ńņš. 4
(āńåćī 7)



Age in the Welfare State

deputies to extend family allowances to small independent farmers, share-
croppers, and tenant farmers, but this did not occur until 1967. Yet despite
frequent extensions to particular economic groups, the family allowance
system remained far from universal. People with weak ties to the for-
mal labor market were excluded, as were the self-employed, and by 1960
only about half of all children under age eighteen were entitled to family
allowances (Franco 1993).
Why was the family allowance program in Italy not converted into a
universal, citizenship-based beneļ¬t, despite the fact that it was so popular
and seemed to be on its way toward universalization in the 1950s and 1960s?
Three main reasons stand out. First, family allowances were always linked
conceptually to wages and were not seen as an important tool for poverty
alleviation. As a result, there was little call to extend the system to cover
nonworkers, however needy. Second, neither the Christian Democratic
Party (Democrazia Cristiana, or DC) nor the Left was really interested
in extending family allowances to either non-insured workers or the self-
employed. Finally, the tax system in Italy, which did not adequately capture
income from the self-employed, was not capable of supporting a major
extension of the system.
Unlike in the Netherlands, there was no postwar plan to turn the old
occupationally based family allowance scheme into a Beveridgean, universal
program. The Dā€™Aragona commission on social security, the direct analog
to the Netherlandsā€™ van Rhijn Commission, speciļ¬cally recommended a
non-Beveridgean system of family allowances (Commissione per la Riforma
della Previdenza Sociale 1948). The commission was explicit that family
allowances were to be considered a wage supplement, not a social secu-
rity beneļ¬t, and argued for a work-related beneļ¬t available to people
whose main income came from dependent employment (in the form of
either wages or social insurance beneļ¬ts linked to employment). The self-
employed were to be included in the family allowance plan, but only at a
later date, once the administrative and ļ¬scal obstacles to including them
could be overcome.
The DCā€™s lack of interest in universalizing the family allowance system
sprang from the fact that their key constituencies were either already cov-
ered or derived greater beneļ¬ts from not receiving family allowances than
from receiving them. Public employees received an independent family
allowance scheme in 1952, and thus did not need to be integrated into a
universal beneļ¬t system. Agricultural workers and small farmers were incor-
porated into the wage earnersā€™ system in 1967, on quite favorable terms.
Family Allowances

The self-employed constitute the last major constituency group of the
DC that remained without entitlement to family allowances, but they
derived clear beneļ¬ts from remaining outside the system. In the immediate
postwar years, income from self-employment was not subject to taxation,
and as a matter of basic fairness, it was clear that the self-employed should
not be eligible for tax-ļ¬nanced beneļ¬ts. To incorporate the self-employed
into a system of universal family allowances, it would have been neces-
sary ļ¬rst to establish levies on income from self-employment, and then to
enforce and collect these taxes. While income taxes on the self-employed
were eventually legislated, the social insurance system (including fam-
ily allowances) did not take on a universalist form to match. There was
ample awareness among the public and among policy experts that the self-
employed were not subject to effective taxation, and thus should not be
included in a universal family allowance scheme.
The self-employed agreed, because in the end the DC was able to
offer them a better deal even than family allowances. The self-employed
remained outside the family allowance system, but in return received tacit
permission to evade income taxes, as well as other kinds of protections that
boosted their incomes far more than family allowances would have (for
example, stiff barriers to competition from large retailers; Livi Bacci 1998
interview). There was not a greater push from the DC for universaliza-
tion of the family allowance system precisely because the main group of
potential DC constituents that was left out under the occupational system
perceived themselves as better off without family allowances.
But policy making in Italy was not nearly as dominated by the DC as one
might expect given ļ¬fty years of DC government. Especially in the area of
social policy, and especially after the beginning of the center-left coalitions
in 1963, Socialist Party (Partido Socialista Italiano, or PSI) and Communist
Party (Partido Comunista Italiano, or PCI) politicians were closely inte-
grated into the policy-making process. And indeed, after 1963 there was
an expansion of the family allowance system to cover categories tradition-
ally associated with the Left: unemployed workers, domestic employees,
home-workers (mostly female piece-workers), part-time agricultural wage
workers, and, in 1974, old-age and disability pensioners. The Left did see
the expansion of family allowances as beneļ¬cial to their constituencies, and
thus to their partiesā€™ fortunes. But this expansion remained tightly linked
to the wage principle: family allowances were extended only to those who
had a right to a supplement to their wages (or deferred wages, in the case
of the unemployed and pensioners). The self-employed and those outside
Age in the Welfare State

the formal labor market did not beneļ¬t from this extension, in the former
case because of the perceived impossibility of adequately taxing them, and
in the latter case because as long as poverty alleviation was not the guiding
principle behind family allowances, people with weak attachments to the
labor market had correspondingly weak claims to entitlement.
In sum, during the 1950s and early 1960s family allowances were not
universalized, but nevertheless experienced substantial growth. The level
of the beneļ¬t was raised on numerous occasions, and the pool of beneļ¬cia-
ries expanded to cover new constituencies. Indexation between 1955 and
1961 kept the purchasing power of the allowances more or less constant, and
indicates a consensus that the level was about right. So how can we under-
stand what happens next: the quite sudden decline of family allowances
through the de-indexation of beneļ¬ts and the use of funds contributed to
the family allowance scheme for other purposes?

