Federalism and the Welfare State:
The German Case
ZeS-Arbeitspapier Nr. 8/2004
Zentrum für Sozialpolitik
Dieses Arbeitspapier ist im Zusammenhang des Forschungsprojekts über "Federalism and the
Welfare State. New World and European Experiences" entstanden. Es wurde von Herbert O-
binger, Stephan Leibfried und Francis G. Castles durchgeführt und ist von der Volkswagen-
Stiftung und dem HanseWissenschaftskolleg gefördert worden. In diesem Projekt wurde die
Interaktion zwischen Föderalismus und Wohlfahrtsstaat in den sechs "alten" Föderalismen in
der OECD in seiner historischen Dynamik untersucht, und zwar einerseits in Australien, Canada
und den USA, andererseits in Österreich, Deutschland und der Schweiz. Die Ergebnisse wer-
den 2005 bei Cambridge University Press erscheinen.
Zentrum für Sozialpolitik
- Barkhof -, Parkallee 39
In der vergleichenden Wohlfahrtsstaatsforschung wird der Föderalismus regelmäßig als
Faktor identifiziert, der die sozialstaatliche Entwicklung verzögert und dem Wohlfahrts-
staatswachstum Grenzen gesetzt hat. In Deutschland scheint jedoch der Föderalismus ver-
einbar mit einem der großzügigsten Sozialstaaten der Welt. Der Aufsatz geht der Frage
nach, ob die deutsche Ausnahme die besonderen Bedingungen erhellen kann, unter denen
der Föderalismus den ihm zugeschriebenen hemmenden Einfluss auf die Sozialstaatsent-
wicklung besitzt. Zentrale These des Aufsatzes ist, dass in Deutschland die föderalen
Staatsstrukturen nicht nur nicht hemmend, sondern sogar ausgabenexpansiv wirkten. Die
besondere Spielart des kooperativen Föderalismus begünstigte ‚fiskalische Unverantwort-
lichkeit’ der verschiedenen Staatsebenen und die Beitragsfinanzierung des Bismarckschen
Wohlfahrtsstaats ermöglichte es Bund und Ländern, ihre Einigungsprobleme zu Lasten
Dritter zu lösen.
The comparative welfare state research regularly highlights ‚federalism’ as a factor that has
delayed welfare state development and sets clear ‚limits to welfare state growth’. Yet, ap-
parently German federalism goes together with one of the most generous welfare states of
the world. This paper argues that federalism in Germany not only has not hindered strong
welfare state expansion, but actually has contributed to it. The special variant of ‚coopera-
tive federalism’ has blurred political and fiscal responsibilities, and the contribution fi-
nanced Bismarckian welfare state has allowed the central government and the states to
come to terms at the expense of a third party. A steady process of externalizing costs out of
the public budgets and into the parafiscal budgets of the welfare state has been the result.
1 Introduction ......................................................................................................................5
2 Bismarckian social legislation as a federalist compromise ..............................................8
3 Weimar and the conflict over resources and competencies............................................14
4 The Federal Republic´s welfare state and fiscal joint decision traps..............................25
5 Conclusion .....................................................................................................................43
6 References .....................................................................................................................45
Conventional wisdom strongly suggests that federalism is inimical to high levels of social
spending. Two arguments are prominent in this context: a veto-point thesis and a ‘competi-
tion of jurisdictions’ thesis. The veto-point thesis is quite straightforward: federal systems
have more veto points than unitary systems ceteris paribus (Tsebelis/Money 1997; Tsebelis
1995). This increases the probability that groups opposed to welfare state expansion can
exert some influence in the legislative process. Veto points would then give these groups
the opportunity to block or substantially water down redistributive legislation (Huber et al.
1993; Skocpol/Amenta 1986; Ross 1997). ‘Competition of jurisdiction’ arguments hold that
welfare redistribution is limited in federal systems because those who would pay more than
they would gain in a given jurisdiction (high income earners, ‘capital’) can credibly threaten
to exit highly redistributive and join less égaliste jurisdictions (Brennan/Buchanan 1988;
Weingast 1993; Weingast 1995; Inman/Rubinfeld 1997). At the same time, those who gain
more than they would pay (e.g. low income earners) are attracted to regions with higher
level of redistribution and these would therefore develop into ‘welfare magnets’ (Peter-
son/Rom 1990). Thus, a re-distributional policy stance is self-defeating in a federal context.
Indeed, many econometric studies of the determinants of welfare state spending have found
that federalism exerts a statistically significant, stable and negative influence on social
spending (Wilensky 1975; Crepaz 1992; Huber et al. 1993; Huber/Stephens 2001; Castles
2000; Hicks/Misra 1993; Hicks/Swank 1992; Swank 2001; Swank 2002). Prominent coun-
try cases are Switzerland and the United States, both strongly federalist countries and his-
torically, prominent welfare ‘laggards’ (although since 1980 Switzerland has moved rapidly
from laggard to leadership status). Other cases providing support for these arguments would
be Australia and Canada, again federal countries that, for much of the post-war period, have
had well below average levels of social spending. Germany, however, is a federal polity
which combines big government with generous social expenditure. This ‘anomaly’ may
motivate us to take a closer look at the postulated inverse relationship between federalism
and welfare state development. Is Germany only the exception that proves the rule or does
the German case provide unsettling counter-evidence to the federalism-thesis? In this chap-
ter, I argue that the German case alerts us to the fact that the dampening impact of federal-
ism on welfare state spending is likely to hold only under special circumstances. Two con-
siderations seem to be of particular relevance in this respect:2
1 I thank Herbert Obinger, Stephan Leibfried, Christine Trampusch, Steffen Ganghof and especially
Francis Castles for very helpful > comments, and Annika Schulte for editorial assistance.
2 A third qualification to the ‘federalism as an impediment to welfare state growth’ thesis may be men-
tioned here, but is not exemplified by the German case (see Bonoli 2000): federalism only appears to
have the postulated blocking effect in times of welfare state expansion (the ‘old politics of the welfare
state’), but not in periods of retrenchment (the ‘new politics of the welfare state’).
First, the federalism hypothesis critically depends on the sort of federalism that prevails in a
given country: is it one that establishes separate jurisdictions between the central and the
regional level (interstate federalism; Trennföderalismus), or is it one of interlocked or joint
jurisdictions (intrastate federalism, Verbundföderalismus, cooperative rather than competi-
tive federalism)? Where federal institutions establish joint jurisdiction over social policy,
federalism does not necessarily have a constraining impact on welfare state growth. Race-
to-the-bottom dynamics or beggar-thou-neighbour politics, greater political accountability
and limited opportunities for blame avoidance – the mechanisms said to constrain welfare
state growth in federal polities – all seem to have much less bite in the case of intra-state
federalism. Intra-state systems do not establish clearly separated spheres of fiscal responsi-
bility, legislative competency and political accountability between different layers of gov-
ernment, but rather mesh them between the regional and national units. In this case, the
conventional ‘competition of jurisdictions’ arguments do not apply (Weingast 1993; Wein-
gast 1995; Brennan/Buchanan 1988). On the contrary: Because legislative and fiscal re-
sponsibilities do not fully overlap, but neither are clearly separate, the incongruence be-
tween ‘having a say’ and ‘having to pay’ provides incentives to shift costs and responsibili-
ties between the different levels of government and between the different budgets of the
central state, of the regional states and of the welfare state. This mismatch results in a lack
of political transparency. It may, for instance, provide incentives to claim political credit for
new spending programmes, while – at the same time – avoiding the blame for correspond-
ing increases in taxation or public debt. This puts a premium on ‘fiscal irresponsibility’,
which may become manifest in the form of higher than average expenditure growth dynam-
ics or the ‘political overgrazing of the fiscal commons’ (Wibbels 2000; Rodden 2002; Rod-
den/Wibbels 2002). Alternatively, intra-state federalism may make legislating new (joint)
taxes extremely difficult, so that a demand for increased state spending is met through the
mechanism of para-fiscalism or contribution-based welfare financing – German unification
would be a good case in point (see below). In both cases, expenditure increases resulting
from the common pool resource dilemma inherent in cooperative federalism run directly
counter to the veto-point hypothesis.
Second, the veto-point argument crucially depends on the assumption of significant differ-
ences in policy preferences between the different veto players/ parties. Yet, this cannot be
taken for granted, as again the German case exemplifies. The ‘policy distance’ between
Christian Democracy and Social Democracy has never been very marked on questions of
social policy (Budge et al. 2001). This means that Germany’s multi-veto-point polity cannot
have exerted the hypothesised restrictive influence on government spending during the era
of ‘the old politics of the welfare state’, at least with respect to partisan-political sources of
‘gridlock’. It also means that, in the era of the ‘new politics’ of the welfare state, Christian
Democrats and Social Democrats can either diverge together from the ‘high redistribution’
status quo or not at all. Under these circumstances, what hinders reform efforts is not so
much the blocking effect of federal structures, but rather the dynamics of inter-party compe-
tition (Kitschelt 2001).
In what follows, I will trace the German welfare state’s history of institutional development,
starting with Bismarckian social legislation (Section 2), continuing with the division of so-
cial policy competencies between the central government, the states and the localities in the
Weimar Republic (Section 3) and, finally, analyzing the interplay between the federal gov-
ernment, the states and the welfare state in the Federal Republic of Germany (Section 4).
On the negative side, my main arguments are that federalism did not function to block wel-
fare expansion, because the policy preferences of the main actors were similar and pro-
welfare, and that the dynamics of cooperative federalism served to prevent any kind of in-
terregional ‘race to the bottom’ in social policy. On the positive side, I argue that coopera-
tive federalism actually encouraged expenditure growth, since unclear demarcation lines
between the central government, states and localities invited politicians to adopt credit
claiming and blame avoidance strategies with expansionary consequences. In particular I
argue that given that the states can veto tax legislation, but cannot prevent increases in so-
cial insurance contributions, federal veto structures in the German case translated into wel-
fare state growth. Section 5 summarizes the argument and discusses some implications for a
2 Bismarckian social legislation as a federalist compromise
Usually the foundation of the German welfare state is understood as having been part of a
carrot-and-stick strategy. Repression of the working class through the Anti-Socialist Law
(1878) and the imperial edict of 1881, in which the coming social legislation was an-
nounced, were indeed closely linked. Less well known is that the foundation of the German
welfare state between 1883 and 1889 was also a huge exercise in nation- and state-building.
Germany, along with Austria, was at the time a 'constitutional-dualistic monarchy', in which
a modernizing state elite could respond to the challenges of industrialization and working
class mobilization with the early adoption of social insurance (Flora/Alber 1981).
In nation-building terms, the welfare state was designed to provide momentum for the in-
ternal foundation of the Reich after the German/French war of 1870/71 had brought external
‘territorial consolidation’. The welfare state offered the Social Democratic and the Catholic
camps an opportunity for social integration, after both had been stigmatized as enemies of
the Reich during the Kulturkampf-era and in times of political repression under the Anti-
Socialist Law. Bismarckian social legislation was targeted at workers, especially at the bet-
ter-off strata of the working class. And the party that was second most successful in mobi-
lizing voters and members among the working class besides the Social Democracy party
was the Catholic Zentrum or Center Party.
The new social insurance schemes with their proto-democratic structure, provided new ave-
nues for political participation and social integration to workers (Manow 2001, chapter 2).
Clearly, the new social rights were meant as partial substitutes for workers’ lack of democ-
ratic right – general suffrage in the national election was not worth much given the power-
lessness of the Imperial Diet, and state elections were often still based on a plutocratic sys-
tem censitaire (cf. Flora/Alber 1981). But it is also important to highlight that social insur-
ance itself – through the democratic system of workers’ self-administration – offered work-
ers important new electoral-participatory rights in areas of direct interest for them. State
building, on the other hand, meant that social reform offered the central government a new
arena of political activity and a genuine legislative responsibility. Social reform established
a new administrative domain for the central state, it established the need for a new bureau-
cratic apparatus, and it promised to open up new sources of revenue for the Reich. All this
was an impressive exercise in state building and meant that, from the outset, Bismarck’s
social legislation had a strong anti-federalist momentum. For instance, an immediate conse-
quence of the central state’s assumption of welfare obligations through Bismarckian social
legislation in the period 1883 to 1889 was the need to establish a central bureaucracy
charged with the oversight of the new welfare system (Kahlenberg/Hoffmann 2001).
Much of the need for social reform and of the central government’s more or less undisputed
dominance in this new policy domain can be explained by the apparent inadequacy of the
old local support systems for the poor and destitute. The regionally scattered system of so-
cial assistance had proved inadequate when confronted with steeply increasing worker mo-
bility in the wake of Germany’s feverish industrialization in the last quarter of the 19th cen-
tury. Previously, state aid to the poor and destitute had been provided either according to
the principle of origin or according to the principle of current residence. Neither principle
proved adequate in coping with the new challenges posed by rapid industrialization. The
principle of origin meant that poor rural districts, from which workers had left for the big
cities, had to subsidize the new industrial centres of the Reich. Moreover, the principle hin-
dered worker mobility in the first place and aggravated severe labour shortages in the new
industrial centres. The residence principle, on the other hand, obliged these new industrial
centres to support the poor. Yet, this in turn resulted in fiscal problems as soon as locally
concentrated industries faced a business cycle downturn. Thus, the need for a national re-
sponsibility in social reform and for risk pooling beyond the regional districts was widely
accepted by informed opinion in the 1870s and 1880s.
