The Welfare State and the Economy

8/13/2019 The Welfare State and the Economy

1/26

European Journal of Political Research 11 (1983) 1-26Elsevier Scientific Publishing Company, Amsterdam. Printed in The Netherlands

The Welfare State and the Economy in Periodsof Economic Crisis:

A Comparative Study of Twenty-three OECDNations

MANFRED G. SCHMIDT

Universiryof Konstanz. Federal Repu blicof Germany

ABSTRACT

While all industrial nations have been plagued by the setback to economic growth since the mid-lWOs, awide range of variation marks the economic performance, policy outputs and policy outcomes of Westernnations. The basic question addressed in this paper is the extent to which economic, socioeconomic andpolitical variables account for the differences in economic performance, unemployment, public debt andthe growth of social security expenditure. In general, the forces which shape public policies in periods ofeconomic crisis tend to be different in character from the major determinants of policy-making in periodsof prosperity. The analysis does not support the view that bourgeois and socialist governments produceclear-cut policy differences. It suggests that the major determinants of policies are the power relations inextra-parliamentary arenas, the level of national economic strength prior to the crisis, the extent to whichsolidaristic values characterize the political culture, and the extent to which a correspondence existsbetween the power relationships in the political superstructure and in the party system and industrialarenas. An expanding capitalist order is, in theory, not incompatible with low unemployment and adeveloped welfare state. However, the governments room to manoeuvre is strongly restricted by politicaland technological developments. One of the major political restrictions is the paradoxical outcome ofelections.

1. POLICY MAKING IN PERIODS OF ECONOMIC CRISIS

When the history of the second half of the twentieth century comes to be written, nodoubt the achievements of high growth rates, full employment, low rates of inflationand the expansion of the welfare state in the 1950s and the 1960s will be seen as thehallmark of a successful partnership between economic and social policies in theindustrialized countries with managed capitalist economies. In the 1970s the setbackto economic growth, rising rates of unemployment and inflation and the increase inpost-materialist demands, and the fiscal crisis of the welfare state have produceddifficult problems for the reconciliation of a managed market economy and themaintenance of a developed welfare state.

1

0304-4130/83/ 03.00 @ 1983 Elsevier Science Publishers B.V.


2/26

2The political science literature is packed with descriptive and explanatory accounts

of a deep crisis. According to a widely shared view, the stage was set for long-standing ungovernability problems (Lehner, 1979; Offe, 1979; Rose, 1979; Dror,1981); policy-makers were confronted with self-produced overload (Luhmann,

1981); the Social Democratic consensus (Dahrendorf, 1979, 147-66) and theKeynesian class compromise (Altvater, 1981) were tending to break down; and theredistributive welfare state had overburdened the market economy (Wittmann,1981; OECD, 1981; OECD Economic Surveys: Sweden, 1981; Stephens, 1981)-notonly in the leading welfare states in Scandinavia, but also under bourgeois rule(Dahrendorf, 1981).

Th e economic and labour market crisis from 1973 to 1974 onwards was the result offive interrelated developments which tended to undermine the successful partner-ship of a managed capitalist economy and a welfare state which had emerged in the1950s and 1960s. These were: the wage policy of organized labour; the strongexpansion of non-wage labour costs; the emergence of newly industrialized nationswhich challenged the competitiveness of old industrial branches in the developednations; the increase in raw material prices; and the instability inherent in a marketeconomy. All these factors contributed to a decline in capital return rates. Theou tcome resembles the political business cycle which Michael Kalecki (1943) haddescribed. Entrepreneurs responded to the profit squeeze with lower investmentrates, export of capital, and investment in labour-saving new technology. As a result,the number of unemployed increased, rates of economic growth and revenuesdeclined , and the public authorities were sooner or later confronted with a fiscal crisisof the state and with requirements of a restrictive policy stance and a decrementalbudgetary process. They had thus to face a new challenge which could not any longerbe managed within the old model of incremental and distributive policy-formation(Dror, 1981). The response to economic crisis has not been uniform across thecapitalist democracies. All advanced industrial democracies were hit by theeconomic constraints of the mid-l970s, but there was a wide range of variation inboth political initiatives and policy outcomes.

In which way have the governments of the economically developed democraticOECD nations reacted to the unprecedented challenge? Which policy stance didthey choose? To what extent have they been successful in maintaining and expanding

welfare state provisions and in fulfilling the growth requirements of a marketeconomy? How is the observable variation in policy efforts, outputs and outcomes tobe explained? To what extent is the political composition of government a variablewhich accounts both statistically and substantively for the variation? Under whichcondit ions were the public authorities punished or rewarded by the electorate? Canwe still assume that the notion of a constant and positive relationship between thedegree of policy effectiveness and the degree of political support is a valid formula forthe 1970s and early 1980s? These are the questions to which we will now turn.

We focus our study on 23 OECD nations which are economically more or lessadvanced and which have a democratic order with competitive regular and general

elections in the majority of the years under study. The period under investigation isfrom the beginning of the world-wide recession in 1973-74 to the opening of the1980s. The policy indicators we have chosen are basically quantitative in character.The first indicator measures the economic performance - judged by bourgeoiseconomic common sense. This indicator was taken from the Euromoney study on


3/26

3

the End of the OPEC era (Euromoney, 1981).* A second class of indicatorsmeasures th e extent to which governments have been capable of solving problemswhich are associated with class conflicts and demands made by organized labour.Measures of the rate of unemployment and the trend in civilian employment indicate

the extent to which demands for employment security have been met.3 Third, ourfocus will be on the changes in social security expenditures between 1973 and1979-80. This indicator measures the extent to which demands made by the clienteleof the welfare state have been met.4 Fourth, we will study the extent to whichgovernments have based their policies on an increase of public debt. This indicatorreveals t he extent to which contemporary policy-making is short-sighted in characterand operates at the expense of the f u t ~ r e . ~

Further, the thrust of our investigation will be on the extent to which the results ofcomparative studies on public policy in periods of economic prosperity are also validexplanations for policy-making in periods of economic crisis. Basically, our focus willbe on variables, which are political, sociological and cultural in character (Hibhs,1977; Cameron, 1978; Tufte, 1978; Stephens, 1980; Castles, 1982a, 1982b; Schmidt,1980, 1982a, 19 82 ~) . t is not our intention to direct the focus of our study onexplanations which have been advanced by Keynesian and neo-classical schools ofthought. Generally speaking, economic explanations, by themselves, do notadequately account for the observable variations in policy outcomes. The majorreason seems to be that variables, which can be broadly grouped together under theinstitutionalist and sociological labels (Goldthorpe, 1978; Andrain, 1980; Scharpf,1981) for example, the importance of cultural and legal norms, organizationalstructures, political interests and strategies beyond the reach of the economiccalculus end to be neglected in explanations advanced by economists.

For example, within the context of an institutionalist paradigm, scholars havefrequently argued that the political composition of governments and the distributionof power in the party system are variables which account for a wide range of variationin policy outputs and outcomes. According to this school of thought, the rewardstructure, the laws of motion and the imbalances of a capitalist economy and ademocratic political order are basically amenable to political control. They can bechanged, provided that governments, and especially Socialist governments, arewilling and politically strong enough to overcome the resistance of entrenched

business interests. Th e underlying assumption in this school of thought is thatdifferent political structures have a differential impact on the course, content andconsequences of public policies.

According to another school of thought it is class politics which makes a difference.Au thor s who write in terms of a class politics paradigm argue that conflicts betweensocial classes, the growth and strength of the labour movement and of businesspower, the mode of regulating the class conflict, the strength and cohesiveness ofbourgeois and left-wing tendencies, and strategies of social demand anticipation onthe part of governments, have determined public policies within the scope of therelative autonomy of the state. In this view, it is not party politics that matters, but

ra ther the balance of competing forces in the society as a whole (see, for example,Stephens, 1979; Korpi, 1980; Schmidt, 1982a, 1982c; Madsen, 1981; Wilensky,1981).

Finally, according to a third view, which is typically taken by authors who write interms of structural functionalism, it is the socioeconomic mperatives and constraints


4/26

4

inh er en t in a ca pitalist ec onom y and a dem ocratic political ord er which are primarilyresponsible for sh aping the actionsof governments. Th e problemsof capital-such asbusiness cycles, wasteof resou rces and environmental pollution ombined with thes t ructures of a tax state (Schumpeter) and with the need for legitimation have

cre ate d th e age nd a fo r th e m odern states activities, largely irrespectiveof whichpolitical t end enc y is in pow er (see, for exam ple, Offe, 1972; an d, as an example ofold er versionsof this theo ry, Ba uer, 1928).