1964ā€“82: Inattention and the ā€œSackingā€ of the Family Allowance Funds
The 1961 reorganization of the family allowance scheme consolidated the
numerous accounts for different sectors into three funds and reduced the
variety of beneļ¬t levels to two. It also marked the beginning of the end for
Italyā€™s family allowances. The indexation of beneļ¬ts to the cost of living,
instituted in the legislation of 1955, ceased in 1961. This change, which
would have enormous implications for the future of family allowance ben-
eļ¬ts, came about with relatively little fanfare. Confederazione Generale
Italiana del Lavoro (CGIL), Confederazione Italiana Sindacati Lavoratori
(CISL), and Conļ¬ndustria documents for the year 1961 contain no men-
tion of the de-indexation. Family allowances had increased along with living
costs before they were indexed in 1955, so unions were probably willing to
concede on the indexation issue, especially since inļ¬‚ation rates in the late
1950s and early 1960s were fairly low. The consolidation of schemes for
different categories of workers, which meant that industrial workers ended
up taking on the large deļ¬cits in the artisansā€™ and agricultural workersā€™
schemes, was a much more salient issue. Certainly this is what Conļ¬ndus-
tria was most concerned with, and what the Minister for Labor and Social
Affairs, in his presentation to the Senate in October of 1961, spent the most
breath trying to justify (Sullo 1961).
But there is another reason why de-indexing might not have seemed like
an important change at the time: the massimale was still in place. With the
massimale in place, employers did not have to pay contributions on their total
Family Allowances

wage bills above a certain ceiling. This meant that smaller employers, and
employers in lower-wage sectors and areas, ended up paying proportionally
more. In addition, since the level of the income ceiling was not indexed
and had not been raised since it was instituted, contributions to the family
allowance fund were not growing with the growth in the total wage bill.
This was seen as putting a constraint on further expansion of the system, as
expressed by Minister Sullo and (repeatedly) by the unions. The hope was
that by getting rid of the massimale, there would be more resources to go
around, and family allowances would be able to grow. The 1961 legislation
agreed to abolish the massimale by 1964, and that was where hopes for
further expansion of family allowances lay.
As it happened, after de-indexation in 1961, the real value of family
allowances dropped sharply, beginning a long slide. There were two reval-
uations in 1964, but that year there was another signiļ¬cant event in the
history of Italian family allowances. With surpluses in the family allowance
fund growing as a result of the declining payouts, the unions and employers
reached an agreement in April of 1964 to allocate the full surplus to fam-
ily allowance increases. However, as the economic situation continued to
worsen, other claims began to intrude on the family allowance fundā€™s nest
egg. Tripartite policy concertation, in place since 1962 and viewed by the
unions as a vital contribution to keeping the new center-left government
alive, resulted in agreements that weakened the ļ¬nancial autonomy of the
family allowance fund and signaled the beginning of a new set of policy
priorities. Under pressure from the government, unions agreed to delay
part of the planned increase in family allowances, diverting the surplus in
the Cassa Unica Assegni Familiari (CUAF) to funds for short-term unem-
ployment beneļ¬ts and the construction of workersā€™ housing. In return, the
government agreed to present a serious plan for pension reform by Decem-
ber of 1964. So despite a small uptick in the real value of the beneļ¬t in 1965,
the agreement of 1964 marked the beginning of what key Italian observers
of family policy would term the ā€œsacriļ¬ceā€ (Franco and Sartor 1994, 17)
or the ā€œsackingā€ (Gorrieri 1979, 129ā€“31) of family allowances in order to
obtain other policy priorities.
Despite the extension of family allowances to new constituencies (short-
term unemployment beneļ¬ciaries in 1964, unemployment insurance bene-
ļ¬ciaries in 1965, the agricultural self-employed in 1967, and domestic and
home workers in 1971 and 1973), aggregate spending and average bene-
ļ¬ts would suffer a real decline from 1964 onward. The only respite from
this decline came in 1974ā€“5 when the massimale was ļ¬nally abolished and
Age in the Welfare State

allowances were revalued to roughly 1966 levels before dropping off even
more steeply again in the following year. Also in 1974 family allowances
were extended to INPS disability and old-age pensioners. Because in Italy
dependent spouses and parents, as well as children, are eligible for family
allowances, many pensioners beneļ¬ted, and this in fact marked a notable
expansion of the program. After the enactment of this provision, about
55 percent of recipients of family allowances were pensioners (Sgritta and
Zanatta 1993). Ironically, the only major beneļ¬t increase of the supposedly
youth-oriented family allowance program after 1961 primarily beneļ¬ted
After the crucial decisions ļ¬rst to de-index beneļ¬ts, and then to trade
family allowance increases for progress on the pension front, family
allowance spending underwent a precipitous decline in Italy. At a techni-
cal level, this development is explained by high inļ¬‚ation, which eroded the
value of allowances, and the substitution of other policy priorities for ad hoc
increases that might have maintained the value of the family allowance ben-
eļ¬t. Explaining the politics of this sequence of events is somewhat more
Because of their link to the wage system, family allowances had simply
ceased to be on the top of anyoneā€™s list of social policy priorities. Union
documents and ofļ¬cials testify that the important ļ¬ghts on behalf of work-
ing families in the 1960s had to do with creating jobs and raising wages,
not increasing family allowances (Roscani 1998 interview; Giovannini 1999
interview; Giustina 1999 interview). And as long as wages continued to rise,
family allowances became less and less important relative to either wages or
major social beneļ¬ts such as pensions that were, unlike family allowances,
meant to provide a full replacement for wages. Similarly, although fam-
ily allowances might have been useful to employers because they would
have helped to reduce pressure on wages, employers were loath to offer to
pay more for family allowances in a period when demands for other social
contributions (pensions, unemployment, sickness, etc.) were on the rise
(Mariani 1999 interview). One could say that by the time anyone noticed
that family allowances were being devalued out of existence, it was too
late: at that point they hardly seemed worth ļ¬ghting for anymore. Here
again, pensioners are the exception that proves the rule. Pensioners were
not only organized, vocal, and important constituents within unions and
political parties; their incomes on average were also low enough that family
allowances really did make a difference for them, and they exerted substan-
tial pressure to get them in 1974ā€“5.
Family Allowances

