highrate of private investment (Johnson, this volume).

The 'residual' character of Japan's (public) welfare accordswith expectations deriving from social democratic theory. In apolity in which unions are relatively limited in scope and havedivided political affiliations, and where the labour movement hasbeen excluded from government (not only formally but for themost part also effectively), we would not expect either left-winginitiative or pre-emptive action from the right to drive theenhancement of citizens' rights to social protection (Korpi, 1980).Given also the enterprise context of trade unionism and collectivebargaining, and the unions' concentration in firms with sufficiently wide profit margins to afford company welfare schemes, it ishardly surprising that the organized workers who benefit from thepublic/private split in social policy are generally opposed toextended public provision, while the unorganized labour-forcewhich is most reliant on the welfare state lacks the collective

voice to demand such extension.This makes clear the striking congruence between the Japanese

pattern of modest public welfare co-existing with highly differentiated occupational welfare, and the logic of the dual labourmarket. The exposure to market forces, which is the hallmark ofa residual welfare state, is a powerful incentive to the workersand small business-people of the 'peripheral' sector to performthe roles assigned to them in the economy - as 'flexible' workersand suppliers, and savers. At the other extreme, the doublestranded tie _ of dependency and loyalty - which binds'permanent' workers to their employers in the sheltered privatesector labour market (i.e. the large corporations) is clearlyreinforced by, and even conditional upon, the critical role playedby company housing, health, and retirement schemes. In a worldof high and portable universal pensions, free public health careand cheap and plentiful housing, Japan's industrial giants wouldbe hard-pressed to maintain their paternalist authority, the

76


precedence among workers of ent .ness, and the selflessness of work e~pn~e ov~r class consciousand profits. ers onentation to productivity

So Are Japanese Labour Relations 'Welfare Corporatist'?

In his influential essay 'Th EGoldthorpe (1984) portrays e n~ of Convergence', John

d corporatism and dualis dian to a large extent incompatible d I . .m as istinctIn Goldthorpe's view the most m? e s of ,PolItIcal economy.these two models concern the o~ons~lcu.OUS dlff~~ences betweenpower position of labour U d gamzatI~n, political status, and

. . n er corporatism unio .a national (class-wide) basi . ' ns orgamze on. aSlS, trading off rest . t' ,econormc demands in return for ", , ram m theirfrom the making of ublic 1" participation m and benefitsexplicitly excludes farge ~~t~~~:' I~ c~ntrast, the dualist modelsphere of unioniz ti ,. ,o.t. e work-force from thecollective action ~yIO~rag:~'mddustnakl citizenship', while confining

ize wor ers to purely ti 1economic objectives. With its labou sec,IOna andnently beneath the pinnacles of rmovement parties perma-power, and its exclusionary forms p~r Ia~e?tary and bureaucraticin the market arena, Japan must ~e umomsm ~nd ~areer mobilityquintessential instance of the d regarde,d, m this context as aquestion is whether the coexiste ua l~t polItIcal economy, Thelabour weakness and port' n

lce wlth~n the large enterprise of

I tea exclusion with b th 1 bmanagement cooperation and alb 0 a ourtection justifies the characterizati~neo~ Jorate sys~em Of, social ~roas welfare corporatist. apanese industrial relations

In my view the dualist model is b fprism through which to interpret th YJ ar the most appropraiteJapan has constituted a highly ff / ~panese case, Dualism inproduction and controlling lab e e~ ive ramework, for organizingrapid economic growth ~ur m accor~ance WIth the goals ofdominance of big business ::d t:er~etuatIOn of t~e twinnedaddition to material benefit ' e lL~eral Democ:~tIc Party, Insharing of the fruits of p s, I~C udmg a surprisingly diffuse

brospenty ualism has b . ff .

ecause the blue-collar workin ' een e ectivebetween the less privileged and gthclass was successfully splitfortunate enough to work f th ose securely employed men

or e government and other big

77

Shalev

employers. It follows that overall outcomes like full employment and industrial peace result from the aggregation of twodistinct logics - a condition of linked subordination andprotection in the primary labour market, and quite differentbut equally powerful economic and ideological constraints onlabour in the secondary market. In parallel, social policy has notbeen managed by any wholesale delegation of the functions ofthe welfare state to the firm. Instead, we again observe a splitsystem, of state-subsidized company welfare in the prim~rylabour market co-existing with a public safety net WhIChminimally shelters other wage-earners from economic insecurity.

Over and above the substantive grounds for preferring thedualist to the corporatist paradigm in the Japanese case, therepresentation of labour relations within the firm as corporatistalso suffers from a fatal methodological flaw. Notions of 'microcorporatism,8 commit the error of confounding the causes andconsequences of corporatism in labour relations. The inference ismade that these relations are corporatist because they are accompanied by labour restraint, compensated by economic security ~ndwelfare. Yet scientific enquiry can hardly proceed by defimngcausal categories in relation to their presumed consequences;corporatism should rather be conceived in terms of what it is thatbrings about labour restraint/welfare in a corporatist context.