I t is o ur intention to put the competing explanations to the test. Th e key conceptsof the various hypotheses were translated into the languageof quantitative andqu alitat ive indicators (see A ppendix; and for a more detailed discussion, Castles,1982b; Sch mid t, 1982a, 1 9 8 2 ~ ). h e inference presentedin ou r study is based on awide range of one-country and comparative studies in policy-making and on astatistical analysisof available comp arable d ata (see Appendix).6

2. TH E DISTRIBUTION OF POWER IN POLITICAL SUPERSTRUCTURE ANDPOLITICAL SUBSTRUCTURE

W e will beg in o u r investigation with a brieflook at th e distributionof political powerbe tw ee n th e bourgeois (conservative, Christian-democratic and centre parties) andth e socialist (social democratic an d comm unist) tendencies. W e classify the demo-cratic industrial countries by the strength and durationof bourgeois and socialisttend enc ies: t he ex ten t to which they have participated in governments.

The political compositionof national governments in theOECD area between1974 and 1980 is strongly skewed towards the dominance and hegemonyof thebourgeois tendency (see Figure 1).N o more than three countries - Sweden,Lu xem bou rg an d Finland re characterizedby a balanced complexionof govern-ment. In Denmark, the United Kingdom and the Federal Republicof Germany,Social Democratic parties were the leading governing parties, whereas a SocialD em oc rat ic hegemon y in this period was the exception (Norway and Austria) ratherthan th e rule .

However, the distr ibution of power is not fully characterized by the politicalcomplexion of governments. For example, there is a large difference between abourgeois government confronted with both a strong and powerful trade union

AUSTRALIABELGIUMICELANDI E L A N D F R A N C EI TA LY G R E E C E

DENMARK NET HER LAN DS JA PA NWDERAZ, REPUBLIC FINL AND NEW ZEALAND CANADA

AUSTRIA OF GERMANY LUXEMBOURG PORTUGAL SPAINNORWAY UNITED KINGWM SWEDEN SWITZE RLAND USA- ~~ ~~ ~-Soc a 1 Soc ia l - Ba lance Bourgeo i s Bourgeo i sDemocra t i c Democra t i c dominance hegemonyhegemony dominance

Fig. 1. Political co mpos ition of national governments 1974-80.

Source: Schmidt, 1982a. Keesings Archiv der Gegenwart.For he definition of the categories see note 7


5/26

5

movement and a strong and united socialist opposition, and one which is supportedby a dominance of bourgeois tendencies in the party system and industrial relations.Fur ther, the degree of ideological and organizational cohesiveness of the bourgeoisand the socialist tendencies is a crucial variable which influences the overall distri-

bution of power.The typical pattern in the OECD nations is that of a differential disharmonybetween the power distribution in the political superstructure (measured by thepolitical composition of governments) and that in the political substructure (defined

1

2

3

4

5

political compositionof national governments

Reference Lineof perfect harmony

Canada.USA

Italycel.. elg.

. wAustral.

..Ireld.Port. . Neth.

Political Substructurenumber of conditions

industrial relationsfavourabl e to bourgeoispolicies)

.1 2 3 4 in party system and

Extent of disharmony

increasing 4 decreasingeconomic order

with

Fig. 2. Disha rmon y between power relationshipsin the political sup erstructureand the political sub-s t ructure 1944-79/80.

by the power relationship within the party system and industrial relations, andmeasured by the number of conditions which are favourable to bourgeois policies).Frequently, governments in the OECD area are confronted with adverse powerrelationships in the political substructure, and thus their room to manoeuvre istypically far more limited than might be expected on the basis of their politicalcomplexion (see Figure 2).

3. POLICY-MAKING, ECONOMIC PERFORMANCE AND THE WELFARESTATE

The governments of the industrial democracies reacted to the challenge of theeconomic crisis in the mid-1970s in a way which differed strongly from the policystance which governments generally had taken in the depression years of the 1 9 3 0 ~ ~In the 1970s the public authorities were generally more inclined to contain the


6/26

6

setback to econom ic growth, t o expand the scopeof state-financed program mes, an dt o ex pa nd lab our m ar ke t policies and social security expenditures. It is true that theove rall policy st an ce was in general not m ore than moderately expansive, and it isalso true that go vernmen ts met the secondoil price shock of 1979-80 with a tighten-

ing of agg reg ate de m an d policies. Nevertheless, there was no real parallel to thestrongly restrictive policy-making which had featuredso prominently in the late1920s and ear ly 1930s. The restrictive policy stance taken by the Reagan adminis-trat ion and the Tha tcher government was the exception.

3.1 Variation in the Politics of Crisis Management: Muddling Through andSuccessful Political Control of the Economy

T h e resp onse to the econo m ic crisis has not been uniform among the democraticO E C D - n a t i o n s . A wide range of variation in policy efforts, outputs and outcom esw as ob ser va ble in the industrialized W estern world (see TableI) . A t the bottom wefind na tio ns whose eco nom ies and policy-making are characterizedby a patternwhich m ay be des crib ed as unsuccessful muddling through. Gr eat Britain, Ireland,Italy an d Sp ain belong to this category. In these countries the economic performancew as un sta bl e, with low grow th rates and high rates of inflation, declining competi-tiveness, high rates of unemployment and low growth rates in social expenditures.Moreover, in Ireland and Italy the public debt increased to an enormous extent.N orw ay, A ustr ia an d Jap an are th e nations which rank at the top, with growth ratesbeing on the whole relatively high, comparatively low ratesof inflation, stable orincreasin g com petitiven ess, low unemployment figures, and , in the Japanese case,high growth rates in social expenditures (whereas the Norwegian and Austrianau tho rit ies e xp an de d the ir social security programme at a slower pace). It was inth ese c ou nt rie s that th e increase in the public debt seems to have been successfullyused for the prom ot ionof a stable econom ic growth with low unemployment and anexpansive welfare state. However, it is obvious that the Japanese authoritiesop er at ed o n a qu ite different basis from the Norwegian and Austrian governments.W he re as in t he latte r cases efforts to control the economy politically took place onth e basis of a dev elop ed welfare state , in Jap an economic growth and the expansionof social security e xp en di tur es started from the baseline of an underdeveloped

welfare state.The major i ty of th e O E C D nations rank in between these extremities of un-successful muddling through o n th eone hand and successful econom ic and welfaresta te m ana gem ent on th e oth er. In these nations, governments were in general onlysuccessful in oneor tw o are as and unsuccessful in ot he rs. It was in these countriesthat the requirementsof balanced economic growthor the dem ands forfull employ-m ent an d/ or th e d em and s for social security expenditures were not me t. However,diff eren t stra teg ies and policy results can be observed also in this groupof countries.The first strategy is characterized by the primacy of economic accumulation andinflat ion c ontro l . Low o r moderate ratesof inflation and low ratesof increase in the

public d e bt w ere a chie ved , but this was done at the expense of full employment andsocial security efforts. The United States, Canada and the United Kingdom underth e T ha tch er go vernm ent belong to this group. The overall policy stance taken bythe authorities in Sweden, Denmark and Luxembourg is completely different.Aprimacy of welfare sta te policies- at the expenseof balanced econom ic growth and


7/26

TABLE I. Economic performance and policy outcomes in 23 OECD nations 1974-80

GrowthLabo ur of social

Ran k-order market security

of performance expenditureeconomic and ratesof in percentage Public deb tperformance unemployment pointsof G D P ) a s 7 of G D P1979-80 1973-802 1973-793 1973 and 19794

Australia 8Belgium 10Fed eral Republ icof 4

G e r m a n y D e n m a r k 15Finland 12France 11G r e e c e 14United Kingdom 17Ireland 13Iceland 22Italy 18J a p a n 1C a n a d a 6LuxembourgNew Zealand 19Nether lands 9Norway 3Austria 5Portugal 21

Sweden 16Switzerland 7Spain 20U S A 2

average (4.5)bad (6.8)bad (3.2)

average (6.0)average (4.6)bad (4.8)good ( 1.9)bad (5.4)bad (7.0)5good (0.5)bad (6.7)good (1.9)average (7.2)good 0.5)5good ( 1.9)average (4.1)good (1 3good(1.8)bad (7.0)5

good (1.9)average (0.5)bad (6.1)average (6.7)