Yet even as family allowance beneļ¬ts declined in value, the system was
expanded to include more and more people. A close examination of which
new groups received beneļ¬ts and how these beneļ¬ts were administered in
the period after devaluation suggests that family allowances were still useful
to the two main parties in ofļ¬ce, the Christian Democrats and the Social-
ists. Small farmers were the single most inļ¬‚uential lobby group within
the DC in the 1960s (LaPalombara 1964), and they were awarded fam-
ily allowance beneļ¬ts in 1967. The PSI was also able to please important
constituencies with the extension of family allowance beneļ¬ts to part-time
agricultural workers, domestic workers, and social insurance beneļ¬ciaries.
(Both the DC and the PSI probably perceived it to be in their interest
to award family allowances to pensioners, a fast-growing and increasingly
vocal constituency in the early 1970s.) Beneļ¬ts for these new groups were
administered directly by the state, rather than through employers. Politi-
cians in the ruling coalition thus gained doubly by the expansion of the
family allowance system, despite low beneļ¬t levels. The extension of ben-
eļ¬ts was targeted at their constituencies, and they had direct control over
the disbursement of allowances.
Recalling the interests of clientelist politicians in the ruling coalition
helps to make sense of the fact that family allowances continued to be
extended while their value declined. What mattered to parties distributing
patronage was not so much how big the beneļ¬t was, but how much con-
trol they had over it. Many smaller beneļ¬ts were more useful for securing
votes than fewer, higher ones. At the same time, unions and the politi-
cal Left became increasingly disinterested in defending the level of family
allowances as their value fell and they ceased to be an important part of

Post-1983: Structural Reform, or Old Wine in New Bottles?
By the end of the 1970s, observers frequently bemoaned the inadequacy
of family allowances in Italy. Two minor revaluations of family allowances
occurred in 1980, at least partially in response to the debate sparked by
Gorrieriā€™s inļ¬‚uential 1979 publication The Family Budget Jungle (La giungla
dei bilanci familiari). This work launched a scathing critique of politicians
and bureaucrats who had, in Gorrieriā€™s widely shared view, misused the
funds collected for family allowances to ā€œļ¬nance a whole host of clientelist
policies aimed at those who are not really in need of assistanceā€ (Gorrieri
1979, 131). By the early 1980s, there was general agreement within the
Age in the Welfare State

policy-making community on the need to reform the family allowance sys-
tem more fundamentally.
One reason for the new attention to family allowances was a newfound
concern among policy makers and politicians with poverty among families
with children. The postwar Poverty Commission (1951) and Commission
on Unemployment (1953) had identiļ¬ed families with working-age heads,
especially large families, as prime candidates for poverty when adequate
labor market remuneration was not forthcoming. But they had both seen
boosting employment, especially in the south, as the means to alleviate
this poverty. Problem cases that would not disappear as a result of more
and better jobs were expected to be few, and could be taken care of by
regional- and municipal-level charity programs (Camera dei Deputati 1953;
Commissione Parlamentare dā€™Inchiesta sulla Disoccupazione 1953).
The economic boom of the 1950s and 1960s indeed did dramatically
raise the standard of living of most working-age families, to the point where
poverty among the elderly became the almost exclusive concern of policy
makers, politicians, and union specialists interested in poverty issues. But
when in the 1970s a ā€œrediscoveryā€ of poverty occurred in Italy, as elsewhere
in Europe (see Gauthier 1993), in Italy the policy response was limited to
facilitating local-level efforts to combat social exclusion and poverty among
the hard-core poor such as drug addicts, the handicapped, or the homeless.
The turning point came in 1982, with the publication of a European Unionā€“
sponsored report on poverty directed by Giovanni Sarpellon. Sarpellonā€™s
report echoed the results of the early postwar studies in arguing that unem-
ployment and bad employment were among the most important causes of
poverty in Italy. Sarpellon called attention in particular to the problems of
large families living in the south, arguing that poverty among the elderly,
while it was indeed the most prevalent form of poverty in the north, was
not Italyā€™s biggest problem (Sarpellon 1983).
Sarpellonā€™s report came out in the context of renewed interest in Catholic
circles in the issue of the family wage7 and in the midst of the work of the
government Commission on the Problems of Families (which was commis-
sioned in 1980 and presented its ļ¬nal report in 1983). This commission was
charged by the government with setting a coherent policy agenda for help-
ing families, in particular the families of workers. Its ļ¬nal report advocated

7 The 1981 papal encyclical Laborem excercens and the DC proposal for a large family allowance
increase for nonworking mothers ā€“ a proposal that the PCI attacked as a cynical electoral
ploy since it could not possibly be funded ā€“ were also presented at around this time.

Family Allowances

focusing resources on poor families, given what the commission members
perceived as inherent limits to the resources that could be made available
for helping families. They suggested a Social Allowance (assegno sociale), to
be given to poor families of dependent employees in order to bring them
out of poverty. While the eventual goal of extending this beneļ¬t beyond
employees was deemed desirable, it was argued that the ļ¬scal systemā€™s lim-
ited capacity to collect taxes on income from self-employment made the
prospect of any universal expansion difļ¬cult. The commission also sug-
gested merging the social allowance for families with children into a more
general allowance providing a basic income for the elderly poor and inva-
lidity pensioners (Commissione Nazionale per i Problemi della Famiglia
1983). However, the commission rejected acting immediately on this last
proposal, because they were concerned that its greater cost would result
in money being channeled away from families with children (Franco and
Sartor 1990).
In 1983, the same year as the report of the Commission on the Problems
of Families, the family allowance system underwent a seemingly minor
change that was the beginning of a period of substantial reform. Union-
employer-government concertation in 1983 resulted in the Scotti Accord,
a set of provisions designed to revive the ļ¬‚agging Italian economy. Under
this agreement, the indexation of wages to inļ¬‚ation was reduced, and in
compensation a new supplementary family allowance, paid out of general
revenues, was adopted. However, this supplementary allowance would be
available only to some families, based on their income and family size. For
the ļ¬rst time, a means test was introduced into the Italian family allowance
system, and the goal of poverty alleviation through vertical redistribution
came to the forefront after decades during which the principle that family
allowances should mainly aim at horizontal redistribution among wage-
earners had dominated.
In 1984, cost-containment pressures led to the means-testing of regu-
lar family allowances. Meanwhile, the governmentā€™s Poverty Commission,
instituted in 1984, was preparing its report, which came out in 1985 (Com-
missione di Indagine sulla PovertĀ“ 1985). The Poverty Commissionā€™s rec-
ommendations were similar to those of the Commission on the Problem of
Families, in that they supported the idea of a means-tested beneļ¬t to poor
families. However, the poverty commission favored extending the beneļ¬t
to the poor elderly as well as to families with children. Responding to these
recommendations, in 1988 Parliament passed a government proposal to
replace the family allowance system with a new program, the Allowance
Age in the Welfare State