In this spirit, I would suggest that the essential features ofcorporatist industrial relations must include labour's comprehensive organization, the negotiated basis of labour restraint, and therole of the state in stimulating and facilitating concertationbetween its distributional interventions and the actions of unionsand employers. Viewed in this light, the attempt to draw aJapanese analogy at the level of the large enterprise isunconvincing. As we have seen, labour organization in Japanmirrors the splitting of the working class in the labour market;unions playa subordinate rather than codetermining role in fixingthe quid pro quos received by labour in return for the exercise ofrestraint; and, above all, no parallel can be found within the firmto the state as mediator between labour and capital and regulatorof distributional processes. This is not to argue that corporatismshould be treated as a unitary phenomenon. On the contrary, themost fruitful applications of the concept have been those whichdelineate alternative varieties of corporatism (for example,Katzenstein, 1985; Pontusson, 1983; Regini, 1982). But such

78


diversity is appropriately sought in the different ways in whichlabour ?rga,mzes as a c~ass and the varying extent of its politicalpower in different settmgs - that is to say, in variations withincorporatism's 'essen~ial' ~ramework - rather than by transplantingthe concept to a quite different locus of analysis.

Economic Adjustment in the Seventies: Has Japan BecomeCorporatist?

Irrespective of the adequacy of micro-corporatism as a conceptual~etaphor for Japanese labour relations, there remains the quitedifferent possibility that in recent decades the Japanese politicaleconomy has begun to evolve in the direction of 'real' macrolev~l corporatism. Several observers have singled out the 1975spnng wage round as a watershed which allegedly pioneered aJapanes; va~i~nt of corporatism based on 'information sharing'and an implicit contract' between labour, capital and state elites(Shimada, 1983; Taira and Levine, 1985). At the same timesome political scientists (notably Harari, 1986) argue thatJapane~e labour's t~aditional exclusion from policy-making networks I~ now a thmg of the past - and so, therefore, is thec~nceptlOn of state/society relations in Japan as 'corporatismwithout labour'. It is not possible here to evaluate thesedev~lopm.ents wit~ any ~uthority. Instead, relying only on readilyavailable mf~r~atlOn - mcluding that furnished by the proponentsof ~he 'transition to corporatism' thesis - I shall point to somesenous grounds for questioning the validity of this thesis.S~ecificall~, it is necessary to address three developments whichmight be mterpreted as a transition to neo-corporatist politicalexchange between labour leaders and the state. First, events after1974 include outcomes - the maintenance of close to fullemployment and the state's expansion of its social expenditure :-vhICh presumably could have been rewards provided by the statem re~urn for uni~n moderation. Second, the 1975 wage roundconstIt~tes a genume watershed, in which shunto militancy - theaggressiveness of both demands and tactics - declined dramaticall.y: also a presumptive indication of a turn by labour topO~ItIcal exchange. Finally, we need to weigh the evidence thatumon leaders are no longer frozen out of official policy-making

79

Shalev

forums, and may have been transformed into corporatist 'social

partners'.

Were there quid pro quos?The idea that welfare state expansion in the 1970s was anegotiated reward for labour restraint can be easily disposed of.This expansion was in full swing before 1975, and we havealready had occasion to point out that it is best interpreted as thestate's response to rising demographic need on the one hand andthe increasing political vulnerability of the LDP on the other.More puzzling is the fact that despite the severe conse~uen~esforJapan of the oil crisis, and the subsequent dechne in theeconomy's longterm growth rate, Japanese unemploymentremained below the levels prevailing in all but a handful of theWestern nations. This record is consistent with the view thatorganized labour became party to a job-pr~serv~ng 'so~ialcontract'. However, there are important quahncanons whichsurround 'full employment' in the Japanese context. Even .beforethe 1970s recession, questions were raised about the quahty andcomparability of Japanese labour-force statistics. Official data onunemployment apparently failed to monitor a good deal .of n~wlabour market slack in the period following the first 011 pnceshock, particularly that due to women's invol~ntary exit fromthe labour market or downgrading to part-nme employment(Nishikawa and Shimada, 1980). To illustrate the import of theseomissions on the basis of labour-force data assembled by Ernst(1978), it may be estimated that in 1975 Japan's officialunemployment rate of 1.9 per cent would have re~ched about 5per cent (similar to the OEeD average at the. time) after theinclusion of 'discouraged workers' and corrections for under-

counting.It is true that similar criticisms have also been levelled at the

statistics for other countries - including Social DemocraticSweden (for example, Haveman, 1978). Nevertheless, it mattersgreatly for the labour market strength of w~rker~ whetherdeclining aggregate demand is compensated pnmanly ~y theforced withdrawal of weak sections of the labour-force - migrantsin Switzerland and women in Japan - or by new jobs (or else thebridges to new jobs provided by 'active manpower poli~ies). Anadditional basis of Japanese labour market adJustment IS evident