2.65.72.7

4.71.64.92.12.52.1

-1.50.35.80.87.5

7.11.62.75.4

5.72.52.51.6

53 6118 29

16 366 15

13 16

73 62660 91

33 6913 36

28 25

43 4433 5317 31

36 5025 30

49 45

Based on Euromoney,1981 (rank 1 = best performance,rank22 = worst performance ). S ee footnote2.Da ta in parentheses are ra tesof unemployment (average1974 o 1980). Bad mean s a rateof unemploy-men t which is greater than2% an d a decline in the levelof employment between1973 and 1980. Goodme ans a low rateof unem ploym ent (less than2%) and a n increasein the levelof employment. Averagemea ns either a low rateof unemp loyment and a decrease in the levelof employm ent (for examp le, theSwiss case) or a high rateof unemp loyment and a n increase in the level of employment (for examp le, theUnited States). Source: see footnote3.See foo tno te4.See footnote 5.Estimate.Relative decrease between1974 and 1979 (see Ridley, 1979,1004).

competitiveness - s typical for these cases. Luxembourg, Denmark and Sweden arenations in which the relative political weight of the welfare state bureaucracy and thewelfare state clientele is relatively strong. Hence, it is not too surprising that thecourse and the content of policy-making in periods of economic crisis were stronglyinfluenced by the entrenched interest of the welfare state class. On the other hand, a

strong welfare state class does not automatically produce an expansive social securitypolicy process at the cost of the accumulation process. For example, in otherdeveloped welfare states (the Netherlands, West Germany and Finland) the policystance chosen by the public authorities was quite different from the Swedish one. Themain thrust was on the modernization of the economy and a modest growth of the


8/26

8welfa re state, with a mo derate expansionof the public de bt . Th e overall costs of thisstrategy were clearly visible in the labour market: the rates of unemploymentincreased very strongly. Rising unemployment was alleviatedto some extent byrelatively g en ero us unemp loymen t insurance funds. However, the absolute number

of un em plo ye d a nd th e fiscal crisisof the sta te tended to undermine the monetarybase of th e u nem ployme nt insurance funds. Frequently, and to an increasing extent,the provision of unemployment benefits was applied much more restrictively andwith a n inc rea se in social control.

T h e Swiss case stand s apart from the othe r states inso far as the good econom icperformance and the extremely low rate of unemployment hides an enormouslystrong reduction in the labour force, which took place at the expense of foreignw ork ers. Po rtug al experienced a strong increase in social security expenditures and abad economic record with low growth, high ratesof inflation and high ratesofunemployment. Elsewhere, it has been pointed out that the expansionof socialsec urity pro gra m m es was the com bined result of the consequencesof the revolution,immigration from the former colonies, and the economic depression(OECDEc on om ic Surveys: P ortugal, 1981).

3.2 The Political Composition of Governments Does Not Make a Difference

These d ata provoke a hostof questions: why d o the nations varyso widely in the waythey resp on de d to the setback to economic growth in the mid-l970s? How can weaccount for th e difference b etween th e muddling through nations and those stateswhich w ere m or e o r less successful in their efforts to control the economy politically?And to what extent is the political compositionof government a variable whichexp lains obs ervab le differences in policy o utpu ts and outcomes?

C on tra ry t o w hat many political scientists and politicians believe, and contrary tom an y findings in stu di es on policy-making in the 1960s and early 197Os, there is noclear-cut relationship between the partisan compositionof government and ourindicators of social and economic performance.In other words, there is not muchsu pp or t f or th e view tha t co untries under social-democratic rule produced policiesth at we re co nsistently different from those with bourgeois governments (see Figure3). o N or is ther e m uch evidence for the view that the size and unityof th e partiesofth e right (Castles, 1982b) prod uce clear-cut policy differences.

T his is no t t o say tha t th e political complexionof governments does not m atter forpolicy formation processes. The crucial point is, however, first that the politicalcomposit ion of governments produces different policies if, and only if, someadditional conditions are fulfilled, and, secondly, that governments with ideo-logically widely divergent parties in power produce equi-functional policy outputsand outcomes if, and only if, additional conditions are met.Of these additionalconditions, it is the structureof class politics arrangements, the structure of thepolitical cu ltu re, t he econom ic strength and levels of econom ic productivity in theper iod be fo re th e crisis, which make a difference- argely, though not exclusively,

irrespectiveof which pa rty is in power.E co no m ic perform ance after the first oil shock: levelsof productivity priorto the cr kkand extra-parliamentary distributions of power make a difference. The economicper formance of the industrialized democracies after the first oil shock is largely


9/26

9

independent of the size and the growth rates of the public economy and the welfarestate. There is not much evidence for the view held by the new fiscal orthodoxyaccording to which the economic miseries were the product of government, inparticular the rapid expansion and the excessive levels of government spending (see

additionally Cameron, 1981). Nor is there much evidence for the view that the blamefor market failure ungovernability problems can be laid at the door of either left-wingo r right-wing governments.

The explanation for the differential performance of the capitalist democracies hasto be sought elsewhere. It is to be found in a number of economic, socio-cultural andpolitical variables that our empirical analysis demonstrates to be of substantialimportance in determining the performance of the economies. The most successfuleconomies were those which were economically powerful and economically highlyproductive before the world-wide recession began; and in which the dominance ofthe bourgeois tendency in the sphere of production was in harmony with the power

distribution in the party system and in industrial arenas or in which powerful, butideologically moderate, trade unions, whose strength was institutionalized in acorporatist mode of class conflict regulation, pursued moderate wage policies whichwere compatible with accumulation requirements on the part of capital.

On the other side, states with an economy haunted by low growth rates, high ratesof unemployment and inflation have, in general, been characterized by a low weightof the economy on the world market and relatively low levels of productivity beforethe crisis began; and by a system of industrial relations characterized by a fragmentedstructure, an authoritarian or competitive mode of regulating the class conflict, andfrequent labour conflicts. In these states, the economic resource base was too weakto allow a peaceful, gradual and economically successful restructuring of theeconomy.

La bo ur m arkets and rates of unemployment: the impactof productivity corporatismsolidaristic and paternalistic values. The performance of the labour market, thecontent of labour market policy and the rates of unemployment are also the result ofthe combined influence of a strong and productive economy in the period before thecrisis began, a more or less corporatist regulation of class conflict and, additionally,of a set of solidaristic values shared by the public and the authorities.

Labour market developments which tend to favour organized labour (low rates ofunemployment and a constant or increasing level of civilian employment) werecharacteristic of Japan and some of the smaller states (Sweden, Norway, Austria,New Zealand, Luxembourg and Iceland, and - f the figures do not lie- Greece).Sweden, Norway and Austria are nations in which a strong corporatist mode of classconflict regulation with powerful trade unions and a powerful and united SocialDemocratic left-wing tendency prevails. The bargaining procedures between thestate , labour unions and organized business within the context of an extra-parliamentary balance of class forces and (in Noyay and Austria) under SocialDemocratic rule resulted in active demand management policies and a labour market

policy committed to full employment.Although the strength of the trade unions and the degree to which they areincorporated in the policy-making structure is lower in New Zealand, Luxembourgand Iceland than in the nations mentioned above, the governments in these smallstates were more or less strongly committed to full employment. The explanation for


10/26

10

-2 0 sC 2 5

D l

o c o . r

uiv

o r -a , + 3c 1

. Ice.

.Por t .

.NZ

- 1 t a l . . u K-Swe. . en. I r e . . i n .

.Bel .Fra .

Can. Swi.

s. RG .Aus t r ia

*Nor.

* J a p .

Aust ..Net. .Fin. . RG

-Nor.*A u s t r i a

I c e .'Gre . . Z *Swe..,Tap. OSwi. -Lux.

1 2 4 ;Bo urgeo i s B a l an ce So c i a l De mocr a ti c

hege- domi- domi- hege-

mony na nc e na nc e mony

31 2 I 5Bo urgeo i s B a l an ce So c i a l De mocr a ti c

heg e- domi- domi- heg e-mony na nc e na nc e mony

. N e t . Lux-

i *Bel. Swe.' P o r t .