The organisational principles according to which the new social insurance was designed
differed from branch to branch, but nowhere did they exactly mirror the federalist structure
of the German Reich. True, Bismarck’s initial attempt to give the new welfare state a dis-
tinct centralist character by establishing a central state agency staffed by civil servants (and
financed not by contributions, but out of the central state’s budget, see below) failed due to
the resistance of the states (Länder) in the Bundesrat, the federal chamber. But nor did the
new social insurance follow a purely federalist design (as claimed by Henning 1977:95;
Ritter 1983:73). Accident insurance was organized along industrial lines, since firms within
the same industry were understood to constitute a distinct risk class. The health insurance
funds, on the other hand, either started as regional organisations (with further differentiation
for specific occupations), as company funds or as institutions of collective self-help for cer-
tain professional groups. Sickness funds quickly grew in number, but soon problems stem-
ming from their small size made themselves painfully felt. A major overhaul of the system
in 1911 reduced the number of funds by more than half from more than 20.000 to less than
10.000. In the case of invalidity and old-age insurance, 31 regional insurance agencies ran
Almost nowhere did the administrative units of the welfare state correspond with state
boundaries. Moreover, when the white-collar movement succeeded in 1911 in obtaining an
exclusive old age insurance branch of their own, this introduced an important element of
competition between the 31 regional insurance offices and the new single national insur-
ance agency for white collar employees. This competition became one of the most impor-
tant catalysts for a uniform legal and centralist institutional development of old-age insur-
ance. Similarly, sickness funds for white-collar workers were organized as supra-regional,
sometimes even nation-wide organisations. Naturally, these funds were of much larger size
than the average local or company sickness fund. In 1925, five white-collar funds counted
an average membership of 224,500, a number about 100 times as large as the average mem-
bership in one of the statutory, local or company funds. Again, this exerted a unifying effect
in the health care insurance system, since contributions could be lower in larger funds
thanks to administrative economies of scale and the advantages of risk pooling. The unem-
ployment insurance system, which was founded as the last of the classical social insurance
schemes in 1927, was organised on strictly centralist lines. A central agency with 13 re-
gional sub-branches took over responsibility from the 22 state and 869 municipal employ-
ment offices that, previously, had been in charge of unemployment support and labour mar-
ket policies (Lewek 1992; Führer 1990).
This process of concentration and centralization continues up to the present day and demon-
strates the underlying mechanism that has driven much of the institutional development of
the German welfare state towards ever higher degrees of uniformity and centrality: the ad-
vantages of larger over smaller risk pools. This combined with the central state’s interest in
the extension of risk-pools in order to minimize fiscal liability for ‘bad risks’. Thus, it is
clear that questions of organisational design were simultaneously issues of considerable
According to Bismarck’s initial plans, the new social insurance schemes were supposed to
be, prominently if not entirely, financed by taxes. This would have granted to the new cen-
tral state a right of taxation that only existed previously in vestigial form. Bismarck planned
to use revenue from a newly established state monopoly for tobacco to finance the new so-
cial programs. That the Reich possessed only a small revenue raising capacity was a conse-
quence of the fact that the German constitution of 1871 had resulted from a political com-
promise, which had centralized legislative powers, but had delegated the responsibility for
administrative implementation and enforcement to the Länder governments. The upshot
3 10 additional insurance agencies took care of seamen, railroad workers and miners. See Frerich/Frey
was that policies had to be agreed upon jointly by the central state and the states, effectively
granting the states a veto on all political decisions (cf. Lehmbruch 1998, chapter 6). Yet,
implementation was the responsibility of the states and this meant that most of the tax reve-
nue remained in state hands, while the Reich possessed no substantial sources of taxation of
its own. The Reich could raise revenue from state-owned enterprises (like the post office),
and from tariffs and indirect taxes (on sugar, salt, beer, matches, liquor etc.). However, to
cover its expenses, the German central state of the late 19th century was also dependent on
Länder-contributions. Initially, in the 1870s, such contributions amounted to 15 to 20 % of
all revenue, although later – especially after the Reich embarked upon protectionism via
high tariffs – the central budget became more independent of state subsidies. Yet, Reich tax
revenues remained small (Witt 1970:378 f.). In this period, central taxes as a percentage of
GDP were around 2%, the smallest figure for countries for which we have comparable data
(Flora 1983:268). With respect to the ratio of central to general taxes, only the Swiss central
government was weaker than its German counterpart (39.3% of overall taxation in 1886
compared to 49.2 % in 1881; ibid.).4 In Germany the central state possessed no large admin-
istrative apparatus of its own, with the Reich spending most of its revenues on the army and
the fleet and, to a small but increasing extent, on the newly established welfare state (Nip-
perdey 1992; Witt 1970:380). The inadequacy of the Reich’s tax base meant that an increas-
ing part of total spending was financed out of a rapidly increasing public debt (Witt 1992).
Bismarck hoped that the central state’s new responsibility for the social security of German
citizens would legitimise tapping new sources of direct tax revenue for the Reich. Yet, as in
the case of the institutional design of the welfare state, its final financial architecture was
once again a compromise between the Reich and the Länder. Central tax financing was not
to play a dominant role in the funding of the German welfare state, but nor did the fiscal
structure take on federalist characteristics (which would have introduced the potential for
‘competition of jurisdiction’ dynamics into the system). Instead, social security contribu-
tions became the dominant source of revenue. The welfare state grew rapidly and its in-
creasing fiscal and macro-economic importance is demonstrated by the fact that, already by
1904, overall welfare state revenue amounted to more than two third of the central state’s
general tax revenue. This has meant that there has always been a temptation to tap the fi-
nancial resources of the welfare state for particularistic fiscal purpose – as the Reich did for
the first time on a massive scale in WW I., when it used the financial assets of the social
insurance funds to finance the war. Similarly, the autonomous financial status of the welfare
state has also tempted politicians to achieve fiscal relief by shifting costs out of the public
4 Interestingly, this picture largely continues to hold true. In the 1980s and 1990 Germany still ranked
among the OECD-countries with the highest degree of fiscal decentralization and the lowest degree of
fiscal centralization (see, also for definitions, Castles 2000:179ff., Table 8.1), but also as the OECD
country in which the share of federal grants and revenue sharing receipts of total provincial revenue
was highest (See Rodden/Wibbels 2002:504, Table 1). A third aspect is also important: local plus
central revenue make up only about two thirds of total revenue – which hints at the enormous fiscal
importance of the welfare state in Germany and of the central role of its para-fiscal mechanisms.
budget onto the special budgets of the welfare state. I will return to the determinants of the
long-term erosion of the central state’s financial involvement in social spending and its ef-
fects in Section 4 below.
The federalist character of the German Reich not only had a profound impact on the institu-
tional architecture and the financial basis of the German welfare state, but also on the sub-
stantive nature of national and regional social protection. The division of labour between
the Reich and the Länder, according to which the central state was now responsible for Ar-
beiterpolitik (social insurance for workers), while the Länder and municipalities remained
responsible for Armenpolitik (traditional social assistance for the non-working poor and
destitute), helped to free the new social insurance from Poor Law-traditions. The social
protection system could focus exclusively on the social risks and vagaries stemming from
industrialization and could be targeted primarily at workers, in fact mostly at the ‘labour
aristocracy’ of trained and politically active workers, leaving local communities to take care
for the poor in the traditional way. This dualism, which is most manifest in the largely arbi-
trary demarcation between social assistance (covered by the municipalities and states) and
the lower layers of social insurance, becomes most contested in the case of the long-time
unemployed and remains a highly controversial feature of the German welfare system to the
present day (Sachße/Tennstedt 1980, 1988, 1992).
In summing up, one can say that central institutional characteristics of the German welfare
state were essentially the outcome of a compromise between the Reich and the states. The
compromise had three distinguishing features: 1) the centralization/ nationalization of the
legislative responsibility for social policy, 2) an independent organisational design of the
welfare state which was neither national nor federalist in character (Stolleis 2001:265), and
finally 3) the financial autonomy of the new schemes; their para-fiscal status – already rela-
tively high from the schemes’ inceptions, but steadily increasing in following decades.
These features played a decisive role in the further development of the German welfare
state, which involved a steady process of coverage extension plus organisational concentra-
tion and centralization.
3 Weimar and the conflict over resources and competencies
Social policy moved to centre stage in the Weimar Republic. For the first time, social rights
became constitutionally guaranteed (especially Articles 157, 159, 163, and 165 of the Wei-
mar constitution). Their prominent status in the Weimar constitution reflected the fact that
the ‘coalition of Weimar’ – formed by the Social Democratic party, the Catholic Center
and the liberal DDP – was crucially based upon the parties’ common interest in the domain
of social policy. Social policy was a natural point of agreement especially between the two
big mass-member parties, the Social Democrats and the Zentrum, since both parties had to
please their substantial worker electorate, both possessed tight linkages to the socialist and
catholic unions respectively, and both had established extensive networks with either politi-
cally or religiously motivated collective self-help organisations during the time in which
they had been excluded from political power. Yet, push factors rapidly began to join these
pull factors in the aftermath of the Great War, as Germany found itself coping with a social
and economic crisis of massive proportions. The steeply increased number of war invalids,
of widows and orphans, of those suffering from malnutrition, of refugees coming from lost
territories, of the impoverished elderly and of 6 million soldiers, who returned from the
front, added to the serious economic problems stemming from the abrupt transition from a
war- to a peace-economy. The number of persons dependent on welfare roughly quadrupled
from 1914 to 1924 and per capita welfare expenditure increased nearly eightfold during the
same period (Sachße/Tennstedt 1992:81).
The severe social and economic problems of post-war Germany amplified the need for pro-
found reform and substantial reinforcement of the existing social protection system. In face
of these problems, the central state’s political responsibility for the war combined with the
problem overload of state and local authorities to give a strong centralist momentum to
Weimar’s welfare state development. This trend can be clearly read off from the data.
Whereas only 5.3 % of all central government expenditures was devoted to social spending
in 1913/14, in 1925/26 the share was already at 35.8 % – only to rise further to almost 50 %
at the peak of the economic crisis in 1932/33 (Frerich/Frey 1993:175, table 20; cf. Bürger
1930). Similarly, central government social spending plus social insurance expenditures
made up for a steadily increasing part of total social spending – from 53 % in 1913/14 to
69 % in 1932/33 (ibid., own calculations).
The war economy itself had already had a strong centralizing impact, and to some extent the
Weimar constitution as well as the important tax reform of 1920 only ratified the new
power balance between the Reich and the Länder. Centralization of legislative, financial
and administrative responsibilities, harmonization of regulation and homogenization of ad-
ministrative design, increased generosity and extended coverage of social benefits – in all
these respects the war had figured as an important “pace setter” of national welfare state
development (Preller 1978:85; cf. Titmuss 1976; Skocpol 1992; Kasza 1996). The same
held true for federalism. Already in 1914, in the wake of the start of WW I, all legislative
responsibilities had been delegated from parliament to the federal chamber, but this led not
– as one might have assumed – to a stronger role for the states in policy making, but rather
to a hegemony of the central administration due to the dominance of Berlin’s central Prus-
sian bureaucracy. Once the war was over, there was no going back to the federalist status
After the war, the political and financial balance tilted even more clearly towards the Reich.
The states’ chamber, the Reichsrat, was weakened, since under the Weimar constitution it
enjoyed only the right to a so-called ‘suspending veto’, which could be overridden by the
Diet or popular house of parliament and it lost its former exclusive right of legislative initia-
tive. In fact, an effective veto of the Reichsrat remained a rare event, since the political
alignment of the central and Länder governments was quite close throughout much of the
1920s. In particular Prussia, by far largest and most important state (2/3 of the territory, 2/3
of the population, and with 26 of 66 votes in the upper house), generally supported or at
least tolerated Weimar governments throughout the 1920s (Lehmbruch 1998:70ff.).
Compared to both the pre-World War I Reich and to the post-World War II Federal Repub-
lic, the constitutional foundations of federalism were less strongly developed during the
interwar years5: Under the previous imperial constitution, the federal chamber had pos-
sessed an absolute veto on all legislation and an exclusive right to initiate legislation. Today
the federal chamber can use its veto to suspend around 45 % of all legislation (Ein-
spruchsgesetze) and has an absolute veto on the other circa 55 % of legislation (Zustim-
mungsgesetze) (Schindler 1999:2430, table 3). In the Weimar Republic, the federal house
had only the suspending veto – still a powerful instrument given that the override required a
2/3 majority. To this came the loss of fiscal privileges. The tax reform of 1920 for the first
time gave the central government access to its own substantial tax-revenue (see Table 2),
while the constitution had equipped the Reich not only with the autonomy to legislate in full
sovereignty all tax laws it deemed necessary for its own revenue needs (Article 8), but also
to regulate the states’ taxation. Table 1 identifies and contrasts the essential features and
rights of the federal chamber under the Reich, Weimar and in the post-war federal republic
5 That parties of the extreme Right and Left used their power position in the regional states, like in
Bavaria or Thuringia, to follow secessionist strategies, is another story.
Table 1: The position of the federal chamber in Wilhelmine Germany, the Weimar Re-
public and the Federal Republic (Sources: Huber 1981:378; Huber 1988:855;
Oeter 1998:63; Nipperdey 1992; Schindler 1999:2447; Renzsch 1991)
Bundesrat has absolute veto
(Article 5 of the imperial con-
Reichsrat has only suspending
veto (Article 74 of the Weimar
constitution). To overrule the
veto, a two-third majority in
parliament is required.
Bundesrat has absolute veto in
matters that affect regional
administration (≈ 55 % of all
laws); in all other matters only
a suspending veto (Article 77,
84, 85 and various other arti-
cles of the Basic Law). [A
suspending veto with a two-
third majority in the federal
chamber can be overruled
with a two-third majority in
parliament as a whole; Article
77, paragraph 4.]