80


in .a variety of t.rends during the mid-seventies recession. In largepnvate enterpnses, employer responses included overtime cuts,temporary lay-offs, and job redefinitions or relocations for the'permanent' work-force, while later there was a noticeable shiftin favour ~f more 'flexible' forms of employment, notablysubcontracting (Tokunaga, 1984; Koshiro, 1983). In other words,personnel policy closely followed the dynamics of a dual labourmarket. Initially workers without tenure were termined or cutback to part-time employment, recruitment of 'mid-career'workers from outside was frozen, and at the same time measureswere. adopt~d to r.ation work among tenured employees whilealso increasmg then productivity. When the demand for labour~eco~~r~d: e~~loyers.had learned their lesson, and stepped upflexibility within the internal labour market while relying mainly

on secondary labour from external sources for expansion.Not surprisingly, then, despite low measured rates of un

employment, the coming of the economic crisis threw workers inthe primary labour market onto the defensive. This is ratherdramatically attested by stagnating union membership, even afteremployment recovered; and by a dramatic decline in workstoppages. Indeed, in an 18-country comparison of trends ins~rikes and unemployment through the early 1980s, this author~mgled .out J~~a~ (along with the United States) as exemplifyinga massive crisis-induced retrenchment of strike activity' (Shalev,

1983b, p. 438). From the mid-1970s, labour militancy becamerestricted primarily to declarations of disputation involving littleor no loss of working time.

Was there a move to 'social contract' bargaining?

In 1975 and again in 1980, in the wake of the decade's two oilprice s~ocks, u?ion wage demands fell substantially belowexpectations denved from prevailing economic conditions andprior wage trends. Shimada (1983, pp. 198-9) has interpreted thisas evidence of 'flexible wage moderation' grounded in theemergence of a 'functional equivalent' of corporatist 'informationexchange and sharing among organized actors'. In a similar veinT~ira and Levine (1985, pp. 250 and 248) write that, beginnin~with the 1975 shunto,

There was an implicit contract between the government,

81

Shalev

unions, and workers. If the government k~pt its promise .to .reduce the rate of increase in consumer pnces and to mamtamemployment, workers would keep their promise ~ot ~o demanda greater increase in money wages t~~n could be justified by .increases in productivity.... In addItIOn, ... governme.nt socialand labour welfare measures were strengthened, and tnlateraland bilateral consultation ... seemed to flourish There wasalso a precipitous decline in industrial disputes .

Nevertheless from the accounts of the 1975 wage roundoffered by Shi~ada and Taira and Levine, it is doubtful t~at'information sharing' and 'consultation' were the key to union

wage and strike behaviour:

1 In the background, as Taira and Levine admi~ably demonstrate, a number of structural changes combmed to alterprofoundly the outlooks of bot~ labour and managementin large-scale private enterpnses. For ~mployers th~problems of inflation and growing internatIonal compention brought to a head their accumulatin~ unea~e with adecade-long rise in industrial conflict, the mcreasmg scopeand cost of shunto demands as labour ~arket~ ~adtightened, and the growing expense of paymg semontybased wages to an aging 'permanent' work-force. At thesame time, on the labour side the threat of unemployme~tafter 1973 joined other growing constraints on wor~ers inthe big export-oriented manufacturing firms. E~rher onthe leading edge of economic expansion, these umons hadbecome the de facto vanguard of the shunto movemen~.But soaring energy prices and other global e~ono~lcchanges, combined with a pronounced techn~loglcal dnftto automation, was paring the number of primary sectorjobs in manufacturing and greatly increasing the vulnerability of 'vanguard' labour to ~mployer. pressure. .

2 The notions of tripartite informatIon shanng and ~oclalcontract bargaining imply a status of equal partnershIp forlabour which is at odds with what actually happened. AsShimada has detailed, between the 1974 and 1975 wagerounds there was a closing of ranks among employers andhigh government officials and clo~e coordinati?n betweenthem, culminating in the adoptIOn of speCIfic targets.

82


Apart from displ~ys of 'formal' and 'symbolic' tripartism,t?e gove~nmentdid not consult the major labour organization until after its wage targets were determined andoffered the unions only vague promises of 'preparingfavourable environments' in return for wage moderation(Shimada, 1983, pp. 189 and 193).

3 Again relying on Shimada, it is clear that sticks substantially outweighed carrots in the state's behaviour towardsthe u~ions".Taking their lead from the government, andbehaving with unprecedented formal and informal disc~p!ine and unity, the leading industrial companies issuedrigid wage offers. As indicated, these companies'employees had every reason to fear declining employment- and the government's actions only reinforced employerpressure on the unions. First, a tight money policyImposed a steep wage-employment trade-off on thenegotiators. Second, a revision of unemployment insur~nce legislation (Nishikawa and Shimada, 1980), introdu~mg new.employment subsidies for distressed employerswh~le prunmg benefits to the unemployed, transmitted apo.hcy commitment to preserving the status quo in thepnmar~ labou~ market. Third, the government pursued anaggr.e~slve pohcy towards labour in the public sector, thetradltI~nal stronghold of union militancy, which mightotherwise have been an embarrassment to the posture ofself-restraint which suited labour in the privately ownedexport industries.