.Fra. . en.. us t FRG . u s t r i a*Swi. .UKI r e .

'UScan Fin .

. I t a l .

. or.-1 . Ice.

1 2 3 1 5

Bourgo i s Ba l ance S o c i a l Demo cr a ti cheg e- domi- domi- heg e-mony na nc e na nc e mony

Fig. 3 . Policy out com es and the political composition of governments 1974-79/80


11/26

11

15

10

5

I t a lI r e .

re.I t a l .

UK

2;: ' A u s t . en.: an. . B e l g . e w e . .Nor.US

. N e t .L u x

wi . A u s t r i aFRG.

ap.

. en.we .5e ;g . .F in- .FRGLUX.

6 W . l .. r a . +Jet'us

. K

. N o r .

. A u s t r i a

c

1 2 3 4 5B o u r g e o is B a l a n c e S o c i a l D e m o c r a t ic

domi - h e g e -nance mony

hege- domi-

mony nance

1

2 3 4 5B o u r g e o is B a l a n c e S o c i a l D e m o c r a t ic

domi- hege-nance mony

heqe- domi-mony nance

~ S w e . .Den.8 Lux.

.Net.

; p

w e UK

&u s i aY st* in . *FRG

: 55:. r a .* s p a .. a n .5

0 .usI c e .

2 3 4 51

B o u r g e o is B a l a n c e S o c i a l D e m o c r a t ic

hege - domi- domi- hege -mony na nc e na nc e mony


12/26

12

the se cas es h as t o t ak e socio-cultural an d political concomitants of small states intoaccount .A high densityof com munication between the Clites, a strongly interlockingnetwork of decision-m aking structure s, and a political culture in which solidaristicvalues a re dee ply em be dd ed , seem to be important explanatory variables.

In the Japa nese c ase, the patternsof a paternalistic capitalism held the rate ofunemployment below the average level in the OECD area. The extremely lowun em plo ym en t figures in Sw itzerland hide the dramatic decrease in civilian employ-m en t be twe en 1973 an d 1980, which took place at the costof the foreign labour forceand th e d iscouraged female w orkers and elder workers covered by early retirementschemes.

Sp ain , Portug al, Ireland , Belgium, France, Gre at Britain and West Germany werestrongly hit by the labour market crisis. Rates of unemployment went up and thelevel of civilian employment declinedor stagnated. T he effortsto modernize theeconomy via a mixture of supply-side and demand-orientated crisis managementw ere in t he se co un tries strongly facilitated by the weaknessof the trade unionsand /o r by th e unions mo de rate behaviour (whose consitutencies were not yetaffected by the lab ou r ma rket crisis in the same way as were the marginal groupsonthe labour m arket (see ,for exam ple, Armingeo n, 1981)), andby the peace-keepingfunctionsof th e unemploy ment insurance system.

Determinants of social security expenditure growth in periods of economic crisis:built-in stabilizers dominant parties corporatism openness of he econom y and theclientele of the welfare state. When we come to examine the determinants of thech an g e in soc ial security exp end iture between 1973 and 1979 (the latest year forwhich co m pa ra ble d at a were available), we are confronted with the problem that asubstantial part of welfare state measures automatically increase whenever theecon om y is depressed. Thisis because in periodsof econom ic crisis, unemploymen tinsu ran ce, social assistance an d early retirement schemes expand the costsof socialsec urity , bot h in a bsolute term s and relative to G D P . Thus, it is hardly surprisingtha t, o verall, nation s with low levelsof economic growth tended to expand socialsecurity exp enditures at a faster pace than nations with a more resilient e c ~ n o m y . ~Ho we ver, a n examinationof th e residuals suggests additional insights into the policyprocesses that characterized the crisis period. Once we have accounted for thepropor t ion of the variance that is explained by economic growth, there remainimportant political factors at work. The expansionof social security expenditureswa s, o ve rall, m ark edl y overproportionate in nations in which the Social Democraticparty was th e c en tre of gravitationof the party system (and thus in a position where itcould strongly influence and constrain the room to manoeuvre by its bourgeoisop po ne nts) ; in which th e regulationof the class conflict was corporatist in na ture(and thus increased the probability of equal-exchange-bargaining proceduresbe tw ee n t he sta te, t he unions an d organized business); in which the age structureofth e p op ula tion ha d e stablished a numerically and politically important stratumof oldage pensio ners to whose dema nds governments were forced or more or less

automatically willing- o respond in their social security policies; and in which thedom estic repercussionsof a n open economy required an active social security policys tance on the par t of the governm ent (Pearsons r with imports and exports as apercentage of G D P in 1972 is .37 in 1972 and .43 in 1979).

A m on g t he OE CD -nation s two different strategies for the expansionof social


13/26

13

security expenditures were observable: first, the politics of welfare state expansiono n the basis of a highly developed welfare state (Belgium, the Netherlands, Sweden,Denmark and Luxembourg are the most notable examples); and second, a strategyof expansion which may aptly be described as the catching-up phenomenon. Japan

and Portugal are the most important examples for this developmental path. Thereasons are idiosyncratic in character. The major reason in the Japanese case seemsto be the fact that the world-wide recession had undermined the privately organizedwelfare provision schemes and that the period of rapid economic growth in the 1960shad produced high and politically dangerous social costs. The Liberal Governmentsought to restrain the consequences of the crisis through the expansion of stateactivities in environmental pollution control, in labour market schemes, and in socialsecurity efforts (Lorcher, 1981; Foljanty-Jost, 1981; Ike, 1981). The expansion ofsocial security provisions in Portugal was also related to a catching-up phenomenon.The major background factors in the Portuguese case were clearly the consequences

of the regime change in the mid-l970s, the requirement to integrate the immigrantsfrom the former colonies, and the effect of the economic depression which tended toinflate the expenditures on social security programmes.

Determinants of economic and political performance in the industrial democracies.Our empirical analysis lends support to the view that the following conditions arefavourable to a pattern of balanced economic development, a relatively good labourmarket record with low unemployment, and a further expansion of the welfare state- even under restrictive circumstances:

a distribution of power in industrial relations which comes close to a balance ofclass forces. Under this condition no class is strong enough to dominate itsopponent and thus the probability of an equal-exchange bargaining betweencapital and labour increases;

highly organized, ideologically united and moderate trade unions whose power isinstitutionalized in corporatist arrangements;

- a set of socio-cultural values which stresses the importance of solidaristic(re-)distribution;

- an economy which had the privilege of entering the period of economic crisis with

productivity levels above the average;a small state not primarily because of the sheer impact of size but primarilybecause of the solidaristic values embedded in the political culture and the moredensely integrated communication network between the elites.On the other hand our empirical analysis would lead us to argue that economic and

political performances tend to be worse (that is, low economic growth, high inflation,high levels of unemployment, and a low rate of welfare state expansion) if thefollowing conditions are met:

an authoritarian or competitive mode of regulating the class conflict (within thiscontext distributional conflicts between labour and capital occur more frequentlyand intensely, and inhibit both the successful restructuring of the economy and theexpansion of a welfare state);a dominant position of the bourgeois tendency both in industrial relations and inthe party system;


14/26

14

a political culture which is strongly influenced by competition and possessiveindividualism;

- an econ om y which entered the periodof economic crisis with comparatively lowprodu ctivity levels;a large stat e with a low deg reeof interlocking communication and decision-makingstructures.Th ese factors sha pe the con tent and the impactof public policies to a large extent,

tho ug h not exclusively, irrespectiveof which party is in power.