25 states (26a) 18 states 11 states (16d)
Total of 58 seats/votes (61a)
¾ Prussia 17
¾ Bavaria 6
¾ Württemberg and Saxony
¾ Baden and Hesse each 3
Total of 66 seats/votes(68b)
¾ Prussia 26 (27c)
¾ Bavaria 10 (11c)
¾ Saxony 7
¾ Württemberg 4
¾ Baden 3
Total of 41 seats/votes (68d)
¾ North Rhine-Westphalia,
Württemberg and Lower
Saxony each 5 (6d)
¾ Hesse, Rhineland-
Palatine, Berlin and
Saxony each 4d
Members are delegates of the
regional governments; no free
Members are delegates of the
regional governments; no free
Members are delegates of the
regional governments; there is
no free mandate.
States receive all direct taxes,
the central state all indirect
taxes. Tax laws need the con-
sent of the states.
Taxation is completely a re-
sponsibility of the central
state. Income- and sales-tax
are shared between Reich and
states. The central government
has Kompetenzkompetenz, i.e.
it can decide who can levy
which taxes. The fiscal equali-
zation scheme is only moder-
Basically the same as under
the Weimar constitution. Ex-
tensive fiscal equalization
scheme, which guarantees that
no state has a per capita tax
revenue lower than 95 % of
the national average.
The right to legislate lies with
the federal chamber.
The right to legislate lies with
the Diet or popular house of
The right to promulgate laws
lies with the Bundestag or
popular house of parliament,
yet laws need the consent of
the federal chamber.
The federal chamber has the
exclusive right to initiate leg-
The federal chamber has the
right to initiate legislation (co-
The federal chamber has the
right to initiate legislation (co-
* from 1911; ** between 1926 and 1928; § since 1926; $ since 1990.
However, it is also necessary to emphasize that the centralist momentum of the Weimar
years was not just an outcome of war and economic crisis, but was also an expression of a
more secular trend. In this context, it is important to note that fiscal federalism and strict
inter-state federalism have never been distinguishing features of the German polity
(Lehmbruch 2002: 80ff.). From the very start, German federalism had a strong bias in fa-
vour of national unity and legislative centralization – even if this tendency remained short
of administrative centralization. To establish strict interstate federalism was never seriously
contemplated outside the southern-Catholic camp with its fear of Prussian-protestant he-
gemony. Revealingly, in the constitutional debates of the second half of the 19th century, the
United States and Switzerland served as negative reference points to exemplify how Ger-
man federalism should not develop (Oeter 1998:32, 34, 40; Lehmbruch 1998:80ff.). But
what then explains the German federalist compromise? The federalist structure of the ‘be-
lated German nation’ simply was an acknowledgment of the fact that the German states
already were autonomous, sovereign entities. Therefore, the constitution was clearly meant
to be an institutional structure that would allow and foster national integration and help
overcome federalist fragmentation (Hesse 1962). In other words, federalism was not de-
signed as a constitutional safeguard for regional particularism, but rather as the best avail-
able instrument for achieving the goal of national unification – given that German unifica-
tion could realistically only happen through the delegation of power from the already firmly
established and sovereign states to the Reich (Oeter 1998). It was in line with this underly-
ing unitary tendency that the states under the Weimar constitution enjoyed only a reduced
veto power in the second chamber and lost some of their former fiscal privileges.
The 1920 tax reform finally provided the material basis for the integrative goals that all the
major political actors shared. In all matters of taxation, the Reich could now exert influence
through its right of concurrent legislation (konkurrierende Gesetzgebung), which stated that
the states could regulate their tax affairs autonomously only insofar and only as long as the
central government abstained from enacting nation-wide standards (Article 8 of the Weimar
constitution).6 Furthermore, the tax reform prohibited the localities from levying additional
charges in addition to direct taxation. This not only levelled the enormously varying tax
burdens between the different regions, but also had an important welfare side effect: it made
local social policy (i.e. social assistance) less dependent on the local tax power, and this, in
turn, eased nation-wide regulation of local social policy and standardization of benefits.
However, at the same time, the reform intensified the fiscal interdependencies between cen-
tral state, Länder and localities. It radically changed the relationship between the central
state and the Länder: whereas before the Reich had depended on Länder-transfers, now the
states became dependent on transfers out of the general budget (Sachße/Tennstedt
1980:176ff.). As Table 2 shows, in the mid-1920s, the central government became, for the
first time, the biggest recipient of tax revenues.
6 Therefore the term ‚concurrent legislation’ is misleading. The central government has the prerogative
of legislation, the German term Vorranggesetzgebung might be more precise.
Table 2: Percentage tax revenue shares of national, regional and local government,
1885-1970 (Hidien 1998:338, 460, table 18. See also OECD, Revenue Statis-
tics and Figure 2 below)
Year National government States Municipalities
1885 18 57 25
1913 30 29 41
1925 38 26 36
1936 66 10 24
1950 52 31 17
1970 54 34 11
As in the field of tax law, the central government now, for the first time, also enjoyed the
privilege of concurrent legislation in the domain of social assistance. These new powers
were soon used to harmonize widely differing standards and levels of local support for the
poor and needy. This was not the only proof of the close nexus between tax and social pol-
icy. In the case of social assistance, earmarked financial transfers were the government’s
initial ‘foot in the door’ in a policy domain in which it had not enjoyed much influence pre-
viously. Financial transfers out of the national budget often came with quite precise policy
directives. Therefore, it was not accidental that the central state enacted a tax emergency
decree at the same time as the important decree on welfare early in 1924 (cf. Stolleis
2001:280). Whereas the tax decree increased the states’ share of personal income taxes from
75 % to 90 %, the welfare decree for the first time substituted uniform national legislation
for the various state laws that previously had existed in this policy sector. But legal har-
monization and central standard setting went hand in hand with administrative reform:
Devolution (of central government responsibilities) and centralization (of local responsibili-
ties) conspired in to assign most of the administrative and financial responsibilities attach-
ing to the provision of welfare at the Länder- or regional-level.
The central state’s intrusion into the states’ own policy territory met with little resistance so
long as it relieved local authorities from some of the pressing financial stress resulting from
mass poverty and well-nigh permanent economic recession. Prescriptive notions of uniform
living conditions and, hence, uniformity of provision also played an extremely important
legitimizing role for central policy initiatives in the welfare sector. Different rules and regu-
lations, tariffs and laws had been the prime target of a forceful bourgeois-liberal critique
castigating them as indicators of outdated parochialism and as barriers to national market
integration which it was the true mission of German national unification to overcome
(Lehmbruch 2002:80ff.). An encompassing national system of taxation reinforced centralist
tendencies in the development of the German nation state, and also ruled out all ‘competi-
tion of jurisdictions’ dynamics in all tax-financed areas of the welfare state. For instance,
tax-financed local social assistance largely followed nation-wide rules and was largely fi-
nanced from the national budget by way of direct earmarked transfers to the states’ budgets.
In Germany social legislation was deliberately used as an instrument of national unification
and was explicitly designed to prevent a federalist fragmentation of living conditions.
Nor was this unifying effect of social policy an issue of much political dispute. Conserva-
tives, liberals and Social Democrats alike were in favour of national unification led by a
strong central state and leading to ‘uniform living conditions’. Social Democracy had been
anti-federalist ever since universal and equal (male) suffrage had been granted in 1871 for
national-level elections to the imperial Diet, while eligibility to vote in most regional elec-
tions before 1918 (especially in Prussia) was on the basis of property qualification. The
Catholic Zentrum was not as single minded in its support for central state responsibility.
The party represented both federalist and centralist currents, with the faction representing
the latter closely linked to the powerful movements of Social Catholicism. In the first dec-
ade of the 20th century, Social Catholicism had developed into a powerful nation-wide in-
terest group independent from the official church bureaucracy with its still regional struc-
tures (Müller 1996). Despite the high degree of party fragmentation in the Weimar Repub-
lic, no major party voiced decisively regional interests or favoured the organisational devo-
lution of the welfare state.
This preference for a common national response to the ‘labour question’ resonated well
with the ministerial bureaucracy. Both civil servants and German legal thought clearly fa-
voured nation-wide uniform regulation. Social policy was no exception. Contemporary ob-
servers characterised the Weimar Labour Ministry as revealing “markedly centralist tenden-
cies”(Preller 1978:290, my translation). This centralism brought the Ministry into conflict
with the Länder and local authorities. Yet the substance of their objection did not concern a
loss of policy discretion, but rather involved disagreement about the basis for allocating
costs between the central government and the states. And the resolution of this conflict by
an increasing reliance on contribution financing became something of a natural option
within Germany’s federal polity. This suggests that the German welfare state’s substantial
dependence on contribution finance should not solely be interpreted as the institutional em-
bodiment or expression of ‘conservatism’ (Esping-Andersen 1990). To a very important
extent it was the political solution to a conflict between the central government and the
states and municipalities over how best to minimize their financial involvement, whilst
leaving their rough fiscal balance of power untouched.
The one major social policy innovation in the Weimar Republic was the establishment of
unemployment insurance in 1927 (Führer 1990; Lewek 1992). This too was indicative of
the centralizing tendencies of the era. Here, the Reich finally established its dominance by
integrating existing local and regional labour agencies into a centralist administrative struc-
ture with the status of dependent Reichsunter- and Reichsmittelbehörden (local and inter-
mediate state agencies). This was the endpoint of a long and protracted battle between the
central state and the communities over administrative competencies and financial responsi-
bilities. The unemployment insurance scheme of 1927 finally established a hierarchical
chain of delegation and largely left states and communities without discretion in a domain
in which they formerly had enjoyed dominant influence. The administrative structure thus
became a prominent example of a deliberate break with the old federal compromise by
which legislative responsibilities had been centralized and nationalized, while the Länder
retained the prerogative for administrative execution and programme implementation. To
the present day, the labour office and the tax administration remain the two most important
deviations from the established federalist ‘division of labor’ between centre and regions. In
the Federal Republic, a quarter of all labour office employees are civil servants, the unem-
ployment insurance is run by one central agency with regional branches and, in the govern-
ing bodies of the labour office, the government represents a third of all delegates, whereas
in the sickness-, accident- and pension-insurance schemes, there is equal representation of
worker- and employer-delegates. Moreover, the financial nexus between the unemployment
insurance scheme and the public budget is quite close, given that the central state is obliged
to cover all deficits of the Labour Office (sickness funds are almost completely financially
autonomous, while the government covers a fixed share of total pension spending).
Yet, even this strictly hierarchical design of the unemployment insurance scheme could not
end conflicts between the centre and the localities in this sector. Since the localities re-
mained responsible for local welfare (social assistance), conflicts between the local authori-
ties and the central government were far from settled with the 1927 enactment of the unem-
ployment insurance law. The central points of conflict were the length of time that the un-
employed should receive benefits on an insurance basis and the basis on which the locali-
ties could use insurance funds for active labour market policies. Given the legislative and
tax-levying powers of the central government, states and municipalities were disadvantaged
in the distributive conflict over who should bear the costs of the economic recession that hit
Germany extremely hard in 1928 and subsequent years. Both with respect to taxes and the
fiscal mix in the welfare sector, the central government used its privileged position to bur-
den the states and localities with welfare tasks without equipping them with the necessary
financial means. Compensatory direct financial transfers to the communities out of the pub-
lic budget remained an insufficient remedy in times of weak economic growth and general
fiscal stress. Whereas prior to World War I, the central government lacked the resources to
match its increased responsibilities, now states and municipalities were the levels of gov-
ernment with inadequate financial means.
Weimar’s politics of welfare were driven by (mainly financial) conflicts between centre and
periphery and these conflicts were nurtured by the mismatch between fiscal, legislative,
administrative and political responsibilities within the German ‘cooperative’ federal system.
This made the “struggles over resources and competencies between Reich and communities
… an essential feature of the social policy dynamics of the Weimar Republic” (Sa-
chße/Tennstedt 1988:89). In this conflict, the states, in turn, were not always acting as the
true advocates of the interests of localities and municipalities. These, as the ‘protectors of
last resort’, increasingly were left alone in their struggle against the consequences of the
economic crisis, as the phrase ‘Kommunalisierung der Armut’ [freely translated: devolution
of poverty] suggests. Often, central government and states could compromise within Ger-
many’s system of cooperative federalism only at the expense of a third party. At times, con-
tributors lost out; at other times, the local authorities. Shifts between state financed unem-
ployment aid and locally financed social assistance indicate the increasingly fierce distribu-
tive conflicts between the different layers of the German state (Sachße/Tennstedt 1988).
The struggle between centre and localities over the devolution or centralization of policy
responsibilities also had an important partisan-political aspect. The centralization of policy
responsibilities was also part of a political attack against Germany’s (allegedly red) munici-
palities. The Catholic Center Zentrum, the bourgeois parties and the bourgeois and religious
voluntary welfare associations all tried to instil fears of so-called ‘creeping socialism’ at the
local level (Munizipialsozialismus), and corporatist cooperation between the Labor Ministry
and welfare peak associations was understood to be an antidote to this allegedly perilous
trend (Sachße/Tennstedt 1988). The centralization of social assistance and the integration of
voluntary associations as components of the national welfare state were thus also part of an
attempt to contain Social Democracy’s local influence in the larger cities and industrial cen-
tres. Once again, it is apparent that a central distinguishing feature of the German welfare
state, the dualism between public financing and state regulation of the welfare sector and
semi-private welfare provision by ‘third sector’ organisations , has been, in large part, an
outcome of federalist struggles between Reich and the localities.
Finally, it is worth asking how the German welfare state compared with others of the same
era? Comparative data for this early period are fragmentary, but according to the available
evidence, theWeimar welfare state was the most developed in Western Europe in respect of
both spending and programme coverage (Flora et al. 1983:460f., 365ff.; Alber 1982:152).