Shinka~a's (198~) comprehensive study of wage restraint in1975, WhIch. ~lso reJ.ects the thesis of a transition to corporatism,throws ~ddlt~onal hght on the origins of contemporary unionmoderation .m Japan.. His account reveals a profound gapbetween the internal sohdarity of capital and the state on the onehand, and labour on the other, and implies that this was a majorf~ctor behind t~e unions' acceptance of what appeared at thetime to be a painful cut in real wages. For quite specific politicalreasons, the new government formed at the end of 1974 awardeda high ~egre~ ~f autonomy to Mr Fukuda, its austerity-mindedeconOl~lc~ mimster. Fukuda succeeded in overcoming oppositionfrom within the state and close to it (government ministries smallbusiness, and important LDP circles) to a policy of cooling down

83

Shalev

the economy and pressing aggressively for wage restraint.. As aresult, by the spring of 1975 both managers an? .workers 10 keymanufacturing sectors were feeling the .chIlhng eff~c.ts ofdeflationary policies. The business co~mum~y. also .exhlblted. aremarkable degree of discipline. WhIle divided 10 b~th I~Sinterests and its formal associational activity, Japanese business ISloosely coordinated by the Zakai - an ~nfluential alliance o~ topindustrial and financial leaders spanmng all the key bus1Oes~organizations. In the inflationary crisis of. 1974, the Zakazsucceeded in forging a class strategy and 10bbY1Og for a toug~ andconcerted management line on wages. In contrast, on the SIde.oforganized labour longstanding divisions between the majornational groupings had been deepened in the. early 1970s by thesuccess of Sohyo's efforts to 'politicize' uniomsm and the ~hun.to.A crucial motive on the part of the 'bread and butter uniongroupings (and one wholeheartedly supported by both em~loyersand the government) was their desire to isolate Sohyo. ThIS t~eysucceeded in doing by accepting the government/emp.loyer h?eon wages and thereby cultivating excellent p~blic r~l~tIons whl~eleaving Sohyo to fight a losing battle with militant pubhc

employers.

Has labour become a partner in policy networks?

Union leaders in 1975 and thereafter did have grounds for hop~ngthat big employers and the state would respect. the norm of Jobsecurity in the primary labour market, provIded that labourexercised pronounced wage restraint.9 But r~the~ than ~epresenting a tripartite accommodation, the umons readiness f~rconcessins stemmed from the short- and longterm ~conom~cvulnerability of the principal players on the labour side, theirdetermination to alter the balance of power w~thin the la?ourmovement and concerted action against the unions by businessand the g~vernment. These forces combined t~ vitiate the oldshunto strategy of a highly visible national campaIgn for large real

wage increases. . 'This at least is the scenano which I have deduced from the

secondary evidence cited. Nevertheless, doubt wo~ld b~ cast onthis scenario if it could be shown that, far from being SImply theobjects of employer and state initiatives,. top uni~n leaders ~adactually become policy-making partners 10 a setting of genume

84


tripart~sm. Some o.bservers have indeed implied something alongthese hnes. In particular, Ehud Harari (1986 and this volume) hasargued that labour's exclusion from policy networks wasexaggerated in past scholarship and has become untenable as acontemporary description. He contends that for a mix ofideological and instrumental reasons, union federations hadthemselves been at least partly responsible for their own pastabsence from consultative frameworks; but that in any event, thisgap has to an important extent been corrected. Citing data on thecomposition of statutory advisory bodies, Harari shows that inthe decade after 1973 union representation on such bodiesexperien.ced . a m~jor quantitative increase accompanied bypenetration into hitherto closed circles. Moreover, it is shownthat labour representatives have included the 'militant' as well asthe 'moderate' stream of unionism, and that the delegatesthemselves do not regard their role as tokenistic.

~arari's findi~gs, while an important corrective to simplisticnotions of organized labour as a political pariah, nevertheless fallsh~rt of s~bstantia~i~g the. kind of policy-making partnerships inwhich unions participate 10 the corporatist nations of WesternEurope ~see Katzenste.in, 1985; Taira and Levine, 1985, p. 266).It remains the case 10 the 1980s that business interests arero~tin~ly. represented on advisory bodies, whereas only amI?Onty include representatives of labour. The latter in any caseenjoy only limited participation, especially in the 'sensitive'are~s fro~ which unions were formerly excluded altogether. Themajor union confederations receive their weightiest representationon committees dealing with employment and welfare - areas inwhich state intervention is modest, to the satisfaction oforganized workers, sheltered as they are by the primary labour~arket. In :ontrast, as Harari points out, union delegates havelittle or no influence where they need it most - on committeesoccupi~~ with the for~ation of broader macroeconomic policies.In addition ~o th~se hmitatio~s, Itoh's (1986) independent studyof the relationship between interest groups and state agenciesreveals quite striking qualitative differences between labour andmos~ ot~er types ?f associations. On the one hand, many unionofficials involved 10 such contacts (although proportionately lessthan ~he representatives of business and most other groups) didperc~Ive open channels of communication. But they were also farless likely to report stronger indications of interpenetration, such