4. THE IMPACT OF POLICY ON ELECTION OUTCOMES:PARADOXES OF MUDDLING THROUGH AND SUCCESSFUL

CRISIS-MANAGE MEN T POLICIES

It is a w ell-established fact that governments ar e fa r from being altruistic providersofpublic goods. T h e conten t and the timingof gove rnmental policy-making is quitefre qu en tly sh ap ed by th e electoral calendar and th e effort to increase the probabilityof re-election. The literature on governability and legitimation problems in latecapitalist societies has been guidedby the assumption that a government withasuccessful and favourable record inits social and economic policies would bepolitically rewarded. Conversely, a bad record would fuel social conflicts andgrad ually un de rm ine political suppo rt for the governing parties. W e will argue thatthis assumption is not compatible with what we empirically observe in Western

nations.O u r c on ten tion is tha t th e relationship between th e degree of policy effectivenesso n th e o n e ha nd , a nd political support and social peace on the othe r, should betrea ted as a variab le. Ou r contention is based ontwo interrelated empirical obser-vations. First, we can observe across the industrial democracies that politicalreactions to high a nd increasing ratesof unemploym ent and decreasing margins forredistribution of reso urce s have been less sh arp than expected. Undoub tedly, this ismainly due to higher levelsof income protection and to the fact that long-termun em plo ym en t is falling to a large extent on weak and unorganized groups in society.But i t may also be the case that to some extent attitudes to work, leisure and

emp loym ent are changing,so tha t less employ men t security in the formal economy isacc epte d by som e in return fo r greater leisureor for work in the informal economy.Sec on d, th e reward-punishment hypothesis is called into question by the results

of a re ce nt stu dy con du cted by the a uth or on the impact of the electoral calendar onec on om ic policies an d econom ic outco m es, and on the extent to which the state ofthe econom y determines the fa teof governing and opposition parties in elections(Schmidt , 1983). For ex am ple , governing parties which were, by and large, success-ful in ste erin g the e con om y and generous in welfare spending efforts were quite oftenpunished. Th e proport ionof the vote they mobilized declined below a thresholdwhich was critical for re-election. A nd more often than no t, governing parties which

ha d a n u nfa vou rable record successfully survived an election. O ne can thus arguethat four different combinations of policy effectiveness and reward-punishmentou tco m es are to b e observ ed among Western nations (see Figure4).

The most interesting cases are in the upper right-hand cell (it is here that aparadox of a successful welfare state appears) and in the lower left-hand cell (here


15/26

15

we observe the paradox of muddling through). The elections in Sweden in 1976 andNorway in 1981 might be taken as examples of the paradox of a successful welfarestate. Th e active labour market and welfare state policy stance taken by the SocialDemocrat ic governments in these countries was - udged by the aggregate outcome

of the vote punished. One can plausibly attribute this result to the impact oflong-term partisan forces and short-term non-partisan variables, with post-material-istic demands being on the whole conducive to a punishment effect. The elections inItaly in 1979 and West Germany in 1981 might be taken as examples of the paradoxof muddling through: the policy records of the government were, in balance and ingeneral, unfavourable for re-election prospects, and much worse than the records ofprevious governments, yet the governing parties were nevertheless re-elected. As faras the West German election of 1980 is concerned, it seems as if the candidate of the

Japan 1979Austria 1975,1979

Federal RepublicofGermany 1980

Italy 1979

High policyeffectiveness

Low policy effectiveness4

Election results

Reward Punishment

Sweden 1976Norway 1981

UnitedKingdom 1979

USA 1980

I Re-elec t ion .Turnov er in political comp osition of governments.Successful political contro l of the econom y measured by our indicators for econo mic performance,labour mark et dev elopm ents, and social security policy).Partially suc cessful or unsuccessful political control of the econ om y.

Fig. 4. Policy perform ance, reward and punishment.

Christian-Democratic party (Franz Josef Strauss) polarized the electorate to such anextent that even those voters who were very critical of, and disappointed with, theSocial Democrat/Free Democrat coalition government still voted for one of thegovernment parties (for example a large proportion of green voters).

It is not o ur intention to investigate fully the determinants of voting behaviour inthe Western nation^.'^ Our focus is on the aggregate outcome of the vote. In thisrespect we ar e confronted with a politically highly explosive paradox. The govern-ments of the Western nations are to an increasing extent faced with a dilemma: the

probability of re-election and the probability of gaining or losing votes is some 50 ,largely regardless of the economic performance, and largely irrespective of a good ora bad record of economic and social policy-making. We are thus inclined to arguethat governments which take this equation into account content themselves withmuddling-through policies.


16/26

16

5. CONCLUSION: THE STATE OF THE ART REVISITED

In the introduction to this article we outlined someof the competing explanationsth at fe at ur e in th e literatu re on public policy in advanced industrial democracies.Tocon clud e o u r p resen tation we shall summarize briefly what ou r analysis contributesto o ur understandingof th e validity of these explanations.1. One general position in the literature is that increasing state intervention hascreated the basis for successful political control of the economy (for example,Ste ph en s, 1979). Th is view has been opposedby Marxist writers who have taken thepos ition tha t t he policies ado pted in industrial democracies were unable to cope withth e str uctu rally given instabilitiesof the cap italist orde r (for example, Kirchheimer,1981; Offe, 1972; Ronge, 1977). Our analysis does not confirm eitherof theseex tre m e positions. R at he r, th e evidence seems to point to what may be described as apartial political control hypothesis. The imbalance of capitalist economies is

am ena ble t o varying strategiesof more or less successful political contro l in certainar ea s n particu lar, interventions to expand the size of the tax and welfare state andt o limit la bo ur m arke t fluctuations. However, the successof intervention in on e areado es n ot usually spill ov er into all othe r aspectsof policy. Inde ed , in the majority ofthe OECD na tions th er e se em s to b e an empirically identifiable tendency for there tobe a trade-off by which success in one area limits successful control elsewhere.Because the au tonomyof th e capitalist economy remains considerable in all thesenations, governments must necessarily order their priorities in the field of macro-eco no mic policy.

Ho w ev er, th e or de r of the priorities varies widely among theOECD nations. For

exa m ple, Sw eden might be tak en a s an example of the primacyof a welfare state anda c om mitm ent to ful l employment at the expenseof a sound capitalist accumulationprocess. Norway and Austria are examplesof a political stalemate betweenbo urg eo is an d socialist forces both in the industrial and the party systems arenas ofpo w er , w ith th e socialist gov ernm ents being o n th e whole quite successful in recon-ciling the requirem entsof a capitalist econom y with dem ands for full employm entan d the deliveryof m or e o r less generous welfare provisions. Japan is unique inso aras a bourgeois hegemony both in the political superstructure and the politicalsubstructure produced a sound economic development, low unemployment levels,an d a rem ark ab le expansion in social security expenditure. However, these nationswill probably have to face increasing difficulties in maintaining low levels ofunem ploym ent over the long run. I tis very likely that the tightening of demand-ma nag em ent in the othe r m ajor O E C D economies and, in particular, the increasingavailability of extremely productive, ch eap and labour-saving new technologies willstrongly, and to an increasing extent, undermine the efforts to keep the rateofunemploym ent down.2. A no th er pos ition, frequently adduced in the literature, is that the political com-position of go ve rnm en ts will be the decisive factor in determining the prioritiesofgo ve rnm en tal policy-making. In o ur analysis the evidence for this party control does

m atte r h yp othe sis is we ak. T he re is no clear-cut dividing line between the policiesad op te d by Social Dem ocra tic and bourgeois governments. This is not to say thatpar ty control of the reins of power is irrelevant for social and economic policy andpolicy outcomes. Rather, our analysis repeatedly demonstrated that party controlwas of importan ceif ot he r conditions are m et, as in the pronounced tendency for a


17/26

17

st rong combination of a balance of class forces in the industrial arena and strongSocial Democratic government to lead to active labour market policy. The real issueis not whether party control matters, but rather to what extent and, in particular, inwhat circumstances it matters.

3. On the whole, those explanations which rest on the primacy of economic factorsare t o some extent compatible with the data. However, the crucial point that emergesis that in many capitalist democracies the links between the economy and publicpolicy are, to a varying extent and in a variety of ways, mediated by interveningsocietal and political factors. For example, the setback to economic growth in themid-l970s, by itself, does not account for the differential performance of the OECDnations in the late 1970s and early 1980s, though we frequently had the opportunityto demonstrate that the level of economic productivity in the period before the crisisbegan crucially influenced the choices open to governments.

4. Perhaps the most interesting and important finding is the degree to which thosevariables that intervene between the economy and public policy are of the kindsuggested by the class politics paradigm. The view that extra-parliamentary politics isan important determinant of policy is supported by our analysis. The strength oftrade unions relative to the strength of organized business, the mode of regulatingthe class conflict, the degree of harmony between the political superstructure and thepolitical substructure in both party system and industrial arenas, and the degree towhich solidaristic values characterize the political culture seem to be vital factors,taken in conjunction with economic variables, in determining the success or failure ofpolitical control of the economy.