for the Family Unit (Assegno per il Nucleo Familiare, or ANF). It would,
like family allowances after 1984, be means-tested, and beneļ¬ts would be
graduated by family size and income.
The concept of a ā€œdependentā€ was abolished in favor of the concept of
family or household ā€œneed.ā€ Beneļ¬ts would be granted no longer on the
basis of a personā€™s status as a provider for dependent children or adults but
on the basis of the overall level of need in the household. Allowances for
children were to be given to parents, but allowances for dependent parents
went directly to the elderly themselves. And single elderly persons became
eligible for the ANF.
But despite the fact that family allowances had been transformed from
an insurance measure into a poverty-alleviation measure for individuals of
all age groups, they were still ļ¬nanced primarily through employer contri-
butions. And despite the fact that the principle of need had taken over for
the essentially insurance-based concept of the old family allowances, non-
elderly families were eligible for beneļ¬ts only if they were headed by employees,
no matter how needy they might be. As a result, under the ANF, occupa-
tionally based contributions paid for citizenship-based assistance beneļ¬ts
for the elderly, but the workers ļ¬nancing the system could receive the
beneļ¬t themselves only if they passed a means test. And poor non-elderly
families had no entitlement to beneļ¬ts unless they were headed by workers
in the formal labor market.
This rather bizarre situation came about because of union concerns about
the capacity of a tax-ļ¬nanced system to support adequate beneļ¬t levels.
Unions withheld their objection to ļ¬nancing the ANF through payroll
contributions because they were aware that switching to a system ļ¬nanced
out of general revenues would imply a universalization of beneļ¬ts. And if
entitlement became universal, union leaders feared, the system would prob-
ably become too expensive to maintain beneļ¬ts at current levels. Shifting
away from payroll ļ¬nancing might in fact result in beneļ¬ts dwindling even
further. Union leaders calculated that if they could maintain a contribution-
ļ¬nanced system, at least decent beneļ¬ts would be assured for the poorest
wage workers and pensioners (Saraceno 1999 interview).
What does this episode of reform tell us about the trajectory of ben-
eļ¬ts for young people in Italy? After the early 1980s key political actors
recognized that families with children were not doing well and agreed
that something needed to be done to combat poverty among this group.
This perception continued and strengthened through the early and mid-
1990s. But at the same time, policy makers believed that costs needed to
Family Allowances

be contained and that there was no room for a real expansion of family
allowances (despite the fact that the family allowance fund was in surplus,
paying out less than half of what it took in in contributions in the period
1980ā€“90). The reforms of the mid-1980s responded to concerns that the old
system of untargeted beneļ¬ts had not worked well, and directed resources
more effectively to needy families, many of which had children. On the
other hand, these reforms reļ¬‚ected a continuing reluctance to spend even
the money that was actually being collected for family allowances, let alone
to take funding away from other priorities. Quite the opposite, in fact: the
1995 pension reform made ofļ¬cial the decades-old practice of transferring
funds from the family allowances account within INPS to cover part of
the deļ¬cits in the pension system. The 1980s saw more attention paid to
poor young families, but as was the case for the 1974ā€“5 increase in family
allowance levels, the family allowance system could be improved only on
the condition that more resources also be directed to the elderly poor.
Why do family allowances in Italy follow a trajectory of declining beneļ¬t
levels and patchwork expansion of family allowances to new clienteles? In
the immediate postwar period, policy makers were concerned about the
well-being of working families, and no oneā€™s interests were hurt by family
allowances. Expansion made good political and economic sense. But family
allowances were never seen as the main way of ensuring the ļ¬nancial stability
of young families: that task was up to the labor market. Furthermore, the
reliance of the Christian Democratic Party on spotty taxation of the self-
employed made it undesirable for any party to push for a universal family
allowance system. The de-indexing of family allowances and prioritization
of other policy goals (namely, pension reform) in the mid-1960s set the stage
for a rapid devaluation of family allowances. This period was marked by
increasing electoral competition between Christian Democracy and parties
of the Left, but all parties in government could gain electoral support simply
by extending the beneļ¬t to new social groups, without regard to the beneļ¬tā€™s
declining real value. Before anyone noticed it, family allowances were so
small that they had ceased to be worth ļ¬ghting for, and the transfer of family
allowance funds to other policy priorities, backed by a strong elderly lobby,
met with little resistance.

This chapter has examined one program, family allowances, that constitutes
an essential component of the overall age orientation of social policies in
Age in the Welfare State

Italy and the Netherlands. The interests of groups commonly believed to
dominate welfare policy making ā€“ labor unions, employers, the Catholic
Church ā€“ do not appear to determine the fate of family allowances in the
postā€“World War II period in either Italy or the Netherlands. Political pres-
sure from age-based interest groups ā€“ the family lobby in the Netherlands
and the elderly lobby in Italy ā€“ does inļ¬‚uence the development of spending
on family allowances, but rarely decisively. The key political actors whose
behavior determines spending patterns on family allowances in Italy and the
Netherlands are politicians working within political parties. In some cases
the demands of age-based lobbies do overlap with the strategic choices of
politicians, and in these situations it is relatively straightforward to inter-
pret policy outcomes as a result of the pressure of electorally important
constituencies for speciļ¬c policy goals. In other cases, however, the vote-
seeking behavior of politicians affects family policy outcomes in a less direct
way. Politiciansā€™ choices in other policy arenas (e.g., taxation, wage pol-
icy, pension policy) shape the potential constituencies of family allowances
and the growth potential of family allowance policies, which in turn affect
how attractive family allowances are as political currency for vote-seeking
If the behavior of politicians helps to create the institutional framework
within which family allowances develop, it is also true that this framework,
once set in place, exercises considerable inļ¬‚uence on the behavior of politi-
cians. In particular, the early choice of either universalist or occupationally
based family allowance programs sets the boundaries within which fam-
ily allowances may grow. Universalist family allowance schemes, whether
means-tested or truly universal, result in welfare regimes that are rela-
tively generous to young people. Universalism in welfare programs has
been conceptualized as an indicator of the degree of cross-class solidarity
achieved in welfare states (Baldwin 1990; Esping-Andersen 1990; Ferrera
1993). But if solidaristic principles underlie universalism, they do not also
generate the age proļ¬le of the social policies that result from universalism.
Rather, universalism and occupationalism in welfare programs condition
how politicians may use these programs as tender in bargaining for votes,
which is in turn responsible for their expansion or contraction.
The breakdown of pillarization in the Netherlands, with the concomi-
tant increase in electoral competition, coincided with a major expansion
of family allowance beneļ¬ts. Likewise, in Italy, there was an expansion of
beneļ¬ts to cover key constituents of the DC and the PSI in the 1960s and
1970s, accompanied by greater government discretion in the disbursement
Family Allowances