Obvious reasons include the relative maturity of German insurance programmes initiated in
the late 19th Century and the extreme levels of need characterizing interwar Germany.
Summing up, one can say that the dynamics of German social policy in the interwar period
essentially reflected the conflicts between the central state and regional and local authori-
ties. Both financial issues and partisan conflict between the different layers of government
loomed large in the social policy debates of the time, but the most important conflicts were
not between the national government in Berlin and the states, but between central and local
governments. A genuine federalist dimension of political conflict was not strongly devel-
oped in the interwar years, given that the states were less powerful than they had been be-
fore World War I, and given that conflicts were muted by a high degree of partisan-political
alignment between central and state governments. Conflicts were also less fierce because
the Berlin government and the states often could compromise at the expense of third parties
– primarily the insured contributors and the localities. Yet, overall, the Weimar period saw
the substantial centralization and nationalization of social policy. Partly, this was due to the
extraordinary socio-economic challenges with which the 1st German republic was con-
fronted. Partly, it was due to the overall centralist tendencies of German federalism itself.
Ultimately, however, both the republic and the welfare state broke down when confronted
with social, political and economic problems of enormous proportions. Arguably, no wel-
fare state, however designed, would have been able to cope with the economic catastrophe
and social dislocations of the late 1920s and the early 1930s. The advent of the National
Socialist rule marked the end of federalist state structures and its replacement by a ‘dual
state’ (Fraenkel) constituted of parallel and often redundant central state bureaucratic struc-
tures and of their Nazi party organisational counterparts.
4 The Federal Republic´s welfare state and fiscal joint decision traps
We turn now to the development of the German welfare state in the post-1945 era, asking
initially how the Nazi interregnum had influenced the structure of the welfare state that
post-war Germany inherited. The answer is that the direct impact was surprisingly small
and the lasting influences relatively subtle, with the literature substantially agreeing on the
lack of success of the radical reform visions of the National Socialists in the social welfare
field (Teppe 1977; Sachße/Tennstedt 1992; Scheur 1967). A powerful coalition of forces,
comprising the conservative ministerial bureaucrats of the labour ministry, the administra-
tive apparatus of the social insurance branches, business interest groups and competing
party and executive factions, struggling for influence within the chaotic regime, effectively
delayed the enactment of the grandiose Nazi reform plans until the war began. After 1939,
the government shied away from fundamental reforms,because it feared such reforms would
endanger social peace and would stand in the way of the smooth working of the war econ-
omy by opening a home front. In organisational terms, the ‘classical’ core of the German
welfare state, the social insurance schemes, thus, remained, to a remarkable extent, un-
Yet, by abolishing most elements of joint union/ employer administration of the social in-
surance schemes, especially by destroying the unions, and by integrating the social insur-
ance schemes into the state apparatus, the National Socialists gave an already centralist or-
ganisational development a further strong étatist momentum. And with the abolition of the
Länder in 1934, German federalism officially ceased to exist. However, since the German
welfare state possessed an organisational structure that was largely independent of the struc-
tures of German federalism, the end of federalism had no important direct impact on the
welfare state itself. Yet, the long-term consequences were quite substantial: once the states
were re-constituted after WW II, they were confronted with a Bismarckian system that in
the meantime had become more centralized and autonomous, more uniform, more encom-
passing, more bureaucratic and less politicized (Leibfried/Tennstedt 1979).
These tendencies may be best exemplified with respect to the old-age insurance system.
Here the fact that the fundamental reform plan of the Nazis (e.g. introduction of a uniform,
tax-financed people’s insurance that would have fused the white-collar and blue-collar
branches) was never enacted should not distract from the fact that subtle, but long lasting,
modifications did take place under the Nazi regime. In this respect, we may mention the
following five changes: 1) the Nazis started to pay one global (augmented) state subsidy for
the pension insurance scheme instead of the fixed subsidy per individual pension paid pre-
viously, 2) they issued a central state guarantee for the pension insurance, thereby taking
over fiscal responsibility from the Länder and communities which previously had guaran-
teed pension payments, 3) for the first time, contribution payments were automatically de-
ducted from wages as an exact and equal percentage share for both blue collar- and white
collar-workers7 (previously, payments had to be paid in personally at a post office and were
differentiated according to different wage groups and with different contribution rates for
white- and blue-collar worker), 4) during the Nazi era, social insurance for handicraft work-
ers became obligatory (1938) and 5) pensioners were integrated into the health insurance
scheme (1941). Thus, much of what happened in organisational terms during 1933 and
1945 can be summarized as rationalization, membership extension and further centraliza-
tion, but did not qualify as a radical break with traditional organisational principles. All of
these changes survived the Nazi-regime and became part of the re-founded social protection
system of the Federal Republic after the war.
When it comes to the post-war period, the most revealing observation about the relation
between the federal Bundesrepublik and the reconstituted welfare state is one of the ‘why
the dog did not bark’ variety: why did the states not protest against their loss of policy re-
sponsibilities in the social policy sector, when the central state reclaimed political responsi-
bility for social policies between 1949 and 1953 after the immediate post-war years had
witnessed a significant organisational devolution of the German welfare state? This lack of
conflict is, on the one hand, strong evidence for the tacit consensus among the major politi-
cal camps and between the states and the central government about the national role – pro-
moting unity and equality – that the German welfare state is supposed to play. On the other
hand it was also the result of necessity. States differed too much with respect to economic
starting conditions and were too differently affected by the war (by destruction and the
refugee problem) to make welfare state decentralization a feasible option (see on this espe-
cially Scharpf 1989).
The process of administrative and territorial consolidation of the states had ended by mid-
1947. The first states were constituted as early as September 1945, only half a year after the
military surrender (Kilper/Lhotta 1996:83; Boldt 2003). The states were thus at least two
years ahead of the Federal Republic, which was re-inaugurated not earlier than late 1949.
Once again, as in 1871, the states were already in place before the nation state formed. Dur-
ing this time, the states became responsible for those social protection programmes that had
formerly been run at the national level, i.e. especially for the white collar-worker pension
and the unemployment insurance schemes (Kahlenberg/Hoffmann 2001). For all other pro-
grammes they preserved the right of administrative and legal oversight (but not of legisla-
tion) that they had under Weimar. The by and large non-contentious re-delegation of organ-
7 In their effort to create völkische unity (unity of the German people) the Nazis also fought against
what they perceived as outdated ständische (status in the sense of social estates) welfare privileges.
Their main target was the hitherto privileged status of white collar-workers. See Prinz 1986.
isational responsibilities from the regional level to the national level after 1949 was legiti-
mized by the commonly accepted idea that all Germans should enjoy ‘uniform living condi-
tions’ (Oeter 1998:532ff.). As was true of the period of initial Bismarckian social legisla-
tion, after WW II, nobody forcefully campaigned for an organisational or fiscal devolution
of the German welfare state.
When the German parliament, the Bundestag, convened for the first time after the first free
post-war elections in the three Western military zones in 1949, it had to decide on the future
of the German welfare state, while the division of powers between the central state and the
Länder had already been part of the constitutional compromise that had preceded the elec-
tions. The preliminary constitution, the West-German Basic Law, which was supposed to
last until the French, British and US-American zones – constituting the Federal Republic –
re-united with the Soviet-zone constituting the German Democratic Republic, deliberately
refrained from prescribing the institutional details of the future welfare state. Only a vague
reference to Germany being a ‘democratic and social federal state’ (Article 20) could be
found in the Basic Law. Yet, with respect to the design of German federalism, the Basic
Law was much more concrete. Essentially, the Basic Law prolonged with small but impor-
tant modifications the Weimar balance of powers. For all legislative initiatives that affected
administration at the state level, the second chamber, the Bundesrat, as the representative of
the states at the federal level,8 possessed an absolute veto (see Table 1, above). For all other
legislation, the second chamber possessed only a suspending veto. Thus, the states had re-
couped some political influence lost in the Weimar years. The same picture emerges in the
question of taxation.
The Catholic ‘South-West’ current of German Christian Democracy, which became domi-
nant in the west after Germany was divided into two nations, had always been in favour of
federalist devolution – mainly motivated by its opposition to the centralist hegemony of
protestant Prussia. These preferences for greater regional autonomy had been reflected in
proposals at the constitutional convention to return to the fiscal status quo of Imperial Ger-
many. According to these proposals, the central government would have again become to-
tally dependent on transfers from the states’ budgets (cf. Renzsch 1991:55-74). Yet, the
German Social Democratic Party – not least because of its expectation of winning the first
national election in 1949 – struggled vehemently for a more centralist distribution of reve-
nue – while the pro-federalism stance of the Christian Democratic Party (the CDU) was, in
turn, motivated by the very same expectation that the first democratic government of the
Federal Republic would be led by Social Democrats. But resistance to a decentralized solu-
tion also cut across party-lines. Large differences between city-states like Hamburg or Bre-
men on the one side and states like Bavaria on the other, different economic starting condi-
tions as well the varying extent to which states were affected by the war (Schleswig-
8 The German Bundesrat does not follow the Senate-model with elected senators, but is composed of
delegates of the state governments, with votes weighted differently according to population size.
Holstein was flooded with refugees from the East, the city states Bremen, Hamburg and
Berlin had been heavily destroyed by Allied bombing) rendered strict fiscal devolution al-
most impossible. Under these circumstances, and given that the states resisted all attempts
to rearrange borders and fuse West Germany’s 11 states into a few viable entities, federalist
principles could only be defended through a regime of centralized but joint taxation. There-
fore, the welfare state- and the compromise over federalism can be seen as two sides of the
same coin. Given the varying degrees of expenditure need, welfare state devolution was
highly problematic. Given the differing revenue potential of the states, fiscal federalism was
not a realistic option. This established the paradoxical institutional setup of the Federal Re-
public: autonomy of the states could only be secured by strengthening their influence on
nationally uniform legislation and by establishing a complex national tax-and-spend ar-
When the CDU formed the first German government under Chancellor Konrad Adenauer,
the basic compromise between ‘centralists’ and ‘federalists’ was already in place. The most
important taxes including value added tax (VAT), income tax and corporate income tax
were either shared between central state and states (income and corporate tax) or were the
exclusive province of the central government (VAT). All in all, a rough and relatively stable
55 per cent (central government) to 45 per cent (states) distribution of total tax revenue was
established (Renzsch 1991:13). Today the federal government, the states and local authori-
ties share the personal income tax according to a 42.5/ 42.5/ 15 % formula and the corporate
income tax and the VAT are split between central and regional governments according to a
50/ 50 and a 51/ 47 ratio (with local authorities receiving 2 % of the VAT), respectively.
Regional differences in tax revenue due to differences in economic development are equal-
ized by a complex system of vertical transfers from the central state to ‘needy’ states com-
bined with horizontal transfers from rich to poor states (Renzsch 1991; Hidien 1999). These
fiscal equalization schemes guarantee that no state has a per-capita tax revenue below 95 %
of the national average. With respect to tax-legislation, the Weimar status quo remained in
place: for taxes comprising 30 % of total revenue, either central government, the states or
municipalities can claim exclusive responsibility. For the remaining 70 % of joint taxes,
legislation is national, but states have a quite powerful right of co-legislation due to their
position in the Bundesrat. In fact, since no major tax – other than social insurance contribu-
tions – falls into the exclusive competence of either the central or regional government(s),
the central government’s tax policy is dependent on the consent of the states – and in times
of ‘divided government’ also on the consent of the opposition (Ganghof 2004).
The fact that the political decisions concerning the fate of the German welfare state came
much later than the fiscal compromise between the central government and the states again
demonstrates that the states conceived of social policy as not being genuinely part of their
own policy domain. In the early 1950s, with a couple of rather unspectacular legislative
measures, the Adenauer government reclaimed central authority for the social insurance
schemes that before 1945 had been organized at the national level. In 1951, the accident
insurance scheme was re-integrated into the administrative domain of the Federal Labour
Ministry, in 1952 a national Labour office was constituted and in 1953 the national white
collar worker pension scheme was re-established in Berlin as the successor of the former
Imperial Insurance Office for white collar-workers (Kahlenberg/Hoffmann 2001:118f.; Fre-
rich/Frey 1993:43, 84). While there was conflict concerning all three measures – mainly
over questions concerning the representation of employers and unions in the administrative
bodies of the schemes (‘self-administration’) – a manifest centre/ periphery conflict dimen-
sion was largely absent from the agenda (Hockerts 1980). True, in the case of the accident
insurance, the states warned that centralization of responsibilities would lead to a further
“puffing up of the federal bureaucracy” (Kahlenberg/Hoffmann 2001:118), and in the case
of unemployment insurance system, they favoured a more decentralized solution, in which
independent regional labour offices would only be engaged in a loose working cooperation
at the national level. However, as soon as the federal state promised to take over the entire
administrative staff of the regional labour offices, the resistance to centralist solutions
quickly evaporated (Trampusch 2000). Apparently, the Länder were not very principled in
their pro-federalism position.
The successful claim of the Adenauer government for supremacy in the field of social pol-
icy was not a consequence of a weakened position of the Länder after 1949. Indeed, the
opposite was the case. Partly as a consequence of their institutional primogeniture, but also
in terms of constitutional authority, the states were in a seemingly powerful position. This
was especially due to their right to an absolute veto on all legislation with an impact on
Länder-administration or Länder-finance. Initially, the veto was thought to apply to no more
than 10 % of all legislation, but today – as a result of extensive legal and political interpre-
tations – about 55 % of all laws are seen as requiring the consent of the second chamber
(Schindler 1999:2430f.). In other words: if no compromise can be found between parlia-
ment and second chamber, the law cannot be enacted. True, Article 72 of the Basic Law
established far reaching legislative competencies for the central government, but these were
balanced by no less far reaching rights of co-legislation for the states (Oeter 1998:123f.).