85

Shalev

state subsidies or so-called 'administrativeas receipt of direct

gu~~a~acc~':as ~~ink~wa (198~) poin::~~~~ t:~ ~~~~~: c~::::it~~::to partial leghltm~a~lOn1;;0~ 1l~~:~ influential figures within thedates from teae, b li e that (a) theruling Liberal Dem~c~atic.~:rr% c:~:e~byee:;loyers and theaggressive labour pohclesdhdl. ~erating a 'settlement' consis-t t had already succee e m ge b

s a e . . h: (b) this settlement could be upset ytent WIth economIC growt uch the Mitsui coal-miner's strike oftraumatic confrontations. (SUC

dpa~litical alienation of left-leaning

1960)' and (c) the contmue . . f th, liti lly embarrassing imphcatlOns or e

unions could ha:e po ~~ca r conflict over the US-Japan Securitygovernment (as m the ~tte was reinforced in the lateTreaty). This pre-emptIve rehsponse ption on the part of an

d I 1970s by t e perce . . .1960s an ear.y. f rowin shunto activism as signifyingelectorally dechmng LDP 0 g I t 'k

g[which could] turn into

" threat of a genera s n e . . . .an ormnous he uni -supported left-wing opposi-a political chall~nge from t. e u~~~~ 258). It was not until thetion parties' (Taira and Lev~n~, ti ':~s backed up by the fist ofglove of limited policy partIClpa ~on dered both their high-

. t inty that the umons surreneconomIC uncer at d d broader mobilization andr y and the tren towar swage po lC k f the 'spring offensive'. There ared d in the framewor 0 .

eman s nds for believing that expanded umontherefore no strong grou k stituted a qualitative break. I' . the policy networ con . dme usion l~. . h conomic crisis, when orgamzewith condItIonS pnor to tee t abandon mass mobilizationslabour evidently sa~ no rea~on 0 The retreat from the labourfor the lure of pohtIcal ;;c. a~:~~er explained as the product ofmilitan.cy of the ea~ly 19 d

SIS unified external opposition and

damagmg economIC tr,en S~hamselves than as a transition to adisunity among the umo~s (~ h'1977) rooted in politicalnew bargained corporatlsm roucn,inclusion of the labour movement.

Conclusion

t of distributional conflicts in the labour marketThe manag~me~ b in th Japanese context from thoserests on quite dIfferent ases m e

86


which underpin corporatist arrangements in the Social Democraticnations of Western Europe - even though in the last decade anda half these countries have shared with Japan the distinction ofrelatively favourable economic performance (particularly inrelation to unemployment) achieved in a relatively consensualatmosphere. I have criticized each one of the routes along whichscholars have attempted to bring Japan under the analyticalumbrella of the corporatist model. At the outset it was shownthat quantitative studies of corporatism from a cross-nationalperspective exhibit uncharacteristic disagreement over whetherJapan is or is not 'corporatist'. This disagreement reflects notonly the confused conceptual and methodological state of the artbut also real ambiguities. Notions of corporatism which link it topowerful and solidaristic labour movements are clearly alien toJapanese realities, and so are those which single out thecharacteristic institutions of corporatist policy-making. Theoretically and methodologically, it is more fruitful to treat thesedifficulties as grounds for seeking an alternative model tocorporatism for the Japanese case, than to constrain concept andcase to converge.

If such a convergence is nevertheless to be achieved, it requireseither that labour's position in the Japanese political economy (asevidenced in patterns of employment, labour relations, and socialpolicy) can be shown to be in some sense the functionalequivalent of European-style corporatism; or that Japan hasactually been moving towards the European model of strategicclass action by labour on the basis of political incorporation andexchange. This essay's critical reconsideration of the evidence foreach of these propositions casts doubt on both. The 'welfarecorporatism' said to characterize large private enterprises is bestunderstood in terms of the dualism of the Japanese economy,labour market, and working class, and the linchpin role played bythe private/public split in social welfare. On the macro-societalplane, the participation of Japanese labour associations in themaking of domestic public policy remains severely bounded, andtrade union elites are still far short of engaging in active socialcontract bargaining. Labour quiescence since the middle of the1970s is better explained by the whip of market conditions, thepressure of unified opponents, and the reinvigoration of the duallabour market, than by the elevation of peak labour organizations to the status of guardians of public order.

87

Shalev

All this suggests that, for comparativists interested in locatingJapan on a cross-national map, greater attention to the context ofdistributional conflict is required, in place of continued attemptsto impose a single-stranded conceptual model on Japan'sempirically inhospitable environment. This certainly need notnecessitate a retreat to the particularistic case-study approach.One country which shares with Japan a hegemony of big businessand conservative politics co-existing side by side with fullemployment, labour-management collaboration and successfulpolicies of economic adjustment, is Switzerland (Katzenstein,1980). Like Japan, Switzerland is characterized by a comparativelylow rate of labour organization in divided and decentralized tradeunions, and a modest socialist vote. But the Swiss model oflabour movement incorporation into institutions of state in effectsignifies the upper limit to the Japanese trend towards endinglabour's exclusion while perpetuating its inability to rule. No lessimportant, the labour markets of both countries are deeplydualistic, with large strata of relatively powerless workersoccupying undesirable job slots. The union movement, reflectingthe concerns of primary sector workers and their employers,excludes secondary sector labour and consents to a dual welfaresystem which perpetuates residual social provision by the stateand helps preserve harmony in the enterprise. The synthesis ofco-optation in politics and dualism in the market arena commonto Switzerland and Japan appears to offer a more enlighteningmodel of labour in the Japanese political economy than attemptsto constrain the corporatist model to fit Japanese realities, orvice versa.