of allowances. Both of these indicate a use of family allowance beneļ¬ts
for particularistic purposes, notwithstanding that the target groups in the
Dutch case are far more encompassing than in the Italian case. But how
politicians use programs to buy votes, and how this affects spending on
family allowances, is altered by the structure (universalist or occupational)
of the programs.
The Dutch family allowance programā€™s character as a universal bene-
ļ¬t after 1962 divorced it rhetorically from its roots as a wage supplement
for workers with large families. Thus, rising wages and standards of living
did not, as in Italy, make the allowances less important. Rather, as family
allowances became an entitlement of parenthood, using them as the cur-
rency of electoral exchange became more costly. The only way to expand the
program was to increase per child spending, and ļ¬nally the family allowance,
like other Dutch citizenship-based entitlements, was linked to the indexed
minimum wage ā€“ even though family allowances by this time had lost their
role as a wage supplement.
In Italy, by contrast, a major expansion of the number of family allowance
beneļ¬ciaries coincided with a substantial drop in spending (both aggregate
and per child) and a dramatic decline in the real value of the allowances.
This decline in beneļ¬ts for salaried workers occurred despite the fact that
this group was the key beneļ¬ciary of the family allowance scheme as it was
laid out in the preā€“World War II period, and perhaps because it was a source
of the Left oppositionā€™s staunchest supporters. The erosion of beneļ¬ts for
such a large group of voters was politically tenable, though, because just as
inļ¬‚ation began seriously to impinge on the value of the un-indexed family
allowances, wages were indexed to inļ¬‚ation. The smaller family allowances
became as a share of take-home pay, the less interested unions were in
making them a policy priority, as long as the ā€œearnedā€ component of wages
was rising steadily and eventually became indexed to inļ¬‚ation. And for those
groups that were not included in the original Fascist-era family allowance
system or in the postwar Christian Democratic expansion to small farmers ā€“
groups such as pensioners, the unemployed, domestics, and home workers ā€“
even small family allowances were a nice bonus and sometimes a signiļ¬cant
component of incomes that were largely unprotected from inļ¬‚ation.
The distinction between universalist and occupational program struc-
tures makes a difference for family allowance spending, then, for two main
reasons. First, it is more expensive to expand a population-wide beneļ¬t
than a piecemeal one. Second, objections to devaluation of the beneļ¬t are
weaker when family allowances are viewed as a component of the wage
Age in the Welfare State

rather than an independent entitlement. So then the important question is
why family allowances were universalized in the Netherlands in 1962, but
not in Italy. These program structures themselves are a result of politiciansā€™
choices about how to get the most out of public policies, in particular, tax
and spending policies.
Extending family allowances to the self-employed, which in the Dutch
case was accomplished by universalizing the family allowance system, was
desirable because it granted cash beneļ¬ts to a segment of the electorate that
was crucial to the fortunes of both Protestant and Catholic parties during
a period of electoral realignment. Universalizing the beneļ¬t was possible
because the comprehensive Dutch tax system allowed beneļ¬ts for the self-
employed to be paid for out of tax revenues, rather than requiring employer
and employee contributions to ļ¬nance beneļ¬ts for the self-employed as
In Italy the electoral situation was quite similar, with intense intra-DC
competition for the rural self-employed electorate. But in contrast to the
Dutch situation, Italyā€™s famously unenforceable and unenforced tax system,
a system whose only reliable source of revenue was automatic deductions
from the pay of wage earners and salaried employees of large companies,
did not permit universalization of the family allowance system. To make
extending family allowances to the self-employed politically acceptable,
the tax system would have had to be reformed to include small-business
owners and the self-employed, as virtually every commentator on the social
security system since the close of the war argued. While the DC had been
able to ā€œsellā€ the unfunded expansion of the family allowances system to
small farmers as a necessary solidaristic gesture on behalf of industrial work-
ers (many of whom had relatives living in the south), covering the better-
off self-employed without subjecting them to contributions would have
stretched that solidarity to the breaking point. But the inadequate taxation
of the self-employed in Italy was in itself a valuable political resource for
the DC. The family allowance system was never fully universalized because
both the DC and the self-employed found it advantageous to forgo fam-
ily allowances in order to maintain tax privileges. Of course, this did not
preclude the piecemeal extension of inexpensive family allowances to less
privileged constituencies who could be bought off for less.
In the end, it was politiciansā€™ strategic choices with regard to the tax
system (non-enforcement) and the wage system (indexation of wages but
not allowances) that stymied the growth of family allowances in Italy. In the
Netherlands, the opposite choices made for universalization and expansion
Family Allowances

of beneļ¬ts, such that by 1990 the Netherlands had the most generous cash
beneļ¬ts for families in the OECD. This focus on the strategic behavior of
politicians in response to demand from a variety of societal groups (includ-
ing labor unions, employers, family organizations, and pensionersā€™ groups)
usefully directs attention toward an under-emphasized phenomenon linked
to the development of welfare states. It casts politicians as crucial actors in
the design and magnitude of social programs, through vote-seeking behav-
ior that inļ¬‚uences both spending and features of the institutional landscape
that impact social policy outcomes more indirectly.