Now that hegemonic Prussia had ceased to exist, this right of co-legislation was of much
greater potential significance. For the government, it became harder to form majorities in
the upper chamber and the possibility of veto in Bundesrat became more real.
With respect to social policy, the late 1940s and early 1950s essentially saw the re-
foundation of the German welfare state along the traditional Bismarckian principles, with
obligatory social insurance, employment-based membership, contribution finance, organisa-
tional fragmentation etc. (Hockerts 1980). Many perceived this as a scandalous anachro-
nism. While contemporaneous reforms in Britain and Scandinavia established allegedly
‘modern’, uniform and universal, central, tax-financed systems of social protection, the
defeated and economically weak Germany obviously thought it could afford to stick with
the old, expensive, Byzantine system with all its outdated status differentiation and bureau-
cratic overkill. Especially contentious was the restoration of the ‘feudal’ differentiation be-
tween a blue collar- and a white collar-worker insurance branch (both in the pension and
health insurance schemes). From early on, two modifications of the pre-war status quo
emerged as a kind of institutional response to this critique: 1) although organisationally
separated, entitlements in both schemes became legally ‘assimilated’,9 and 2) internal fiscal
transfers, as well as transfers between the different schemes, were introduced to ensure that
equal welfare entitlements would not translate into varying contribution levels between
regions or social insurance schemes. Initially these financial transfer schemes were volun-
tary; later they became obligatory (for an overview see Schmähl 2001; Henke 2001).
The financial transfer schemes proved to be of special importance for the further develop-
ment of the relationship between German federalism and the Bismarckian welfare state,
since uniform entitlements and equal contributions backed by fiscal transfers that levelled
regional or economic disparities largely ruled out ‘race to the bottom’ dynamics in the so-
cial policy arena. A nationally uniform contribution rate in the pension insurance made fis-
cal transfers between pension funds necessary. Finally, in 1969 in the wake of Germany’s
first post-war recession, the ‘good’ risk pool of the white collar workers’ insurance fund
was fully fiscally fused with the comparatively ‘bad’ blue collar work insurance scheme
(Frerich/Frey 1993:52f.). Another motive complemented the ‘equal contribution rate’ argu-
ment. Fiscal pooling was introduced largely to provide financial relief for the central state,
because transfers from the relatively well-off white collar-worker pension fund to the defi-
cit-ridden blue collar-worker scheme reduced the need for state subsidies from the general
state budget. Similar fiscal interests motivated the integration of the miners’ pension
schemes (Frerich/Frey 1993:51). Thus, the apparent picture of institutional fragmentation
conceals the almost complete pooling of financial resources (VDR 2003). With nationally
uniform contribution rates, however, social insurance contributions developed more and
more into national quasi-taxes. The high political barriers for tax legislation in Germany can
then explain why subsequently much of Germany’s fiscal adjustment to a more unfavour-
able economic environment since the first oil-crisis in 1973 happened through the adjust-
ment of welfare state revenue and spending.
The most striking example of the often problematic interplay between the regulative idea of
nationally uniform living conditions, the high degree of financial autonomy of the German
welfare state and the particularistic fiscal interests of both central and states governments
was provided by German unification in 1990 (Czada 1995). On the one hand German unifi-
cation exemplified how the German “unitary welfare state easily overcomes federalist di-
viding lines“ (Oeter 1998:534, translation by the author), given the fact that much of the
East’s rapid catch-up in terms of living conditions and income has to be attributed to the
operation of the German welfare state. On the other hand German unification also demon-
strated the price that had to be paid for this rapid catch-up; namely, a vast expansion of so-
9 It is, therefore, wrong to ascribe to the Adenauer pension reform the intention to “restore status differ-
ences”. See Esping-Andersen 1990:25. In a number of respects, blue collar-workers were for the first
time put on an equal footing with white collar-workers. See Prinz 1991.
Massive disparities existed between wages and labour productivity in the area of the ‘old’
Federal Republic and the five new Eastern Länder. However, the electoral appeal of a quick
fix of Eastern living standards without substantial tax increases or a significant growth in
public debt made it more politically attractive to impose ‘hidden’ tax increases by way of
substantial increases in social insurance contributions. But burdening the social insurance
funds with the costs of German unification was not only the result of a populist ‘read-my-
lips’ strategy on the part of the Kohl-government motivated by the exceptional density of
elections in the decisive year of unification, in 1990 (with a total of 17 national, regional
and local elections) (Schwinn 1997; Sturm 1998; Zohlnhöfer 2001). The other side of the
story is that in the process of unification the Western Länder were quite successful in pro-
tecting their fiscal interests (Altemeier 1999). As Table 3 shows, according to conservative
estimates roughly a quarter of all financial West-East transfers in the wake of German uni-
fication had to be borne by the social insurance schemes, i.e. ultimately by the Western con-
tributors. West/East welfare transfers were more than three times higher than transfers out
of the Western states’ budgets. This is a spectacular instance of the way in which federalist
veto structures in taxation and revenue sharing (through the Finanzausgleich, the fiscal
equalization scheme) translated directly into ‘welfare state growth’.
Table 3: West/East Transfers in billion DM, 1991-1999, central and regional govern-
ments and social insurance funds10
1. Federal Govern-
ment 75.1 90.0 115.7 115.9 136.7 136.7 129.7 130.8 136.3 1066.9
2. States 5.3 5.7 10.3 13.5 11.2 11.3 11.6 11.5 11.6 92.1
3. Social Insurance 18.7 34.2 23.0 29.8 33.3 30.9 34.7 31.9 36.0 272.2
4. Net Total Transfers 109.9 133.5 150.6 148.8 141.3 137.7 135.1 132.6 140.3 1229.9
5. Social Insurance
as Per cent of Net
17.0 25.6 15.3 20.0 23.5 22.5 25.7 24.1 25.7 22.2
As a consequence, contribution rates to social insurance schemes steeply increased and wel-
fare state spending revealed a strong upward trend. Whereas the combined social insurance
contribution rates of employers and employees amounted to around 35.6 % of net wages
10 With respect to transfers to the East data differ. I report here rather conservative estimates, taken from
Ragnitz 2003; Bach/Vesper 2002, report slightly higher numbers; again slightly different estimates
are provided by the Deutsche Bundesbank 1996
(up to an upper limit) in 1990, today they are at a substantial 42 % (BMG, table 7.7). When
it comes to real financial burdens, not nominal rates, Figure 4 shows that the overall finan-
cial welfare burden on employers and employees had remained fairly stable over the 1980s
with even a small downward trend at the end of the decade, whereas after unification in
1990 employers and employees had to bear steeply increasing costs.
Figure 4: Absolute volume of contribution payments of employers and employees, as a
Percentage of GDP, 1975-2002 (BMG, table 7.6)
Yet, this is only a particular flagrant example for a much more long-term trend of political
cost externalization at the expense of the contribution-financed welfare state – a trend we
already had been able to observe for the Weimar period (see above, Section 3). In fact, since
the worsening of general economic conditions after the first oil-shock German governments
invariably have sought to shift costs onto the contribution financed welfare state rather than
taking the electoral unpopular and politically complicated step of raising taxes or of cutting
expenditures (Nullmeier 1992; Manow/Seils 2000; Trampusch 2003). From 1965 to 2002
combined revenue from taxes and social insurance contributions has risen from 32.8 % to
38.2 % of GDP, but whereas tax revenue decreased from 23 to 20.8 % during this period,
revenue from social insurance contributions increased from 9.8 % to 17.4 % or nearly dou-
bled (cf. Manow/Plümper 2003). As Figure 5 reveals, the tax revenue of local and regional
governments remained relatively stable over the last 30 years. The major revenue shifts
have taken place between the central state and the social insurance funds: “insurance-based
programmes … have absorbed a steadily growing share of social expenditures since 1970
… entirely tax-funded programmes … lost a lot of ground” (Leibfried/Obinger 2003). Ger-
man unification accelerated, but did not initiate this process.
Figure 5: Total Tax Revenue, relative Shares of Central, State, Local Government and
Social Security Funds, Germany 1973-2002 (Source: OECD, Revenue Statis-
As figure 5 makes clear, there is indeed ample evidence of expansionary expenditure ten-
dencies in German-style ‘cooperative’ federalism. These findings with respect to welfare
state finance mirror the more general point that federalism can “exacerbate collective action
problem” and “might undermine fiscal discipline” if the fiscal and political responsibilities
of the different layers are not clearly separated from each other (Rodden/Wibbels
2002:495f.). It is, therefore, hardly a surprise that the veto-point hypothesis so strongly em-
phasised in the literature on the political consequences of federalism does not seem to fit the
German case very well. And this is so despite the fact that Germany can be considered a
‘critical case’ for the veto point thesis with a strong second chamber and the frequent occur-
rence of countervailing majorities (Lijphart 1984; Money/Tsebelis 2003).11
Yet, from 1972 and 1994, only 34 laws has been successfully blocked by the federal cham-
ber, either because the majority in the Bundesrat exercised an absolute veto (N=29) or be-
cause the parliament abstained from overruling the federal chamber’s suspending veto
(N=5). Only 10 of these 34 laws had anything to do with social policy, and none of them
represented a major reform act (Schindler 1999:2434ff.). This is not a particularly impres-
11 In Germany the central government has enjoyed a political majority in the federal chamber for only 9
of the 31 years since the end of the “golden age” of the welfare state, from 1972 to the present. These
‘happy years’ of political alignment of the first and second chamber fell basically into the first two
(and a half) terms of the Christian Democratic government under Chancellor Kohl, from 1982 to
sive demonstration of federalism’s veto power. Of course, looking at outright legislative
failure gives us only a small part of the picture. Since the impact of federal veto structures
presumably lies in that they have the capacity to ‘water down’ substantive reforms and
force the government to agree on lowest common denominator policies, it is clearly not
enough to focus exclusively on failed legislative initiatives. Even with only little strategic
capacity, a government can anticipate resistance in the second chamber and will draft legis-
lation accordingly, in order to avoid complete political stalemate. Therefore any ‘evidence’
for the veto-point thesis must be based on hard-to-prove and hard-to-falsify counterfactual
reasoning of the kind: ‘would government x have proposed a bill substantially different
from the bill it introduced had it acted in a unicameral rather than a bicameral system?’
Stalemate politics can be motivated either by calculations of partisan advantage or by states’
special interests. If we take into account that the two mass-member parties, the German
Social Democratic Party and the Christian Democratic Party (SPD and CDU), are pro-
grammatically quite close to each other in questions of social policy (see for instance Budge
et al. 2001:235), the rarity of partisan-motivated policy blockade is not that surprising. In
fact, almost all major social policy reform acts since 1949 have been enacted with the sup-
port of the major opposition party, no matter whether this was the SPD or the CDU. This
might be not seen as particularly surprising in the case of the generous expansion laws of
the 1950s and 1960s. However, there was also a broad consensus between SPD and CDU in
respect of many of the major retrenchment laws in the decades after 1980, including the
health reform of 1988, the pension reform of 1990, and the present health reform of the
second Schröder government. Moreover – contrary to the ‘new politics of the welfare state’
thesis – the introduction of a completely new social insurance branch, the long-term care
insurance (LTCI) in 1994, two decades after the ’golden age’ of welfare state expansion
was over, also relied on consensus between Christian and Social Democrats. Indeed, in the
case of long-term care, the conservatives were the initiators of expansionary reform. 12
True, as Jens Alber has shown, broad political consensus is less likely if welfare cuts are on
the agenda than if the legislation is expansionary (1989:262, table 43), a finding that ap-
pears trivial only as long as one does not fully recognize how much this tells us about the
non-trivial fact that Social Democrats and Christian Democrats largely share a general pro-
welfare state attitude. However, it is also the case that major legislative cost containment
efforts have frequently been enacted by super-majorities. In international comparison Ger-
many is a country in which welfare retrenchment started early – already under Social De-
mocratic chancellor Schmidt in the second half of the 1970s – and where retrenchment “ac-
tually was quite substantial” (Leibfried/Obinger 2003: 214; see for further evidence among
12 The introduction of the LTCI is another example for cost-externalizing social policy at the expense of
the welfare state. Financed in the traditional Bismarckian fashion through contributions paid by em-
ployers and employees, the LTCI brought substantial financial relief for the budgets of the munici-
palities and states, since support for the handicapped, frail or elderly that previously had been fi-
nanced by tax-financed social assistance was now to be covered by the new, ‘fifth’ social insurance
branch; see Götting et al. 1994.
many Siegel 2002, chapter 9; Alber 1998; Alber 2001). Both the health sector and the pen-
sion insurance were subjected to continuous cost containment measures (see also Jochem
2001). As a welfare state spender Germany had receded from a leading position to the mid-
dle-field by the late 1980s. German unification then once again supplied a strong expan-
sionary momentum. But both before and after 1990, successful cost containment was
thwarted by two other countervailing trends. First, high and persistent mass-unemployment,
costly medical progress and – lately – also the first impact of rapid population ageing have
operated together to significantly increase the demand for social spending. Hence, success-
ful cost-containment was successful only in reducing the growth rate of expenditure. Sec-
ond, and more important in the context of this analysis, cost-cutting had its main effects on
the tax-financed state share of total social spending, but had a much lesser impact on contri-
bution-financed share. In fact, the importance of contributory finance increased, since the
central government sought to ease its own fiscal stress by shifting social spending obliga-
tions out of the public budget and into the special budgets of the social insurance funds, and
by accepting (automatic) contribution hikes as substitutes for legislating unpopular and
politically blocked tax hikes.