Notes

1 Much of this chapter is in the nature of an incursion into terrainusually reserved for experienced Japanologists, by an author withlittle primary expertise on Japan. I am grateful to the editors of thisvolume for encouraging my audacity and to participants in seminarswhich they organized for helpful comments; to T.J. Pempel forpointing me to valuable sources; to Toshimitsu Shinkawa and CarlLe Grand for sharing their unpublished work with me; and to WalterKorpi and the Swedish Institute of Social Research for generous

88


hospitality while the paper was in preparation.2 The Pearson correlation between the Cameron and Cd'

091 (n-17) (N za a scores ISo~iginal-dat . ?te that here, as in all subsequent calculations thefrom -1 to a+s;)les were transformed into a common scale ra~ging

3 For a rich collection of Sources relevant to the' ..Moore and Oldfather (1976). Moore (1983) pen~~, se~ LlVings.t~n,acc?unt of the period of labour resurgence e~~~vline~h~ ~c~~fi:~~~~

TPen od. I rely here on informative articles by Sumiya (197:) d

okunaga (1983) and also 0 A an(1984 pp 136--40)' '1' n rmstrong, Glyn and Harrison's

4 ' . . compi anon of contemporary Sources.T~e claim that these constraints on worker solidarity have been thpnce exacted by management for its conditi 1 e'. . I iona acceptance of tradeun~08~sm IS substantiated by Sumiya (1974, p. 63), Taira (1970p'. ), I?ore (1973, p. 325), and, regarding the imposition of a 'n~umo~lzatlOn. clause' on subcontractors, by Carl Le Grand of thSwedish Institute for Social Research who ki dl h d wi eresu~ts of his unpublished fieldwork ~n sUbc~nt:a~ti~re With me the

5 ~~;~Cu~rlr2~e)vea.ling in this respect is that, accordin~ to Tokunaga,. .' SInce 1965 the leading iron and steel firms hav

pr~sented a smgle take-it-or-leave-it wage offer to unions in th .e

~~ ~~:~;oa~e~i~:e:~C\~p~ars to have influenced the overall le:~~. IS IS not to argue that Japanese k

would have been just as well off without shunto. Koshiro,;0~9:;spp. 23Q-.8) econometric evidence indicates that shu t 'l~ 'in the public and . t n 0 mi itancyh . pnva e sectors has raised wages - although he 1

sows that spnng wage ttl a so. se ements have been consistent with th

mamtenance of profitability. I e6 These barriers apply to all but the 'h If- kers' .bo . d a way wor ers recruited in

emo~ peno s by l~rge enterprises from smaller firms or the self-

1p oyed. But, tellIngly, these mid-career recruits are penalized by

ower pay unless they pr d .(Koshiro, 1983, p. 257). 0 uce an Impeccable performance record

7 Regabrding rewards: (a) under the pressure of labour shortages thegap etween men's wages in fi f dif 'd . rms 0 I ferent sizes narrowed~~~:t the ~960s; (b) from a life-cycle perspective, primary labourare low wor ers ~re n~t ~lways advantaged - their starting salariest ' and so IS their mco~e after retirement; (c) it is plausibleo assume that household income in the long .

more equal th th run IS much. an e cross-sectional gap between individual k

at any given moment in time (Pempel 1989) thi wor erswhy income I r b ' - IS IS presumably. e mequa ity etween households is lower in Japan thm most OECD countries. an

8 Goldthorpe himself (1984 340)b . f f ' p. uses the term micro-corporatism in

a ne re erence to the Japanese case, while stressing that a Japan-

89

Shalev

style adjustment to dualism necessarily results in 'the fundamentaldivision and effective depoliticization of the working class'.

9 Unions in 1975 agreed to a nominal wage increase below the rate ofinflation. The actual increase in real wages after 1973 remained under3 per cent for the rest of the decade - whereas earlier in the decade itwas consistently above 8 per cent (Taira and Levine, 1985, p. 248).

References

Armstrong, P., A. Glyn and J. Harrison (1984), Capitalism since WorldWar II: The Making and Breakup of the Great Boom. London,

Fontana.Bruno, M. and J.D. Sachs (1985), The Economics of Worldwide

Stagflation. Oxford, Basil Blackwell.Cameron, D.R. (1984), 'Social democracy, corporatism, labour quies

cence and the representation of economic interest in advancedcapitalist society' in J.H. Goldthorpe (ed.), Order and Conflict inContemporary Capitalism. Oxford, Oxford University Press,

pp.143-78.Castles, F.G. (1978), The Social Democratic Image of Society. London,

Routledge.Crouch, e. (1977), Class Conflict and the Industrial Relations Crisis:

Compromise and Corporatism in the Policies of the British State.