Beneļ¬ts for the Unemployed

This chapter addresses two questions: how and why do unemployment ben-
eļ¬ts in Italy treat older workers more generously than younger workers,
while in the Netherlands the opposite is true? The answer to the how
question is straightforward. The youth orientation of the Dutch system
of unemployment protection is due largely to two policy features: gen-
erous regular unemployment insurance beneļ¬ts and universal coverage
that protects ļ¬rst-time job seekers. Italyā€™s highly elderly-oriented system
of unemployment protection, on the other hand, is characterized by mea-
ger regular unemployment insurance beneļ¬ts and no beneļ¬ts for youth
The answer to the why question is more complicated and more interest-
ing. Italy and the Netherlands share features that we might expect would
lead them to develop similar kinds of unemployment policies. Both display
features of what have variously been labeled Conservative-Corporatist or
Christian Democratic systems, combined with a distinct leftist presence in
policy making. The result is a male-breadwinnerā€“centered labor market,
few active labor market policies, and low female labor force participation.
The two countries also share a history of high unemployment, numerically
rather weak labor unions, and high levels of self-employment. Yet Esping-
Andersenā€™s (1990) characterization of the Netherlands as an almost Social
Democratic welfare state, a characterization driven largely by the generos-
ity of Dutch unemployment beneļ¬ts, highlights an important difference.
Italy and the Netherlands differ dramatically in the extent to which their
welfare states protect elderly versus non-elderly citizens from a variety of
risks, and nowhere is this difference as pronounced as in their unemploy-
ment policy regimes.

Beneļ¬ts for the Unemployed

The elderly orientation of the Italian system of unemployment protec-
tion, and the relative youth orientation of the Dutch system, are a result of
interactions between the structure of labor market policies that politicians
and policy makers have inherited from the past and the characteristics of
the competitive environments in which they ļ¬nd themselves. The quite
different age orientations of the unemployment protection systems in the
Netherlands and Italy are difļ¬cult to explain if we consider only the ide-
ologies or power resources of political actors, but not once we include the
institutional and political environment within which they act.
Standard explanations for the extreme generosity of Dutch unemploy-
ment insurance beneļ¬ts, for example, attribute high replacement rates to
either Christian Democratic dominance of policy making and the resulting
male-breadwinner focus or, more commonly, to the societal consensus on
Social Democratic ideals of universal coverage and generous beneļ¬ts. Yet
a closer look at the evolution of the Dutch system reveals that the pol-
icy choices that led to a relatively youth-oriented system of unemployment
protection are more strongly inļ¬‚uenced by the legacies of preā€“World War I
experiments with unemployment insurance and the electoral pressures asso-
ciated with the breakdown of pillarization in the 1960s. Politicians eager
to claim credit for a generous system of regular unemployment beneļ¬ts
maintained high replacement rates that dated back to 1916, and electoral
pressures contributed to the creation of universalistic unemployment ben-
eļ¬ts in the mid-1960s. Once this system was in place, it generated across-
the-board entitlements that grew with the pace of the rest of the welfare
state until the mid-1980s. And, as we see in more detail in the following
section, the combination of high replacement rates for regular unemploy-
ment insurance and universal coverage are the most important contributors
to the Netherlandsā€™ youth-oriented system of unemployment protection.
In Italy, as in the Netherlands, policy legacies and the dynamics of parti-
san competition emerge as crucial determinants of the age orientation of the
system of unemployment protection. The low level of regular unemploy-
ment beneļ¬ts and the lack of coverage for youth unemployed in Italy are
most often attributed to the impotence of unions and the Left in the face of
employer preferences for state-subsidized short-term income replacement
schemes over contribution-ļ¬nanced regular unemployment beneļ¬ts. As in
the Dutch case, however, this power resources argument begins to break
down under closer scrutiny. In Italy, legacies of clientelist political competi-
tion, including a weakly enforced tax system, precluded setting replacement
rates for regular unemployment insurance at a high level and scuttled plans
Age in the Welfare State

for universal unemployment beneļ¬ts. Crucially, unions and the Commu-
nist Left altered their preferences to account for the clientelist behavior of
Christian Democratic and eventually Socialist politicians. The Left consis-
tently concurred with policy choices that led to an elderly-oriented system
of unemployment protection, often despite their stated goals of improving
conditions for younger citizens. As in the Dutch case, policy legacies and
the environment of political competition are important determinants of the
age orientation of the Italian unemployment protection system.
This chapter begins with a comparison of unemployment beneļ¬ts
regimes in Italy and the Netherlands in the 1990s. The comparison goes
beyond legislated replacement rates and aggregate spending to examine the
consequences for different age groups of national ensembles of different
types of unemployment-related programs. The second part of the chapter
traces the development of the differing age proļ¬les of unemployment policy
regimes in Italy and the Netherlands, which are not a straightforward result
of political pressure from groups of potential beneļ¬ciaries, employers, or
trade unions. Rather, they are a long-term effect of program structures
that are driven by politiciansā€™ behavior during periods of intense electoral

Comparing Unemployment Beneļ¬ts in Italy and the Netherlands
A comprehensive comparison of the age orientation of unemployment-
related policies should begin with a global comparison of coverage and
beneļ¬t levels, to determine how well an average unemployed person is
protected against income loss due to joblessness. But we also want to know
how younger versus older members of the work force fare within the full
range of programs offering income maintenance for unemployed people.
This section reviews the various kinds of unemployment beneļ¬ts available
in Italy and the Netherlands and the populations covered by each scheme.
Next follows a closer examination of the beneļ¬ciaries of different kinds of
unemployment beneļ¬ts to see how older and younger workers fare under
these programs.