This last point is of especial importance since it highlights the important nexus between
taxation and welfare state finance. This linkage helps explain the otherwise puzzling com-
bination of relatively successful cost-containment before 1990 and significant cost increases
thereafter. In principle, in response to fiscal stress, the German government either could
increase debt, cut costs or increase revenue. The strong Bundesbank and, since 1992, the
Maastricht criteria have set quite clear limits to the debt option (see for the general argu-
ment Manow/Plümper 2003). On the other hand, legislating taxes hikes in order to raise
revenue is not only notoriously unattractive in electoral terms, but, in the context of the
German federalist system of joint taxation, strong bicameralism and frequent divided gov-
ernment,13 also extremely complicated in partisan political terms. Welfare retrenchment did
occur to some degree, but was never sufficient to close the ever widening revenue / expen-
diture gap. The welfare state provided a fiscal pressure-valve in two respects. First, public
budget cuts plus task-delegation at the expense of the social insurance funds reduced the
fiscal involvement of the central state. Second, semi-automatic increases in social insurance
contributions substituted for legislated tax hikes. This further diminished the importance of
tax-financed social spending with strongly regressive consequences. Social contributions
are already levied from low incomes and given that welfare transfers are either not taxed at
all or only at a very low level, income taxation does not correct for the regressive effects of
the Bismarckian system. In terms of net social spending Germany – on a par with Sweden
– is still the world’s most ‘generous’ welfare states and is, yet at the same time one of the
world’s least redistributive social policy systems (Adema 1999; Ganghof 2004).
13 In this context it seems worth mentioning that only shortly after unification in October 1990 the Kohl-
government lost its majority in the upper house because of the Hessen-elections in April 1991.
Inter-party consensus with respect to steady incremental retrenchment has not prevented
fierce clashes between government and opposition on issues with a high electoral salience.
The massive campaign of the Christian Democratic opposition against pension cuts in 1976
is one example as is also the campaign of the SPD against the retrenchment proposals of the
Kohl government in 1996 (reduction of sick pay, a lowering of employment protection for
small firms, pension cuts by introduction of a ‘demographic formula’, an increase in retire-
ment age, etc.). In 1976, as in 1996, electoral motives were clearly dominant. In 1998 the
Social Democrats benefited from the fact that finally the Christian Democratic– liberal coa-
lition had begun to engage in more than incremental tinkering with welfare state costs. For
the first time, the Social Democrats’ continuous lament that the Kohl-government followed
a neo-liberal course of deregulation could point to, at least, some evidence. Achieving of-
fice with the promise to modernize Modell Deutschland without neoliberal neglect of equal-
ity and social security, the Schröder-government quickly delivered the promised ‘counter-
reform’, i.e. it undid the few retrenchment measures of the last Christian Democ-
ratic/Liberal administration.14 It soon became clear, however, that this was not enough.
Space prohibits a thorough description of all reform measures, but I think it is fair to say
that taken together the tax, pension, health and labor market reforms of the first and second
Schröder government cut much deeper into the existing structures of the German welfare
state and political economy than all the reforms enacted during the 16 years of CDU/FDP
If at all, only a modified version of the veto-point theory is compatible with this finding,
given that the red-green coalition only enjoyed a very short period of 6 months majority
status in the Bundesrat compared to the 9 years of concurrent majorities in both houses dur-
ing the Kohl chancellorship from 1982 to 1991. Since April 1999, the Schröder government
has been confronted with a hostile upper house, as had been the Kohl government from
1991 to 1998. How, then, can one explain their different reform records? Intense political
competition between two mass membership parties, both of which attempted to woo the
pro-welfare median voter would lead one to expect at best incremental changes, more often,
perhaps simply political stalemate. Furthermore, we would expect governments to shy away
from harmful measures or to take action only where the opposition could be brought on
board, given that the many state elections (16), spread over the parliamentary term, provide
strong political disincentives to any programme of painful reforms. Governments try to
avoid unpopular cutbacks since they fear electoral punishment, which might result in the
loss of power at the state level and, as a consequence, in an oppositional majority in the
upper house (Lehmbruch 1998; Burkhart 2003; Dinkel 1977). Differences in reform intensi-
ties therefore have to be mainly attributed to the changed party constellation: the Christian
14 For a (highly critical) overview over the welfare reforms of the first Schröder government see
Schmidt 2003. See also the contributions by Rose (labour law), Heinelt (labour market), Buhr (social
assistance), Nullmeier (pensions), Bleses (family), Brandhorst (health), West (education), Leitner
(gender) and Vogel/ Wüst (immigration in Antonia Guhr/ Martin Seeleib-Kaiser (eds.), 2003).
Democrats in government were electorally more vulnerable to the charge of being the
‘gravedigger’ of the welfare state. Now, in turn, the SPD is forced to be an initiator of re-
form, while the bourgeois parties find it difficult to ‘leapfrog’ the left by protecting the
status quo (Kitschelt 2001). What before 1998 was a situation of mutual stalemate has
turned into an overbidding game in which the CDU attempts to top the reforms of the red-
green coalition with even more radical proposals.
Compared to antagonistic party competition, political gridlock in the federal chamber as a
result of special Länder-interests seems to occur more often. Case studies often stress that
the veto power of the regional governments in the upper house has regularly prevented
more far-reaching reform measures from being enacted. The health care sector in particular
has regularly been mentioned as exemplifying this tendency (Webber 1988; Webber 1989;
Rosewitz/Webber 1990). However, even here, the empirical evidence is not particularly
strong. Neither is it true that health sectors in which the states have direct interests (hospital
sector, pharmaceutical industry), seen fewer cutbacks than other sectors such as physicians’
reimbursement, where the states’ interests are weaker (cf. Döhler/Manow 1997). Nor is
there much of a difference between social insurance branches in which the states have a
veto position and those where they have none. With respect to the first point we can observe
that in the hospital sector, sickness funds spent 29.4 % of their total outlays on hospitals in
1974, at the onset of the ‘period of permanent austerity’. In 2000 they spend only slightly
more, 32.2 %. Change in the pharmaceuticals sector is even less evident, with the share of
total expenses remaining more or less stable with 23.2 % in 1974 and 24.7 % in the year
2000 (Statistical Yearbook, various years). As to social insurance branches, much the same
picture emerges. In respect of most laws in the health care sector, the states have an abso-
lute veto, since here legal measures usually affect regional administrative responsibilities in
one way or another. This is typically not the case in the pension insurance scheme. Yet, if
we compare the spending dynamics of these two sectors (both similarly affected by demo-
graphics), we cannot see much of a difference. The ratio of pension to health care expendi-
tures in 1974 was 1.8, in the year 2000 it was at 1.73 (Statistical Yearbook, various years).
Thus, health care costs grew relative to pension costs, but not strongly enough to justify a
convincing argument about the influence of federal veto points.
In more general terms, the federalism as a veto-point thesis has substantial problems in ac-
counting for Germany’s post-war welfare state development, given the fact that federal state
structures seem to not have hindered Germany from becoming the most generous welfare
state of Western Europe in the 1950s and 1960s (Alber 1982, see Table 4) and did not stand
in the way of quite substantial retrenchment relative to other OECD nations in the period
1970 and 1990.
Table 4: Total Social Expenditures as a percentage of GDP, 1950-1970, 11 countries
(Source: ILO, Costs of Social Security)
1950 12,57 7,84 6,64 15,34 10,93 8,73 8,65 5,70 8,72 7,32
1955 15,76 9,40 6,77 14,81 11,23 n.d. 9,55 7,44 9,89 n.d.
1960 15,74 11,09 8,55 n.d. 12,53 12,22 11,93 9,35 10,94 9,00
1965 18,35 11,20 10,33 16,56 17,93 13,87 17,99 16,00 10,95 13,67 10,10
1970 17,86 15,69 12,48 18,35 15,80 17,69 20,89 15,46 18,63 14,32
The overall German spending trajectory – first fast growth and very high spending levels,
then moderate growth at medium spending levels, finally again strong expansion in the
wake of German unification – is not well accounted for by the veto-point thesis which
would predict below average growth in the golden times of welfare state expansion up to
the mid-1970s, but, from that time onwards, above average growth rates due to the political
difficulty of effecting substantial welfare retrenchment in federalist polities.
“Decentralized redistribution is self-defeating” (Prud´ homme 1995:202). In this sense, it is
true that federalism and welfare states represent potentially antagonistic institutional set-
tings. Federalism can be a powerful institutional impediment to substantial redistribution.
And this is exactly how federalism seems to have impacted on welfare state development in
the English-speaking liberal nations with systems of interstate federalism. Yet, as the Ger-
man case made clear, as soon as political actors pursue extensive redistributive aims by
superimposing nationwide tax and transfer systems on still federally fragmented political
accountability structures, the strategic logic of the interplay between federalism and welfare
state may change profoundly. Here, the diffusion of responsibilities and the incentives for
shifting costs between central government, states and the welfare state instead are condu-
cive to an ‘overgrazing of the fiscal commons’ especially if the welfare state is – as is the
case in Germany – largely financed by contributions. Instead of the restrictive influence
predicted by the standard versions of veto-player theory, federalism here leads to expan-
sionary spending dynamics and a low degree of fiscal discipline. I have argued that federal-
ist veto structures has played a role in Germany’s welfare state development in the period of
retrenchment, but in a way not foreseen in the literature. Given that the German states have
a powerful veto position when it comes to legislation on joint taxes, which account for 70 %
of total tax revenue, but have no say in matters of social insurance contributions, increasing
welfare state revenue offered relief in a situation of severe fiscal stress – particularly, but
not exclusively in the wake of German unification.
Founded with the aim of furthering national unification, the German welfare state indeed
has become the great national unifying institution. The feedback effects of the welfare state
on both living conditions in Germany and on the functioning of German federalism itself
are enormous. The welfare state together with the national tax system has effectively pre-
vented inter-regional economic competition. These national achievements have been so
successful that proposals to reverse the trend and to introduce what has been called federal-
ist competition not only in education and taxation, but also in the welfare state arena, have
become ever more articulate (cf. Münch 1998; Pitschas 1994; Papier 1995). Competition
would also – so it is hoped – re-introduce experimentation and learning into German feder-
alism, now largely absent because of nationwide regulation and encompassing fiscal balanc-
ing schemes.15 However, political support for these calls for welfare state devolution is
weak and will remain so in the future, given that the relevant actors are fully aware of the
distributional consequences. The new Eastern states as well as the less affluent Northern
states can only lose from political and fiscal devolution. Thus, even if, in Germany, federal-
ism does not hinder welfare retrenchment, it still makes it difficult to effect major structural
reforms. In Germany’s “grand coalition state” (Schmidt 2002), there is still no alternative to
If we refer back to the two ‘points of qualification’ noted in the introduction, the preceding
paragraphs have shown that neither the implicit ‘interstate-competition’-assumption nor the
implicit ‘incongruence of political preferences’– assumption of most contributions to the
subject have had much bite in the German case. It is, therefore, hardly surprising that Ger-
man federalism has not prevented the Bismarckian welfare state from becoming one of the
world’s most expensive, generous, and encompassing systems of social provision. Corre-
spondingly, and contrary to the received wisdom of the federalism thesis, federalism has
also not prevented substantial retrenchment in the era of the ‘new politics of the welfare
15 Hence it is not surprising that the only instances of policy experimentation and mutual learning are to
be found outside the traditional social insurance framework, especially in local care for the elderly
and local social assistance; see Schmid 1990; Alber/Schölkopf 1999.
Adema, Willem, 1999: “Net social expenditure”, Labour Market and Social Policy Occa-
sional Papers 39. Paris: OECD. (http://www.oecd.org/els/papers).
Alber, Jens, 1982: Vom Armenhaus zum Wohlfahrtsstaat. Frankfurt am Main: Campus.
Alber, Jens, 1988: „Die Gesundheitssystem, der OECD-Länder im Vergleich“, in: Manfred
G. Schmidt (Hg.), Staatstätigkeit. International und historisch vergleichende Analysen.
Opladen: Westdeutscher Verlag, 116-150.
Alber, Jens, 1989: Der Sozialstaat in der Bundesrepublik 1950-1983. Frankfurt/New York:
Alber, Jens, 2001: “Recent Developments of the German Welfare State: Basic Continuity of
Paradigm Shift?” ZeS Working Paper 6/2001. Center for Social Policy Research, Uni-
versity of Bremen: Bremen.
Alber, Jens; Schölkopf, Martin, 1999: Seniorenpolitik. Die soziale Lage älterer Menschen
in Deutschland und Europa. Amsterdam: Fakultas.
Alesina, Alberto; Glaeser, Edward; Sacerdote, Bruce, 2001: Why doesn’t the US have a
European-style welfare state? NBER working paper w8524. Boston, Mass.: National
Bureau for Economic Research.
Álvarez, Pablo B., 2003: Decentralization and income Inequality. Madrid: Instituto Juan
Altemeier, Jens, 1999: Föderale Finanzbeziehungen unter Anpassungsdruck. Verteilungs-
konflikte in der Verhandlungsdemokratie. Frankfurt/New York: Campus.
Bach, Stefan; Vesper, Dieter, 2002: „Finanzpolitik und Wiedervereinigung – Bilanz nach
10 Jahren“, DIW Vierteljahresheft 2/2002, vol. 69: Zehn Jahre deutsche Währungs-,
Wirtschafts- und Sozialunion. Berlin: Duncker & Humblot, 194-223.
Benz, Arthur, 2001: „Themen, Probleme und Perspektiven der vergleichenden Föderalis-
musforschung“, in: Arthur Benz; Gerhard Lehmbruch (Hg.), Föderalismus. Analysen in
entwicklungsgeschichtlicher und vergleichender Perspektive. PVS Sonderheft 32/2001.
Opladen: Westdeutscher Verlag, 9-50.