5 . The hypothesis of ungovernability in democratic states is generally not confirmedby our analysis. Up to the present a clear tendency towards ungovernability seems tobe th e exception rather than the rule. It is only in those nations where a set of otherconditions is met (for example, an economy which was weak before the crisis began;a fragmented system of industrial relations; weak trade unions) that the probabilityof governability problems increases dramatically. On the other hand, there is someevidence fo r the result of Phillippe Schmitters empirical contribution to the study ofgovernability problems (Schmitter 1981). It seems as if nations under SocialDemocratic rule and, additionally and probably more important, nations with abalance of class force in the industrial and party-system arenas are more governablethan their bourgeois-dominated counterparts.

6. Our partial-political-control hypothesis indicates that governments which aresupported by a stalemate in class conflict arenas (and, in addition, governmentswhich act within the context of a paternalistic capitalism) were, up to the present,able to handle the problems associated with the instability inherent in a capitalistor de r. This is not to say that these governments will in the future automatically beable to continue the policy stance they adopted in the mid-1970s. Indeed, there issome evidence for the view that the low unemployment nations (Austria and Swedenmight be taken as examples) have reached the upper limits of successful political

control of the economy, with unemployment rates in the early 1980s being on thewhole on the increase. This is a not too surprising observation, since the monetaryand fiscal policy stance of the major OECD nations has tended to be more restrictivein character after the second oil shock than it was before, and since labour-savingnew technology is on the advance.


18/26

18

Our analysis would, however, leadus to believe that old issues, that is, issuesassociated with the n atureof a capitalist economy and the dynamicsof latent andmanifest conflicts between labour, capital and the welfare state clientele, can inprinciple b e so lved . H ow eve r, th e most im portant barriers which strongly restrict the

room t o m ano euv re f or this policy are not only economic and technological, butalsopolitical in character, with the paradoxof the successful welfare state and thepar ado x of muddling-through being on the whole oneof the m ajor impediments.

O u r study would lead us to argue that the futureof policy-making, policy outputsan d out com es is dismal in chara cter. Paradoxical dynamicsof party competition andelection ou tcom es, the long-term consequencesof a strong expansion of the welfarestate and the policy process inertia, combined with the momentum of a capitalisto rd er an d , in par ticu lar, th e labour-saving consequences of new technology, have,unintentionally, underm ined the foundationson which the policy process of the1950s a n d 1960s was based.

NOTES1 This paper summarizes some of the results of an ongoingstudy on determinants and consequences of

policy responses to old and new issues in industrial democracies. This research project is a follow-upto the authors studies on comparative partisan differences in public policy (Schmidt, 1980, 1982a.1982b, 1982c), and it is associated with the Future of Party Government project at the EuropeanUniversity Institute.

2 Euromoney, 1981. The summary of economic performance is based on five measures over the period1974 to early 1981: the rate of real growth, the ra te of consumer price inflation, the strength of thecurrency in terms of Special Drawing Rights, the average sue of the current account relative to GNP,and the strength of the export sector. The strength of the export sector was measured by whetherexport s have grown faster than GDP. The statistical method used to combine each countrys five scoresinto one overall score was a principal component analysis. The overall index combines all theinformation on each country. It works by giving a weight to each measure. A good growth score isgiven a weight of 0.45. A poor inflation score is given a weight of -0.55. A strong exchange rate has aweight of 0.52. a good export performance has a weight of 0.49, and a current account surplus has asmall weight of 0.05.

3 The da ta were taken from the OECD Economic Outlook and the OECD Labour Market Statisticsseries.

4 Source: OECD National Accounzs 296 1979 (Paris: OECD, 1981). Table 9, rows 17-20. The datainclude social security benefits, social assistance grants, current transfers to private non-profit institu-tions serving households, and unfunded employee welfare benefits as a percentage of GDP.

Comparable data which include government final consumption expenditures for welfare purposeswere not yet available for all nations under investigation. The data on social security expenditurecorrelate strongly with the data on disbursements of general government (data base: OECD EconomicOutlook 31, 1981) which were separately analysed (see the results of the statistical analysis inAppendix 3).Data were taken from the Statistische lahrbuch der Eundesrepublik Deutschland edited by theStatistische Bundesamt in Wiesbaden (West Germany). The data are preliminary.

6 In general, the data on economic and socioeconomic variables were taken from OECD publications(in particular, OECD Economic Outlook, OECD National Accounts, Main Economic Indicators,Labour Force Statistics and OECD Economic Surveys). Political data were based on Schmidt 1982aand Keesings Archiv der G egenwarr.

7 The classification is based on the left-wing parties and the bourgeois parties share of cabinet seatswithin a given period o n a monthly base. The original values were converted into a rank-order scale.The scale runs from bourgeois hegemony (the bourgeois parties share of cabinet seats equals 100 )through bourgeois dominance (their share ranges from greater than two-thirds to less than 100 ).balance (t he bourgeois parties share and the left-wing parties share is larger than one-third andequal to or less than two-thirds of the total of cabinet seats), Social Democratic dominance(analogous to bourgeois dominance). to Social Democratic hegemony (the share of cabinet seats forthe left-wing parties is equal t o 100%).


19/26

19

8 The selection of the variables is basedon their explanatory power in a cross-national studyof 21cap italist dem ocra cies (Schmidt 1982a). Five indicators were supposed to represent conditions whichare favo urab le t o bourgeois policies: a proportionof the vo te for bou rgeois parties which is greaterthan 50%; a left-wing tenden cy which is organizationally an d ideologically split (i.e. whenevernon -Soc ial-D em ocra tic left-wing parties score more than10% in the majorityof the elections within a

given perio d and the latest election before this period); a bourgeois tendency which is organizationallyand ideologically cohesive (i.e.if not more than one bourgeois party scores 10 of the vote in themajority of the elec tions within a given period, inclusiveof the latest election before this period); aunio n d ensity sco re which is less than50%; an d a tra de u nion move ment wh ich is politically split alongreligio us, eth nic and political cleavages).

9 See O E C D Economic Outlook (Paris: OE CD , bi-annually).10 See the results of the statistical analysis in Appen dix 2. T he d ata confirm a hypothesis which von

Be ym e (198 1) has recently adv anced (and they contradict the inference which Cam eron (1982) hasdra w n fro m his study of comparative econ omic policy-making in the 1970s).

11 Th e interpretation is basedo n the O E C D Economic Surveys andon a statistical analysis of the da ta(see Ap pendix 3).

12 A corporatist mode of regulating the class conflict is the label for those nations in which trad e unions,em plo ye rs associations and the sta te are com mitted to a social partnership ideology, and collaborate inth e policy -forma tion proce ss in som e policy are as. Empirical correlates of a corporatist mode are: verylow str ike v olum e; typically, though not exclusively, the absenceof an authoritarian incomes policyenacted from above; and an ideologically and organizationally united trade union movement. Anauthoritarian mode of regulating the class conflict is characterized by co-operation be tween the sta teand the entre pren eurs against the trade unions.A competitive mode is characterized by a low degreeofinterfe renc e in ind ustrial relationson the part of the state. In this study we con tent ourselves with acorporatism indicator which runs from strong to medium and weak.For the meaning of strongcorporatism see above: weak corresponds to a varying mixture of authoritarian and competitivemodes of conflict regulation while mediumis a residual category.

13 This interpretation is based on a wide rangeof more detailed one-country studies(for example,D ah re nd or f, 1981; Scharp f, 1981; the O E C D EconomicSurveys) and the resultsof astatistical analysis(see Ap pendix 3).

14 Da ta base: see Appendices 1 and3. If on e were to analyse the expansionof the total state expend ituresbetw een 1973 an d 1979. the results would basically be identical. A m inor difference should be pointedou t: the growth ratesof disbursements of governm ent co rrelate weakly with an additional variable-the union den sity measure r = - .35).