London, Heinemann.Crouch, e. (1980), 'The conditions for trade-union wage restraint', fifth

draft of paper for Brookings Institution project on the Politics and

Sociology of Global Inflation.Czada, R. (1987), 'The impact of interest politics on flexible adjustment

policies' in H. Kernan, H. Paloheimo and P.F. Whiteley (eds.),Coping with the Economic Crisis: Alternative Responses to EconomicRecession in Advanced Industrial Societies. London and Beverly Hills,

Sage, pp. 20--53.Dore, R. (1973), British Factory - Japanese Factory: The Origins of

National Diversity in Industrial Relations. Berkeley and Los Angeles,University of California Press.

Ernst, A. (1978), 'Unemployment and underemployment in Japan: anoutline of the state of empirical research' (in German), Mitteilungenaus der Arbeitsmarket und Berufsforschung, 1.

Esping-Andersen, G. (1985), 'Power and distributional regimes', Politicsand Society, 14(2), pp. 223-56.

Goldthorpe, J .H. (1984), 'The end of convergence: corporatist anddualist tendencies in modern western societies' in J.H. Goldthorpe

90


(ed.), Order and Conflict in Contemporary Capitalism. Oxford,Oxford University Press, pp. 315-43.

Goldthorpe, ~.H: (1984), idem (ed.), Order and Conflict in Contemporary Capitalism. Oxford, Oxford University Press.

Gordon, A. (1985), The Evolution of Labor Relations in Japan: HeavyIndustry, 1853-1945. Cambridge, Mass., Harvard University Press.

Hall, R. (1988), 'Enterprise welfare in Japan: its development and role'Welfare. State Programme, Suntory-Toyota International Centre fo;E~ono~lcs and Related Disciplines, London School of Economics,DISCUSSion Paper No. WSP/31.

Hana~i, T. (1~79), Labor Relations in Japan Today. Tokyo, Kodansha.Haran, E. Polley Concertation in Japan. Occasional Paper No. 58/59 of

the Eas~ Asian Seminar, Free University of Berlin's Social andEc~nomlc Research on Modern Japan Series. Berlin: Verlag UteSchiller, 1986.

Have~an, R.H. (1978), 'Unemployment in Western Europe and theUmte~ States: a problem of demand, structure, or measurement?',American Economic Review, 68(2), pp. 44-50.

Itoh, M.. (1986), 'Labor unions as interest groups: the case of Japan'un~ubh~hed paper, Political Science Department, Nagoya Stat~University.

Jacoby, S. (1979), 'The origins of internal labor markets in Japan'Industrial Relations, 18(2), pp. 184-96. '

Kalecki, M. (1943), 'Political aspects of full employment', PoliticalQuarterly, 14(4), pp. 322-31.

Katzenste~n, P. (1980), Capitalism in One Country? Switzerland in theIntern~tlOnal Economy. Cornell University, Western Societies ProgramOccasional Paper No. 13.

K~tzenstein, P. (1985), Small States in World Markets: Industrial Policytn 1!urope. Ithaca, NY, Cornell University Press.

Korpi: v:t. (1980), '.Social policy and distributional conflict in thecapitalist democracies. A preliminary comparative framework' WestEuropean Politics, 3(3), pp. 296-316. '

Korp~, W. and M. Shalev (1980), 'Strikes, power and politics in westernnat.lOns, 1900--1976', Political Power and Social Theory, 1, pp. 301-34.

KoshlfO: !<. (198~),. 'Development of collective bargaining in post-warJapa~ In T. ~hlra~ (ed.), Contemporary Industrial Relations in Japan.Madison, University of Wisconsin Press, pp. 205-57.

Lee,. H.K. (1987), 'The Japanese welfare state in transition' in R.R.Fnedman,. N. Gilbert and M. Sherer (eds.), Modern Welfare States: ACo,:,par.atlve View of Trends and Prospects. New York, New YorkUniversity Press, pp. 243-63.

Lehmbruch, G. and P.e. Schmitter (1982), idem (eds.), Patterns ofCorporatist Policy Making. London and Beverly Hills, Sage.

91

Shalev

Lehner, F. (1987), 'Interest intermediation, institutional structures andpublic policy' in H. Keman, H. Paloheimo and P.F. Whiteley (eds.),Coping with the Economic Crisis: Alternative Responses to Econo~ic

Recession in Advanced Industrial Societies. London and Beverly HIlls,Sage, pp. 54-82.

Lincoln, J.R. and A. Kalleberg (1985), 'Work organization andworkforce commitment: a study of plants and employees in the USand Japan', American Sociological Review, 50, pp. 738--60.

Livingston, J., J. Moore and F. Oldfather (1976), idem (eds.), The JapanReader: Volume Two - Post-war Japan, 1945 to the Present.Harmondsworth, Penguin.