BMG (Bundesgesundheitsministerium/ Ministry of Health), various years: Daten des Ge-
sundheitswesens. http://www.bmgesundheit.de ; or Baden-Baden: Nomos.
Böckenförde, Ernst-Wolfgang, 1980: „Sozialer Bundesstaat und parlamentarische Demo-
kratie“, in: Jürgen Jekewitz et al. (eds.), Politik als gelebte Verfassung. FS für Friedrich
Schaefer. Opladen: Westdt. Verlag, 182 ff.
Boldt, Hans, 2003: „Die Wiederaufnahme der deutschen föderativen Tradition im Parla-
mentarischen Rat 1948/49“, Zeitschrift für Staats- und Europawissenschaften: 505-526.
Bonoli, Giuliano, 2000: The Politics of Pension Reform. Institutions and Policy Change in
Western Europe. New York: Cambridge University Press.
Borscheid, Peter; Drees, Annette (Hg.), 1988: Historische Statistik von Deutschland. Band
IV: Versicherungsstatistik Deutschlands 1750–1985. St. Katharinen: Scripta Mercurae.
Brennan, Geoffrey; Buchanan, James, 1988: Besteuerung und Staatsgewalt. Hamburg:
Steuer- und Wirtschaftsverlag.
Budge, Ian; Hans-Dieter Klingemann, Hans-Dieter; Volkens, Andrea; Bara, Judith; Tanen-
baum, Eric, 2001: Mapping Policy Preferences. Estimates for Parties, Electors, and
Governments 1945-1998. Oxford: Oxford University Press.
Bürger, Hans, 1930: Sozialversicherung und Reichshaushalt. Darstellung und Kritik. Ber-
Burkhart, Simone, 2003: Parteipolitikverflechtung – eine systematische Untersuchung über
den Einfluss der Bundespolitik auf Landtagswahlentscheidungen von 1976 bis 2002.
Castles, Francis G., 1989: “Social Protection by other Means: Australia’s Strategy of Cop-
ing with external Vulnerability”, in: Francis G. Castles (ed.), The Comparative History
of Public Policy: 16-55.
Castles, Francis G. (ed.), 1993: Families of Nations: Patterns of Public Policy in Western
Democracies. Aldershout: Dartmouth.
Castles, Francis G., 1999: “Decentralization and the Post-war Political Economy”, Euro-
pean Journal of Political Research 36, 27-53.
Castles, Francis G., 2000: “Federalism, fiscal decentralization and economic performance”,
in: Ute Wachendorfer-Schmidt (ed.), Federalism and Political Performance. London:
Castles, Francis G.; Mitchell, Deborah, 1993: “Worlds of Welfare and Families of Nations”,
in: Castles, Francis G. (ed.): Families of nations. Aldershot: Dartmouth.
Castles, Francis G.; Mitchell, Deborah, 1996: “Needs-based strategies of social protection
in Australia and New Zealand”, in: Gösta Esping-Andersen (ed.), Welfare states in tran-
sition national adaptations in global economies. London: Sage, 88-115.
Crepaz, Michael L., 1992: “Corporatism in Decline? An empirical analysis of the impact of
corporatism on macroeconomic performance and industrial disputes in 18 industrialized
countries”, Comparative Political Studies 252: 139-168.
Czada, Roland, 1995: Der Kampf um die Finanzierung der deutschen Einheit. MPIfG Dis-
Deutsche Bundesbank, 1996: „Zur Diskussion über die öffentlichen Transfers im Gefolge
der Wiedervereinigung“, in: Deutsche Bundesbank Monatsbericht 48.
Dinkel, Reiner, 1977: „Der Zusammenhang zwischen Bundes- und Landtagswahlergebnis-
sen“, Politische Vierteljahresschrift 18: 349-359.
Döhler, Marian; Manow, Philip, 1997: Strukturbildung von Politikfeldern. Das Beispiel
bundesdeutscher Gesundheitspolitik seit den fünfziger Jahren. Opladen: Leske +
Esping-Andersen, Gösta, 1990: The Three Worlds of Welfare Capitalism. Cambridge, UK:
Esping-Andersen, Gösta, 1996: “Welfare states without work”, in: Gösta Esping-Andersen
(ed.), Welfare states in transition national adaptations in global economies. London:
Evans, Peter; Rueschemeyer, Dietrich; Skocpol, Theda, 1985: Bringing the state back in.
Cambridge: University Press.
Flora, Peter, 1983: State, Economy, and Society in Western Europe, 1815-1975. Volume I:
The Growth of Mass Democracies and Welfare States. Frankfurt: Campus/ London:
Macmillan/ Chicago: St. James.
Flora, Peter; Alber, Jens, 1981: “Modernization, Democratization, and the Development of
welfare states in Western Europe” in: Peter Flora; A. J. Heidenheimer (eds.), The Devel-
opment of Welfare States in Europe and America. New Brunswick: Transaction Books.
Frerich, Johannes; Frey, Martin, 1993: Handbuch der Geschichte der Sozialpolitik in
Deutschland, Bd. 1: Von der vorindustriellen Zeit bis zum Ende des Dritten Reiches.
And Bd. 3: Sozialpolitik in der Bundesrepublik Deutschland bis zur Herstellung der
Deutschen Einheit. München: Oldenbourg.
Führer, Carl Christian, 1990: Arbeitslosigkeit und die Entstehung der Arbeitslosenversiche-
rung in Deutschland 1902-1927. Berlin: de Gruyter.
Ganghof, Steffen; Bräuninger, Thomas, 2003: Partisan Veto Players in Australia, Den-
mark, Finland and Germany: Government Status and Legislative Behavior. MPIfG
Working Paper 03/11.
Ganghof, Steffen, 2004: Wer regiert in der Steuerpolitik? Einkommensteuerreform in
Deutschland zwischen internationalem Wettbewerb und nationalen Verteilungskonflik-
ten. Frankfurt/New York: Campus.
Götting, Ulrike; Haug, Karin; Hinrichs, Karl, 1994: “The long road to long-term care insur-
ance in Germany”, in: Journal of Public Policy 14: 285-310.
Guhr, Antonia; Seeleib-Kaiser, Martin (eds.), 2003: Sozial- und Wirtschaftspolitik unter
Rot-Grün. Opladen: Westdeutscher Verlag.
Henke, Klaus-Dirk, 2001: “Der parafiskalische Finanzausgleich, dargestellt am Beispiel der
Gesetzlichen Krankenversicherung“, in: Klaus-Dirk Henke; Winfried Schmähl (Hg.),
Finanzierungsverflechtung in der sozialen Sicherung, Baden-Baden: Nomos, 77-93.
Henning, Hans, 1977: „Sozialpolitik, Geschichte III“, in: Handwörterbuch der Wirtschafts-
wissenschaften Bd. 7: 85-110.
Hesse, Konrad, 1962: Der unitarische Bundesstaat. Karlsruhe: C.F. Müller.
Hicks, Alexander; Misra, Joya, 1993: “Political Resources and the growth of welfare in
affluent capitalist democracies, 1960 – 1982”, American Journal of Sociology 99: 668-
Hicks, Alexander M.; Swank, Duane H., 1992: “Politics, Institutions, and Welfare Spending
in Industrialized Democracies, 1960-82”, American Political Science Review 86: 658-
Hidien, Jürgen, 1998: Der bundesstaatliche Finanzausgleich in Deutschland. Baden-Baden:
Hockerts, Hans Günter, 1980: Sozialpolitische Entscheidungen im Nachkriegsdeutschland.
Alliierte und deutsche Sozialversicherungspolitik, 1945-1957. Stuttgart: Klett-Cotta.
Hockerts, Hans Günter, 1992: „Vom Nutzen und Nachteil parlamentarischer Parteienkon-
kurrenz. Die Rentenreform 1972 – Ein Lehrstück“, in: Karl D. Bracher (ed.): Staat und
Parteien. Berlin: Duncker & Humblot, 903-934.
Hohorst, Gerd; Kocka, Jürgen; Ritter, Gerhard A., 1975: Sozialgeschichtliches Arbeitsbuch.
Materialien zur Statistik des Kaiserreichs, 1870-1914. München: Beck.
Howard, Christopher, 2002: “Workers’ Compensation, Federalism, and the heavy Hand of
History”, Studies in American Political Development 16: 28-47.
Huber, Evelyne; Stephens, John D., 2001: Development and Crisis of the Welfare State.
Parties and Policies in Global Markets. Chicago: University of Chicago Press.
Huber, Evelyne; Ragin, Charles; Stephens, John D., 1993: “Social Democracy, Christian
Democracy, Constitutional Structure and the Welfare State”, American Journal of Soci-
ology 99: 711-749.
Huber, Ernst Rudolf, 1981: „Die Weimarer Reichsverfassung“, in: Deutsche Verfassungs-
geschichte seit 1789, Bd. 6. Stuttgart: Kohlhammer.
Huber, Ernst Rudolf, 1988: Bismarck und das Reich. Deutsche Verfassungsgeschichte seit
1789, Bd. 3. Stuttgart: Kohlhammer.
ILO, various years: Costs of Social Security. ILO
Inman, Robert P.; Rubinfeld, Daniel L., 1997: “The Political Economy of Federalism”, in:
Daniel Mueller (ed.), Perspectives of Public Choice. New York: Cambridge University
Isensee, Josef, 1990: „Idee und Gestalt des Föderalismus im Grundgesetz“, in: Josef Isen-
see; Paul Kirchhof, Handbuch des Staatsrechts der Bundesrepublik Deutschland, Bd.
IV: Finanzverfassung – Bundesstaatliche Ordnung. Heidelberg: Mueller, 517-691.
Jochem, Sven, 2001: „Reformpolitik im deutschen Sozialversicherungsstaat“, in: Manfred
G. Schmidt (Hg.), Wohlfahrtsstaatliche Politik. Institutionen, politischer Prozeß und
Leistungsprofil. Opladen: Leske + Budrich, 193-226.
Kahlenberg, Friedrich P.; Hoffmann, Dierk, 2001: „Sozialpolitik als Aufgabe zentraler
Verwaltungen in Deutschland – ein verwaltungsgeschichtlicher Überblick 1945-1990“,
in: Bundesministerium für Arbeit und Sozialordnung (Hg.), Geschichte der Sozialpolitik
in Deutschland seit 1945. Baden-Baden: Nomos, 103-182.
Kasza, Gregory J., 1996: “War and comparative politics”, Comparative Politics 28, 355-
Kersbergen, Kees van, 1995: Social Capitalism. London: Routledge.
Kilper, Heiderose; Lhotta, Roland, 1996: Föderalismus in der Bundesrepublik Deutschland.
Opladen: Leske + Budrich.
Kitschelt, Herbert, 2001: “Partisan Competition and Welfare State Retrenchment. When do
Politicians choose unpopular Policies?”, in: Paul Pierson (ed.), The New Politics of the
Welfare State Oxford: Oxford University Press, 265-302.
Kittel, Bernhard; Obinger, Herbert; Wagschal, Uwe, 2000: „Die gezügelten Wohlfahrtsstaa-
ten im internationalen Vergleich: Politisch-institutionelle Faktoren der Entstehung und
Entwicklungsdynamik“, in: Herbert Obinger; Uwe Wagschal (eds.), Der gezügelte
Wohlfahrtsstaat. Frankfurt: Campus, 329-364.
Lane, Jan-Erik; McKay, David; Newton, Kenneth, 1997: Political Data Handbook of the
OECD Countries. Oxford: Oxford University Press.
Lehmbruch, Gerhard, 1998: Parteienwettbewerb im Bundesstaat. Opladen: Westdt. Verlag.
Lehmbruch, 2002: „Der unitarische Bundesstaat in Deutschland: Pfadabhängigkeit und
Wandel“, in: Arthur Benz; Gerhard Lehmbruch (Hg.), Föderalismus. Analysen in ent-
wicklungsgeschichtlicher und vergleichender Perspektive. PVS Sonderheft 32/ 2001.
Opladen: Westdt. Verlag, 53-110.
Leibfried, Stephan; Tennstedt, Florian, 1979: „Sozialpolitik und Berufsverbote im Jahre
1933“, Zeitschrift für Sozialreform, 129-153, 211-238.
Leibfried, Stephan; Obinger, Herbert, 2003: “The State of the Welfare State: German Social
Policy between Macroeconomic Retrenchment and Microeconomic Recalibration”, West
European Politics 26, 199-218.
Lewek, Peter, 1992: Arbeitslosigkeit und Arbeitslosenversicherung in der Weimarer Repu-
blik 1918-1927. Stuttgart: Steiner.
Lijphart, Arend, 1984: Democracies. Patterns of Majoritarian and Consensus Government
in Twenty-One Countries. New Haven: Yale University Press.
Manow, Philip, 1997: The Bismarckian Welfare State and the German Political Economy.
MPIfG Discussion Paper.
Manow, Philip; Seils, Eric, 2000: “The employment crisis of the German welfare state”,
West European Politics 23, 137-160.
Manow, Philip, 2001: Social Protection, Capitalist Production: The Bismarckian Welfare
State and the German Political Economy from the 1880s to the 1990s (PhD, University
of Konstanz, Chapter 2.
Manow, Philip, 2002: „‚The Good, the Bad, and the Ugly’: Esping-Andersens Wohlfahrts-
staatstypologie und die religiösen Wurzeln des westlichen Sozialstaats“, Kölner Zeit-
schrift für Soziologie und Sozialpsychologie 54, 203-225.
Manow, Philip; Plümper, Thomas, 2003: The Opportunity Costs of Welfare Reform: Fiscal
Adjustment in the OECD 1972-195. Max-Planck Institute for the Study of Societies, Co-
Money, Jeanne; Tsebelis, George, 2003: Bicameralism. New York: Cambridge University
Müller, Dirk H., 1996: Arbeiter, Katholizismus, Staat. Der Volksverein für das katholische
Deutschland und die katholischen Arbeiterorganisationen in der Weimarer Republik.