Income Protection Programs for the Unemployed
Since 1916 in the Netherlands and 1919 in Italy, there has existed some
form of state-sanctioned system for protecting workers from loss of income
due to joblessness. Throughout the postā€“World War II period, moreover,
Beneļ¬ts for the Unemployed

a variety of programs supplementing traditional unemployment insurance
have developed in both countries. All of the unemployment beneļ¬t pro-
grams outlined below have undergone signiļ¬cant changes in beneļ¬t levels,
eligibility criteria, duration of beneļ¬ts, and so on since their inception. The
most important changes took place during two periods: from immediately
after the war through the mid-1960s, when prewar systems were adapted
to postwar conditions; and from the mid-1980s onward, as welfare reform-
ers sought to adapt the postwar systems to postā€“oil shock realities. But the
elderly orientation of the Italian system and the youth orientation of the
Dutch system persisted throughout the postwar period. We can summarize
the different types of income-maintenance beneļ¬ts available to the unem-
ployed in the ā€œmatureā€ Italian and Dutch welfare states, noting that while
the systems have undergone some structural changes, especially since the
mid-1980s, their relative age orientations remain distinct.

Regular Unemployment Insurance The ļ¬rst national-level unemploy-
ment insurance schemes were introduced in 1916 in the Netherlands and
in 1919 in Italy. The Dutch program was a voluntary scheme, ļ¬nanced
50 percent by the state, and paid out beneļ¬ts equaling 70 percent of the
prior wage for up to ten weeks. The Italian unemployment insurance law,
Italyā€™s ļ¬rst obligatory social insurance program of any kind, was ļ¬nanced by
employer and employee contributions, with the state stepping in to cover
deļ¬cits when necessary. The beneļ¬t was guaranteed for up to twenty weeks
but, on the model that had just been adopted in Britain, offered only a
ā€œsubsistenceā€ beneļ¬t to recipients. This ļ¬‚at-rate beneļ¬t had three grada-
tions according to the contribution level, but could in no case exceed 50 per-
cent of the previous wage. For average-income earners, this amounted to a
replacement rate of 36 percent of the average daily wage in industry at that
time ā€“ not dramatically meaner than the unemployment provisions extant
in other advanced countries at that time, but still less generous than in the
Netherlands (ILO 1922).
This original divergence in replacement rates continued in the regula-
tion of unemployment insurance beneļ¬ts after World War II in Italy and
the Netherlands. The Dutch introduced compulsory unemployment insur-
ance for all employees (including agricultural employees) in 1949 with the
Unemployment Insurance Act (Werkloosheidswet, or WW). This insur-
ance provided for a (taxable) beneļ¬t equal to 80 percent of the previous
salary (for heads of household), ļ¬nanced jointly by employers, employees,
and the state, and lasting for a maximum of twenty-six weeks. (Beneļ¬ts were
Age in the Welfare State

lower for nonhousehold heads: 70 percent for singles living on their own,
and 60 percent for singles living with their parents.) In 1964, the Dutch
government passed a supplementary unemployment insurance act (Wijzig-
ing Wet Werkloosheidsvoorziening, or WWV) that provided beneļ¬ts of
75 percent of the previous wage for people unemployed up to two years
beyond the original six months covered under the WW. In 1987, as part of
a major overhaul of the Dutch system of labor-marketā€“related regulation,
the beneļ¬t level was reduced to 70 percent of previous salary, still above
the average replacement rate among European countries.
Italyā€™s postwar unemployment insurance law, passed in 1949, maintained
the subsistence beneļ¬t format of the old unemployment insurance system.
But large increases in the cost of living during and immediately after the
war meant that the new daily beneļ¬t (now a ļ¬‚at-rate 200 lire/day) was
nowhere near subsistence level. When the law was introduced, this beneļ¬t
provided for a replacement rate of only 17 percent of an average production
workerā€™s daily wage. The lack of indexation of the beneļ¬t to either wages
or prices meant that the purchasing power of the beneļ¬t continued to fall,
despite sporadic upward adjustments in its nominal value during the late
1950s to mid-1970s. By 1974, the regular unemployment insurance beneļ¬t
had fallen to 800 lire/day, just 8.6 percent of the gross wage of an average
production worker in 1974. It remained at this level until 1988 (in 1987
Italyā€™s Constitutional Court ruled that 800 lire per day was a constitutionally
inadequate beneļ¬t), when the regular unemployment insurance payment
was raised to 20 percent of prior earnings. By 1999, the beneļ¬t had again
been revised upward to 40 percent of prior earnings, a substantial increase,
to be sure, but still much lower than in the Netherlands.
Comparing beneļ¬t replacement rates across countries can lead to inac-
curate judgments about the relative generosity of different programs net of
taxes and other beneļ¬ts (Fawcett and Papadopoulos 1997). Unemployment
insurance beneļ¬ts are taxed as income in some countries and not in oth-
ers, and beneļ¬ts normally associated with income from work (e.g., family
allowances, housing beneļ¬ts, or credit for future pensions) also accrue to
unemployment insurance recipients in some cases. It is useful, then, to use
household income data to estimate post-tax, post-transfer beneļ¬t levels for
unemployment insurance recipients. It is reassuring to note that even after
taking into account taxation of unemployment beneļ¬ts and the addition of
family and housing beneļ¬ts, the Dutch beneļ¬t remains among the most
generous in the OECD, and the Italian beneļ¬t among the least generous,
for a variety of model family types (Table 5.1).
Beneļ¬ts for the Unemployed

Table 5.1 Net replacement rates of unemployment insurance beneļ¬ts, 1997

Couple with
Single two children
Ireland 31 Greece 44
Australia 33 United Kingdom 49
New Zealand 39 Italy 53
Italy 42 Ireland 57
United Kingdom 46 United States 57
Greece 47 Australia 62
United States 58 Belgium 64
Germany 60 Japan 64
Austria 60 New Zealand 68
Canada 62 Germany 70
Denmark 63 France 72
Belgium 64 Denmark 73
Finland 65 Spain 73
Norway 66 Norway 74
Japan 67 Austria 76
France 71 Sweden 78
Sweden 71 Portugal 79
Spain 74 Finland 83
Portugal 79 Luxembourg 87
Luxembourg 82 Netherlands 89
Netherlands 82 Canada 91
Note: Data reļ¬‚ect replacement rates of wages at average production worker earnings level
after taxes and including housing and family beneļ¬ts.
Source: OECD 2002.