Maruo, N. (1986), 'The development of the welfare mix in Japan' in R.Rose and R. Shiratori (eds.), The Welfare State: East and West. NewYork and London, Oxford University Press, pp. 64-79.

Nishikawa, S. and H. Shimada (1980), 'Employment and unemployment' in S. Nishikawa (ed.), The Labor Market in Japan: SelectedReadings. Tokyo, University of Tokyo Press, pp. 124-41.

Panitch, L. (1981), 'Trade unions and the capitalist state', New LeftReview, 125, pp. 21-43.

Pempel, T.J. (1989), 'Japan's creative conservatism: continuity underchallenge' in F.G. Castles (ed,), The Comparative History of PublicPolicy. Cambridge, Polity Press, pp. 149-91.

Pempel, T.J. (1982), Policy and Politics in Japan: Creative Conservatism.Philadelphia, Temple University Press.

Pempel, T.J. and K. Tsunekawa (1979), 'Corporatism without labor?The Japanese anomaly' in P.c. Schmitter and G. Lehmbruch (eds.),Trends Toward Corporatist Intermediation. Beverly Hills and London,Sage, pp. 231-70.

Pizzorno, A. (1978), 'Political exchange and collective identity inindustrial relations' in C. Crouch and A. Pizzorno (eds.), TheResurgence of Class Conflict in Western Europe since 1968 (Volume 2).London, Macmillan, pp. 277-98.

Pontusson, J. (1983), 'Comparative political economy of advancedcapitalist states: Sweden and France', Kapitalistate, (10-11), pp. 43-73.

Regini, M. (1982), 'Changing relationships between labour and thestate in Italy' in P.C. Schmitter and G. Lehmbruch (eds.), TrendsTowards Corporatist Intermediation. Beverly Hills and London, Sage,pp.109-32.

Schmidt, M.G. (1982), 'The role of parties in shaping economic policy'in F.G. Castles (ed.), The Impact of Parties. London and BeverlyHills, Sage, pp. 97-176.

Schmitter, P.C. (1974), 'Still the century of corporatism?', The Review ofPolitics 36(1), pp. 85-131.

Shalev, M. (1983a), 'The social democratic model and beyond: two

92


"generations" of comparative research on the welfare state' Compara-tive Social Research, 6, pp. 315-51. '

Shalev, M. (1983b), 'Strikes and the crisis: industrial conflict andunemployment in the western nations', Economic and IndustrialDemocracy, 4(4), pp. 417-60.

Shimada, H. (1983), 'Wage determination and information sharing: analternative approach to incomes policy?', Journal of IndustrialRelations, 25(2), pp. 177-200.

Shinkawa, T. ~19,87), 'A J.apanese style incomes policy: the 1975 springlabor ~f!ensive., unpublished paper, Toronto University, Departmentof Political SCIence. (English version of an article published in H.Ohtake (ed.), Nipponseiji no Soten. Tokyo, Sanichi Sobo, 1984.)

Steven, R. (1982), Classes in Contemporary Japan. CambridgeCambridge University Press. '

Sumiya, M. (1974), 'The emergence of modern Japan' in K. Okochi, B.K~rsh and S.B. Levine (eds.), Workers and Employers in Japan.Pnnceton and Tokyo, Princeton University Press and University ofTokyo Press, pp. 15-48.

Taira, K. (1970), Economic Development and the Labor Market inJapan. New York and London, Columbia University Press.

Taira, K. and S.B. Levine (1985), 'Japan's industrial relations: a socialcompa~t emerg~s' in !I' juris, M. Thompson and W. Daniels (eds.),Industr~al Relat~ons In a Decade of Economic Change. Madison,Industria] Relations Research Association, pp. 247-300.

Takahashi, T. (1974), 'Social security for workers' in K. Okochi, B.Karsh, and. S.B: Levine (eds.), Workers and Employers in Japan.Tokyo, University of Tokyo Press and Princeton University Press,pp.441-84.

Tarantelli, E. (1986), 'The regulation of inflation and unemployment'Industrial Relations, 24(1), pp. 1-15. '

Therborn, G. (1986), Why Some Peoples are More Unemployed thanOthers. London, Verso.

Tokunaga, S. (1983), 'A Marxist reinterpretation of Japanese industrialrel~ti?ns, with special reference to large private enterprises' in T.Shirai (ed.), Contemporary Industrial Relations in Japan. MadisonUniversity of Wisconsin Press, pp. 313-29. '

TOkunaga, S. (1984), 'The structure of the Japanese labour market' in S.Tokunaga and J. Bergmann (eds.), Industrial Relations in Transition:The Cases of Japan and the Federal Republic of Germany. TokyoUniversity of Tokyo press, pp. 25-55. '

Wilensky, H.L. (1975), The Welfare State and Equality: Structural andIdeological Roots of Public Expenditure. Berkeley and Los AngelesUniversity of California Press. '

93

4 class conflict, corporatism and comparison: a japanese...

Documents