Münch, Ursula, 1997: Sozialpolitik und Föderalismus. Zur Dynamik der Aufgabenvertei-
lung im sozialen Bundesstaat. Opladen: Leske + Budrich.
Münch, Ursula, 1998: „Entflechtungsmöglichkeiten im Bereich der Sozialpolitik. Zur Dis-
kussion um eine Föderalisierung der Sozialversicherung“, in: Ursula Männle (Hg.), Fö-
deralismus zwischen Konsens und Konkurrenz. Baden-Baden: Nomos, 73-91.
Nipperdey, Thomas, 1992: „Machtstaat vor der Demokratie“, in: Deutsche Geschichte
1866-1918, Bd. 2. München: Beck.
Nullmeier, Frank, 1992: „Der Zugriff des Bundes auf die Haushalte der Gemeinden und
Parafisci“, in: Hans-Hermann Hartwich; Göttrick Wewer (Hg.), Regieren in der Bundes-
republik, Bd. IV: Finanz- und wirtschaftspolitische Bestimmungsfaktoren des Regierens
im Bundesstaat – unter besonderer Berücksichtigung des deutschen Vereinigungsprozes-
ses. Opladen: Leske + Budrich, 147-180.
OECD, “Revenue Statistics”, http://www.sourceoecd.org
OECD, 2002: Social Expenditure Database. Paris: OECD.
Oeter, Stefan, 1998: Integration und Subsidiarität im deutschen Bundesstaatsrecht. Tübin-
gen: Mohr Siebeck.
Papier, Hans Jürgen, 1995: „Die Regionalisierung der gesetzlichen Rentenversicherung aus
verfassungsrechtlicher Sicht“, Neue Zeitschrift für Sozialrecht 4, 241-244.
Peterson, Paul; Rom, Mark C., 1990: Welfare Magnets. A new Case for a National Stan-
dard. Washington, D.C.: Brookings.
Pierson, Paul, 1995: “Fragmented Welfare States: Federal Institutions and the Development
of Social Policy”, Governance 8, 449-478.
Pierson, Paul, 1996: “The new politics of the welfare state”, World Politics, 48, 143-179.
Pierson, Paul, 2001: The new politics of the welfare state. Oxford: University Press.
Pitschas, Rainer, 1994: „Regionalisierung der Rentenversicherung aus verwaltungswissen-
schaftlicher Sicht“, Neue Zeitschrift für Sozialrecht 3, 289-296.
Preller, Ludwig, 1978: Sozialpolitik in der Weimarer Republik. Düsseldorf: Athenäum.
Prinz, Michael, 1986: Vom neuen Mittelstand zum Volksgenossen. Die Entwicklung des
sozialen Status der Angestellten von der Weimarer Republik bis zum Ende der NS-Zeit.
Prinz, Michael, 1991: „Die Arbeiterbewegung und das Modell der Angestelltenversiche-
rung. Zu einigen Bedingungen für die besondere Bürgerlichkeit des Wohlfahrtsstaates in
der Bundesrepublik“, in: Klaus Tenfeld (Hg.), Arbeiter im 20. Jahrhundert. Stuttgart:
Prud’homme, Rémy, 1995: “The dangers of decentralization”, The World Bank Research
Observer 10, 201-220.
Ragnitz, Joachim, 2003: Wie hoch sind die Transferleistungen für die neuen Länder? Insti-
tut für Wirtschaftsforschung, Halle.
Renzsch, Wolfgang, 1991: Finanzverfassung und Finanzausgleich. Die Auseinandersetzun-
gen um ihre politische Gestaltung zwischen Währungsreform und deutscher Vereinigung
(1948-1990). Bonn: J.H.W. Dietz.
Ritter, Gerhard A., 1983: Sozialversicherung in Deutschland und England. München: Beck.
Ritter, Gerhard A., 1997: „Sozialpolitik im Zeitalter Bismarcks“, Historische Zeitschrift
Robertson, David Brian, 1989: “The Bias of American Federalism: The Limits of Welfare
State Development in the Progressive Era”, Journal of Policy History 1.
Rodden, Jonathan, 2002: “The Dilemma of Fiscal Federalism: Grants and Fiscal Perform-
ance around the World”, American Journal of Political Science 46, 670-687.
Rodden, Jonathan; Wibbels, Erik, 2002: “Beyond the Fiction of Federalism. Macroecono-
mic Management in Multitiered Systems”, World Politics, 54, 494-531.
Rosewitz, Bernd; Webber, Douglas, 1990: Reformversuche und Reformblockaden im deut-
schen Gesundheitswesen. Frankfurt a.M.: Campus.
Ross, Fiona, 1997: “Cutting Public Expenditures in Advanced Industrial Democracies: The
Importance of Avoiding Blame”, Governance 10:175-200.
Sachße, Christoph; Tennstedt, Florian, 1980: Geschichte der Armenfürsorge in Deutschland
Band 1: Vom Spaetmittelalter bis zum Ersten Weltkrieg. Stuttgart: Kohlhammer.
Sachße, Christoph; Tennstedt, Florian, 1988: Geschichte der Armenfürsorge in Deutschland
Band 2: Fürsorge und Wohlfahrtspflege, 1871-1929. Stuttgart: Kohlhammer.
Sachße, Christoph; Tennstedt, Florian, 1992: Geschichte der Armenfürsorge in Deutschland
Bd. 3: Der Wohlfahrtsstaat im Nationalsozialismus. Stuttgart: Kohlhammer.
Scharpf, Fritz W., 1989: „Der Bundesrat und die Kooperation auf Dritter Ebene“, in: Bun-
desrat (Hg.), Vierzig Jahre Bundesrat. Baden-Baden: Nomos, 121-162.
Scharpf, Fritz W., 1997: “The Joint Decision Trap. Lessons from German Federalism and
European Integration”, in: Klaus E. Goetz (ed.), Germany. Aldershot: Ashgate, 291-330.
Scharpf, Fritz W., 2001: “Employment and the Welfare State - a continental Dilemma”, in:
Bernhard Ebbinghaus; Philip Manow (eds.), Comparing Welfare Capitalism. London:
Scheur, Wolfgang, 1967: Einrichtungen und Maßnahmen der sozialen Sicherheit in der Zeit
des Nationalsozialismus. Dissertation Köln.
Schindler, Peter, 1999: Datenhandbuch zur Geschichte des Deutschen Bundestages. Baden-
Schmähl, Winfried, 2001: “Finanzverflechtung der gesetzlichen Rentenversicherung: inter-
ner Finanzausgleich und Finanzbeziehungen mit dem Bund sowie anderen Sozialversi-
cherungsträgern, in: Klaus-Dirk Henke; Winfried Schmähl (Hg.), Finanzierungsverflech-
tung in der sozialen Sicherung. Baden-Baden: Nomos, 9-37.
Schmid, Josef, 1990: Die CDU, Organisationsstrukturen, Politiken und Funktionsweisen
einer Partei im Föderalismus. Opladen: Leske + Budrich.
Schmid, Josef, 2001: „Sozialpolitik und Wohlfahrtsstaat in Bundesstaaten“, in: Arthur
Benz; Gerhard Lehmbruch (Hg.), Föderalismus. Analysen in entwicklungsgeschichtli-
cher und vergleichender Perspektive. PVS Sonderheft 32/2001. Opladen: Westdeutscher
Schmidt, Manfred G., 2002: „Germany – The Grand Coalition State“, in: Joseph M. Co-
lomer (ed.), Political Institutions in Europe. London: Routledge, 57-94.
Schmidt, Manfred G., 2003: “Rot-Grüne Sozialpolitik (1998-2002)”, in: Christoph Egle et
al. (eds.), Das Rot-Grüne Projekt. Eine Bilanz der Regierung Schröder 1998-2002.
Opladen: Westdeutscher Verlag, 239-258.
Schwinn, Oliver, 1997: Die Finanzierung der deutschen Einheit. Opladen: Leske +
Siegel, Nico, 2002: Baustelle Sozialpolitik. Konsolidierung und Rückbau im internationalen
Vergleich. Frankfurt: Campus.
Skocpol, Theda; Amenta, Edwin, 1986: “States and social policies”, Annual Review of So-
ciology 12, 131-157.
Skocpol, Theda; Amenta, Edwin, 1988: “Redefining the new deal. World War II and the
development of social provision in the United States”, in: Margaret Weir; Ann S. Orloff;
Theda Skocpol (eds.): The politics of social policy in the United States. Princeton: Uni-
versity Press. 81-122.
Skocpol, Theda; Amenta, Edwin, 1989: “Taking exception. Explaining the distinctiveness
of American public policies in the last century”, in: Castles, Francis G. (ed.): The com-
parative history of public policy. Cambridge: Polity Press, 292-333.
Skocpol, Theda; Orloff, Ann Sheila, 1984: “Why no equal Protection? Explaining the Poli-
tics of Public Social Spending in Britain, 1900-1991, and the United States, 1880-1920”,
American Sociological Review 49, 726-750.
Skocpol, Theda, 1992: Protecting Soldiers and Mothers. The political Origins of Social
Policy in the United States. Harvard: Belknap Press.
Statistical Yearbook, Federal Republic of Germany, various years.
Stolleis, Michael, 2001: „Historische Grundlagen. Sozialpolitik in Deutschland bis 1945“,
in: Bundesministerium für Arbeit und Sozialordnung (Hg.), Geschichte der Sozialpolitik
in Deutschland seit 1945. Baden-Baden: Nomos, 199-332.
Sturm, Roland, 1998: „Die Wende im Stolperschritt – eine finanzpolitische Bilanz“, in:
Göttrick Wewer (Hg.), Bilanz der Ära Kohl. Christlich-liberale Politik in Deutschland
1982-1998. Opladen: Leske + Budrich, 183-200.
Swank, Duane H., 2001: “Political Institutions and Welfare State Restructuring: The Impact
of Institutions on Social Policy Change in Developed Democracies”, in: Paul Pierson
(ed.), The new Politics of the Welfare State. Oxford: Oxford University Press, 197-237.
Swank, Duane H., 2002: Global Capital, Political Institutions, and Policy Change in devel-
oped Welfare States. Cambridge: Cambridge University Press.
Teppe, Karl, 1977: „Zur Sozialpolitik des Dritten Reiches am Beispiel der Sozialversiche-
rung“, Archiv für Sozialgeschichte 27, 195-250.
Titmuss, R. 1976: Social Policy. London: Allen & Unwin.
Trampusch, Christine, 2000: Arbeitsmarktpolitik, Gewerkschaften und Arbeitgeber. Ein
Vergleich über die Entstehung und Transformation der öffentlichen Arbeitsverwaltungen
in Deutschland, Großbritannien und den Niederlanden zwischen 1909 und 1999. Disser-
tation, Universität Göttingen.
Trampusch, Christine, 2002: Sozialpolitischer Verschiebebahnhof. MPIfG Discussion Pa-
Trampusch, Christine, 2003: Ein Bündnis für die nachhaltige Finanzierung der Sozialversi-
cherungssysteme. MPIfG Discussion Paper.
Tsebelis, George, 1995: “Decision making in political systems”, British Journal of political
science 25, 289-326.
Tsebelis, George; Money, Jeanette, 1997: Bicameralism. Cambridge: University Press.
Volden, Craig, 2002: “The politics of competitive federalism: A race to the bottom in wel-
fare benefits?” American Journal of Political Science 46, 352-363.
Watts, Ronald L., 1998: “Federalism, Federal Political Systems, and Federations”, Annual
Review of Political Science 1, 117-137.
Watts, Ronald L., 1999: Comparing Federal Systems. Kingston, Ontario.
Webber, Douglas, 1988: „Krankheit, Geld und Politik. Zur Geschichte der Gesundheitsre-
formen in Deutschland“, Leviathan 16, 156-203.
Webber, Douglas, 1989: „Zur Geschichte der Gesundheitsreformen in Deutschland, Teil II:
Norbert Blüms Gesundheitsreformen und die Lobby“, Leviathan 17, 262-300.
Weingast, Barry, 1993: “Constitutions as Governance Structures: The political Foundations
of Secure Markets”, Journal for Institutional and Theoretical Economics 149, 286-311.
Weingast, Barry, 1995: “The economic role of political institutions: Market-preserving fed-
eralism and economic development”, Journal of Law, Economics, and Organization 11,
Wibbels, Erik, 2000: “Federalism and the Politics of Macroeconomic Policy and Perform-
ance”, American Journal of Political Science 44, 687-702.
Wilensky, Harold, 1975: The Welfare State and Equality. Berkeley: University of California
Winter, Thomas von, 1997: Sozialpolitische Interessen: Konstitutierung, politische Reprä-
sentation und Beteiligung an Entscheidungsprozessen. Baden-Baden: Nomos.
Witt, Peter-Christian, 1970: Die Finanzpolitik des Deutschen Reiches von 1903 bis 1913.
Lübeck/ Hamburg: Matthiesen.
Witt, Peter-Christian, 1992: „Finanzen und Politik im Bundesstaat – Deutschland 1871-
1933“, in: Jochen Huhn; Peter-Christian Witt (Hg.), Föderalismus in Deutschland. Ba-
den-Baden: Nomos, 75-99.
Ziblatt, Daniel, 2002: Constructing a Federal State: political development, path dependence
and the origins of federalism in modern Europe, 1815-1871. PhD, University opf Cali-
Zohlnhöfer, Reimut, 2001: Die Wirtschaftspolitik der Ära Kohl. Opladen: Leske + Budrich.