15 A fasc inatin g account h as recently been advanced by Budge and Farlie (1982).


20/26

20

APPENDIX 1. DATA SELECTION)

-0

AustraliaBelgiumFederal Republicof GermanyDenmarkFinlandFranceGreeceUnited KingdomIrelandIcelandItalyJ a p a nC a n a d aLuxembourgNew ZealandNetherlandsNorwayAustriaPortugalSwedenSwitzerland

SpainU S A

Case-N123456789

101112131415161718192021

2223

V0138.00

10.ocl4.00

15.0012.0011 oo14.0017.0013.0022.0018.00

1 OO6.00

19.009.003.005.00

2 1. x)16.007.00

20.002.00

V02611.908.204.80

11.0012.6011.1017.5016.0015.5043.1017.109.909.307.20

14.307.109.006.30

22.7010.304.00

17.909.30

V029 V0332.00 1.003.00 0.002.00 0.003.00 1.002.00 0.003.00 0.001.00 2.003.00 0.003.00 0.001.00 3.003.00 0.001.00 2.003.00 1.001.00 2.001.00 2.002.00 1.001.00 2.001.00 2.003.00 0.001.00 3.001.00 2.00

3.00 0.003.00 1.00

VO46

53.0018.0016.006.00

13.00

73.0060.00

33.0013.00

28.00

43.0033.0017.00

36.0025.00

49.00

-

-

-

-

-

-

V048

43.007.005.004.008.00

54.0060.00

5.0032.0012.00

23.00

23.0025.0010.00

41.005.00

37.00

-

-

-

-

-

Spearmans rank order correlation coefficientr,. The major reason for the selectionof this coefficientwa s pragm atic in ch arac ter: first, one can easily understand the logicof this coefficient, and sec ond, itsvalues can m ore easily be com pare d with Pearsonsr . All oth er coefficients are Pearsonsr .

Number of cases: 22 (com para ble data fo r New Zealand w ere not available).N = 16 (com parable data were not availablefor Australia, Greece, Iceland, Canada, New Zealand,T h e ta ble includes o nly those correlation coefficients which are larger than0.3 (significance level with23cases a t some0.09).

Portugal and Spain).


21/26

S

as

u

ye

n

u

g

wh1

7

np

na

p

n

oGDP

P

c

cmp

o

o

g

nmen

1

8

1

b

He

mo

5 SOCDem.He

mo

Uno

dn

y

-

>5 ,1

6

)

S

nr

u

o

mo

men

1

sp

1

e

s

Co

vn

o

le

wn

en

n

- 1

sr

1

w

Co

vn

ob

s

ten

n

-1

we

1

sr

GNP1

(U

d

a

P

ao 973

1OOO)

C

p

sm

I

we

2

medum,

3

sr


22/26

22APPENDIX 2. CORRELATIONS BETWEEN MAJOR DEPENDENT ANDINDEPENDEN T VARIABLES

EuromoneyLabour market'Social security'Public debt'UnemploymentInflationGovernment'Party'CohesivenessLeft voteCorporatism'GNP 1973Population 1973ProductivityGrowth IGrowth I1Unemployment 1973

Social security 1973Public debt 1973Disbursements 1973Inflation 1966-73ElderDemographyImport export 1972

1.o1 o

1.o

- 85.81 48

29

.45-.40 .63 .34-.48 -.32-.35 -.45-.45 .41

- 29

- 54

-.32.32

.37

1 O1 .o

.31

74.38.47

- 47.43

.65

-.23 .34

.34

.30

1 o- 30

.33

.39

58

- 38

34.37- 30

.85- 36

1 .o1.0

- S O .48 1.0.61 -.77.50 -.54 .35

-.39 .34 .30-.41 -.32

.3430

-.31 .33

.37 .40

.45 -.32 - S O

.59 .44

.51 -.30-.79 .56 .41

1 o.3463

- 54.43

50

.31- 60

VAR IABLE NA MES IN APPENDIX 2Euromoney

(see footnote2).

Labour marketSocial security(first differences) (see footnote 4).Public debtUnemployedInflationGovernmentParty Dominant party in party system ( I means a bourgeois party is dominant, i.e. it is stronger thanthe major left-wing party by at least 5% of the popular vote in the majority of elections from 1974 to 1980(including the latest election before 1974); '0' means 'balance' or 'non-dominant party'. i.e. the criteriastipulated for rank 1 and rank 3 are n o t met; '3' mcans that a left-wing party isdominant (classified in thesame way as rank * l ' ) ,

CohesivenessLeft vote1974.

Euromoney indicator of economic performance 19741980/81 after the first oil price shock

Labour market performance 1971-80 (see note toTable I .Growth in social security expenditure as a percentage of GDP between 1973 and 1979

Change in public debt as a percentage of GDP between 1973 and 1979 (first differences).Average rate of unemployment 1974 to 1980 (percentage of total labour force).

Political composition of national government (average 1974-80). See footnote 7.Consumer Price Index (average 1974 to 1980).

Cohesiveness of the left-wing tendcncy 197440 (see footnote 8).Average of left vote in national elections between 1974 and 1980and the latest election before


23/26

23

1.01 0.97 1.0.32 1 o

-.35 1.0- 37 1.0

52 1 0

.39 - 37 -.31 1.0- 36 .SO .32 1.0.37 .31 .84 .38 1.0

1 o.53 .57 -.34 1.0

- . a .86 .90 .48 .33 - 34 -.42 1.0.36 -.46 -.53 -.37 -.37 .44 .30 .29 -.36 1.0

Corporatism Corporatist modeof regulating th e class conflict (see fo otnote12).G N P1973 Gross N ational Product1973 (US Dollars and current exchange rates).Population 1973 Population 1973 (in 1oooS).ProductivityGrowth I Grow th ra tesof G D P (real terms, average1974-79).Growth I1Unemployed 1973 Rate of unemployment in1973 (as a percentageof labour force).Public debt 1973 Public deb t (general governm ent) as a percentageof G D P 1973.Disbursements Growth of disbursements of general governm ent as a percentageof G DP between 1973a n d 1979 (first diffe renc e). Disbursemen ts mainly consistof governm ent final consumption expenditure,subsidies, social security transfers to households, and intereston public deb t.Inflation 1 S 1 9 7 3Elder Population at the ageof 65 and above a s a percentageof total population in1973.Demography

Level of productivity (GNP 1973/Population1973).

Grow th ratesof G D P (real terms, average1974-80).

Con sum er Price index, average1%7-73.

Chan ge in population at the ageof 65 and above between1973 and 1979 (first differences).


24/26

24

APPENDIX 3. CORR ELATIONS BETWEEN THE POLITICAL COMPOSITIONOF GOVERNMENTS AND INDICATORS OF ECONOMIC PERFORMANCEAND POLICY OUTCOME

Spearmansrank ordercorrelationcoefficient

4

Euromoney-indicator of economic performanceLabour market performance 1974-80Rates of unemployment 1974-80Change in rate of unemployment 1973-80Trend in levels of employment 1973-80 (1 = increase, 0 = else)Rate of inflation 1974-80Change in rate of inflation between 1 9 6 7 3 and 1 9 7 WGrowth of disbursements of government 1974-79Growth of government final consumption expenditure 1974-8WGrowth in social security expenditure 1974-795Increase in public debt 197479(general government)Increase in public debt 1974-794(central government only)

- 012.09

2506

.@I- 30- 34

.24

.282

. l l

,116

.086

See footnote 2.

See footnote 3 to Table I.N = 22.First difference.See footnote 4.N = 16.

REFERENCESAltvater, E. (1981). Der gar nicht diskrete Charme der neoliberalen Konterrevolution, Prokla 11.Andrain, C. F. (1980). Politics and Econ omic P olicy in W estern Democrucies.North Scituate: Duxbury

Press.Armingeon, K. (1981). La Confkdkration des Syndicats Allemands (DGB) Face a la Crise in Klaus

Armingeon et a l . Les Synd icats Europkennes ei la Crise.Grenoble: Presses Universitaires de Grenoble.Armingeon, K. (1982a). Socialist Governments andTrade Unions-The Impact on Wages and Policy, in

Castles (1982a).Arrighi, G. (1981). Der Klassenkampf in Westeuropa im 20. Jahrhundert, in Frobel et al. (1981).Bauer, 0 (1928). Kapitalsherrschaft in der Demokratie, Der KampJ SoziuldemokruiischeMomis

schrift 21.Bauer, 0 . 1970). Das Gleichgewicht der Klassenkrafte, in Hans-Joerg Sandkuhler and Rafael de la

Vega (eds.), Austromarxismus. Frankfurt and Vienna: EVA (first published in 1924).Berger, J. (1981). Wandlungen von Krisenursachen m Wohlfahrtsstaatlichen Kapitalismus, n Moneriire

Restriktionen : D ie Znjlationsbekampfung. Berlin: Argument-Verlag. Argument Sonderband AS 68.von Beyme, K. 1981). Do Parties Matter?, Poliiische Vierteljuhresschrift22.Budge, I. and Farlie, D. J. (1982). Explaining and Predicting Elections: Issue Effects and Party Strategies

Cameron, D. R. (1982). On he Limits of the Public Economy, Annals of the American Academy of

Castles, F. G . (ed.) (1982a). The Zmpuci ofPoliiica Parties. Beverly Hills and London: Sage.Castles, F. G. (1982b). Politics, Public Expenditure and Welfare, in Castles (1982a).

in 23 Democracies. University of Essex mimeo. To be published.

Political and Social Science, No. 459.


25/26

25

Dahrendorf, R. (1979). Lebenschancen. Anlaufe zur sozialen und politischen Theorie. Frankfurt:Suhrkamp.

Dahrendorf, R. (ed.) (1981). Trendwende. Europas Wirtschafr in der Krise. Vienna, Munich, Zurich,New York: Verlag Fritz Molden.

Dror , Y. (1981). Social Policy in a Period of Decrement: A Perspective of Governments, in OECD

(1981).Esser, J. (1983). Gewerkschafren in der Krise. Die Anpassung der deutschen Gewerkschafren n neueWeltmarktbedingungen. Frankfurt: Suhrkamp.

Euromoney (1981). The End of the OPEC Era, in Euromoney, October 1981. London: EuromoneyPublications.

Flora, P. (1979). Krisenbewaltigung oder Krisenerzeugung? Der Wohlfahrtsstaat in historixherPerspektive, pp. 82-136 in J. Matthes (ed.), Sozialer Wandel in Westeuropa. Verhandlungen desSoziologentages n Berlin. Frankfurt and New York: Campus.

Foljanty-Jost, G. Park, S.-J. and Seifert, W. (eds.) (1981). JapansSoziul- und Wirtschajbenmicklung minternationalen Kontext. Frankfurt and New York: Campus.

Frobel, F., Heinrichs, J. and Kreye, 0 . (1981). Krisen in derkapitalistischen Weltokonomie. Reinbek beiHamburg: Rowohlt.

Giersch, H. (ed.) (1981). Macroeconomic Policy for Growth and Stability: A European Perspective.Tubingen: J. C. B. Mohr (Siebeck).

Goldthorpe, J. T. (1978). The Current Inflation: Towards a Sociological Account, pp. 186-213 in F.Hirsch and J. T. Goldthorpe (eds.), The Political Economy o nflation. Cambridge, Mass.: HarvardUniversity Press.

Gould, A. (1981). The Salarized Middle Class in the Corporatist Welfare State, Policy and Politics 9.Hibbs, Jr., D. A. and Fassbender, H., with the assistance of Riven, R. D. (eds.) (1981). Contemporary

Ike, N. (1978). A Theory of Japanese Democracy. Boulder, Col.: Westview Press.Kalecki, M. (1943). Political Aspects of Full Employment, Political Quarterly 14.Kirchheimer, 0 1981). Weimar- und was dann? Analyse einer Verfassung, in 0 Kirchheimer, Politik

und Verfassung. Frankfurt: Suhrkamp (first published in 1930).Korpi, W. (1980). Social Policy and Distributional Conflict in the Capitalist Democracies: A Preliminary

Comparative Framework, West Europeun Politics 3.Lehmbruch, G . and Schmitter, P. C. (eds.) (1982). Panems of Corporatist Policy-Making. Beverly Hillsand London: Sage.

Lehner , F. (1979). Grenzen des Regierens, Eine Studie zur Regierungsproblematik hochindustrialisierterDemokratien Konigstein , Ts Athenaum .

Leibfritz, W. 1980). Die Staatsverschuldung im internationalen Vergleich - einige Aspekte derempirischen Entwicklung, in Probleme der Staatsverschuldung, Beihefre zur Konjunkturpolitik, No.27. Berlin: Duncker und Humblot.

Political Economy. Amsterdam, London, New York: North Holland Publishing Company.

Lorcher, S. (1981). Japans Sozialkapital im internationalen Vergleich, in Foljanty-Jost etal. (1981).Luhmann, N. (1981). Politische Theorie m Wohlfahrtsstaat. Munich, Vienna: Gunter Olzog Verlag.Madsen, H. J. (1981). Partisanship and Macroeconomic Outcomes: A Reconsideration, pp. 269-82 in

Malecki, E. S. (1981). The Capitalist State - Structural Variation and Its Implication for Radical

OE CD (1980). Main Economic Indicators- Historical Statistics 1 1979. Pans: OECD.OECD (1981). The Welfare State in Crisis, An Account of he Conference on Social Policies in the 1980s.

OE CD , Paris, 2 23 October 1980. Paris: OECD.OECD: Economic Outlook (Paris: OECD, biannually).OECD: Economic Surveys (Paris: OECD, annually).OE CD: Labour Force Statistics (Pans: OECD, annually).OE CD (1981b). National Accounts o OECD Member Countries. Paris: OECD.Offe, C . (1972). Strukturprobleme des kapitalisfischen Staates. Frankfurt: Suhrkamp.Offe, C. (1979). I Unregierbarkeit. Zur Renaissance konservativer Krisentheorien, n J. Habermas

(ed.), Stichworte zur Geistigen Situation der Zeit, Volume I: Nafion und Republik. Frankfurt:

Hibbs and Fassbender (1981).

Change, Western Political Quarterly 34.

Suhrkamp.

in H. Ruhle and H. J. Veen (eds.), WachsendeStaatshaushalte.Stuttgart: Bonn Aktuell.Ridley, A. (1979). Die offentlichen Ausgaben in Grollbritannien - Die grollte aller Krisen?, pp. 87-166

Ronge, V. (1977). Forschungspolitik ah Strukturpolitik. Munich: Pipe:Rose, R. (ed.) (1980). Challenge to Governance. Beverly Hills and London: Sage.Sachs, J. D. (1979). Wages, Profits and Macroeconomic Adjustment: A Comparative Study, in

Brookings Papers on Economic Activity, 2.


26/26

26

Scharpf, F. W. (1981). The Political Economy of Inflation and Unemployment n Western Europe: A n

Schmidt, M. G . (1980).CDU und SPD an der Regierung. Ein Vergleich ihrer Politik in den Landern.

Schmidt, M. G. (1982a). Wohlfah rtsstaatlich e Politik unter biirgerlichen und sozialdemokratischen

Schmidt, M. G. (1982b). Die Politik der CD U/C SU - und der SPD-Regierungen, inJ . Raschke (ed.),

Schmid t, M. G . (198 2~ ) .The Roleof the Parties in Shaping M acroeconom ic Policy, in Castles (1982a).Sch mid t, M. G . (1983). Politische Ko njunkturzyklen. Ein internationaler Vergleich, inM . Kaase and

H. D. Klingemann (eds.), Wahlen undpo litisches System.Opladen: Westdeutscher Verlag.Sch m itter , P. C . (1981). I ntere st Intermediation and Regime Governability in Contemporary Western

Europe and North America, inS. Berger (ed.), Organ izing Interests in Western Europe: PluralismCorp oratism and the Transformation of Politics. Cam bridge, Mass.: H arvard University Press.

Schmitter, P. C. and Lehmbruch,G . (eds.) (1980). Trends Toward Corporatist Intermediation. BeverlyHills and London: Sage.

Statistisches Bundesarnt (1973ff.).Statistisches Jahrbuch der Bundesrepublik Deurschland. Stuttgart:Kohlhammer.

Stephens, J. D. (1979).The Transition to Socialism. London : MacM illan.Stephens, J . D . (1981). Impasse and Breakthrough- n Sweden,Dissent Summer.Tufte , E . R . (1978).Political Control of th e Econ omy. Princeton: Princ eton University Press.Wilensky, H . L. (1981). D emocratic Corpo ratism. Consensus and Social Policy: Reflections on Changing

Wit tmann, W. (1981). Der Steu erstaat am W endepunkt,Frankfurter Allgemeine Zeitung No. 229 ,3

Outline. Berlin: Wissenschaftszentrum, Discussion papers IIM/LMP 81-21.

Frankfurt and New York: C ampus.

Regierungen . Ein internationaler Vergleich.Frankfurt and New York: Campus.

Burger und Parteien. Bon n: Bundeszentralef i ir politische Bildung.

Values a nd the Crisisof the W elfare State, in O E C D (1981).

October.

The Welfare State and the Economy

Documents