The generosity of regular unemployment insurance beneļ¬ts clearly dif-
fers markedly between Italy and the Netherlands. But regular unemploy-
ment insurance beneļ¬ts in both countries are supplemented by a wide vari-
ety of other programs for unemployed workers in different sectors of the
economy and with different personal characteristics. So to determine both
the generosity of unemployment beneļ¬ts in general and the speciļ¬c orien-
tations of these policies toward different age groups, it is necessary to look
beyond the basic insurance program.

Beneļ¬ts for Partial Unemployment The Netherlands provides insur-
ance for partial loss of income due to shortened work hours or temporary
layoffs, with the same beneļ¬ts as for full unemployment. This beneļ¬t
may also be used to cover temporary unemployment due to bad weather.
Age in the Welfare State

In 1979 about 24 percent of unemployment insurance payouts were for
partial or short-term unemployment (CBS 1981, 360). In Italy, insurance
for partial unemployment is in fact an important alternative to the ordi-
nary unemployment insurance beneļ¬t and takes several forms: the short-
time earnings replacement fund (Cassa per lā€™Integrazione Guadagni,
or CIG), the ā€œextraordinaryā€ version of the same program (Cassa per
lā€™Integrazione Guadagni Straordinaria, or CIGS), and so-called mobility
allowances (indennitĀ“ di mobilitĀ“ ).
a a
The earnings replacement fund was established in 1945 to compensate
industrial workers for earnings lost due to reductions in working hours.
With the introduction of bans on mass layoffs after the end of the war, CIG
became an important tool used by employers to manage excess productive
capacity without dismissing workers outright. Throughout the 1950s and
early 1960s, CIG was funded by employer contributions and was used in
place of dismissals during brief periods of slack.
In 1968, however, a new, ā€œextraordinaryā€ form of the beneļ¬t (CIGS),
funded out of general tax revenues rather than employer contributions,
was introduced. During the 1970s and 1980s, CIGS became both employ-
ersā€™ and unionsā€™ tool of choice for compensating workers during partial or
full unemployment resulting from industrial restructuring, sectoral crises,
cyclical downturns, and plant closures.1 From 1975 onward, both CIG and
CIGS paid out a tax-free beneļ¬t equal to 80 percent of prior gross wages ā€“
clearly a much more comprehensive form of income replacement than the
ordinary unemployment insurance program.
Mobility allowances were introduced in 1991 to provide an alternative
to the use of CIGS, which was coming under increasing ļ¬re. Mobility
beneļ¬ts are paid to workers laid off by ļ¬rms in the same sectors covered
under CIGS, who have been employed in the relevant ļ¬rm for at least
twelve months, and who are available for work in a ā€œsuitable job.ā€ The
ā€œsuitable jobā€ criteria are more generous than for regular unemployment
beneļ¬ts, however, and mobility beneļ¬ciaries receive special priority for job
placement. The amount of the beneļ¬t is similar to CIGS. The duration
of the beneļ¬t ranges from twelve to forty-eight months, depending on the
unemployed personā€™s age and the region of residence, but is longest for
older workers.

1 Although nominally CIGS was for temporary unemployment, in the late 1970s it became
possible for redundant employees to receive CIGS beneļ¬ts while working for a ļ¬rm that
had ceased operation.

Beneļ¬ts for the Unemployed

Special Unemployment Insurance Beneļ¬ts Special unemployment ben-
eļ¬ts (trattamenti speciali) offer another alternative to the low regular unem-
ployment insurance in Italy. These beneļ¬ts for limited groups of workers
in industry, construction, and agriculture were introduced in 1968, and pay
out a beneļ¬t in most cases equal to 80 percent of prior wages. The duration
of the beneļ¬t varies from scheme to scheme; in most cases the initial grant
of beneļ¬t was for six months, but this can often be extended, as with CIG.

Unemployment Assistance Beneļ¬ts Protection for the long-term unem-
ployed and ļ¬rst-time job seekers, which in other countries is often provided
by unemployment assistance beneļ¬ts, is the most conspicuous absence in
the Italian arsenal against income loss due to unemployment. In the imme-
diate aftermath of World War II, Italy introduced a special subsidy (sussidio
straordinario) for unemployed persons, which offered a basic assistance ben-
eļ¬t for all unemployed persons regardless of whether they held any unem-
ployment insurance coverage. Similar assistance beneļ¬ts were introduced in
many European countries in order to ease the postwar transition. However,
in contrast with most other European countries, Italyā€™s national government
never introduced a permanent, comprehensive unemployment assistance
program.2 As a result, there is no minimum income guarantee in Italy for
unemployed people who have only scattered contribution records, who have
been out of work for long periods, or who have not yet succeeded in entering
the work force. Since both ļ¬rst-time joblessness and long-term unemploy-
ment are phenomena that disproportionately affect younger members of
the work force in Italy, this lacuna biases the system of unemployment
beneļ¬ts in Italy against younger workers, as we discuss below.
The Netherlands does guarantee a means-tested unemployment assis-
tance beneļ¬t to the unemployed. The level of the beneļ¬t was equal through
1995 to 100 percent of the statutory minimum wage for married heads of
households; following important revisions to the social assistance law in
1995, childless beneļ¬ciaries under the age of twenty-two received only
50 percent of the minimum wage. (The minimum wage is itself linked to
average earnings, such that in 1980 it was equivalent to about 80 percent
of the median wage; by 2003, revisions to the system of labor market regu-
lation and income supports had reduced the minimum wage to 52 percent

2 Nor was there any other form of national-level guaranteed minimum income. A few regional
governments have instituted assistance programs for the unemployed, but these beneļ¬ts
remain discretionary and cannot be considered a basic citizenship right even for residents
of those regions.

Age in the Welfare State

of the median wage.) Recipients of unemployment assistance before 1995
had to be available for work, but there was no time limit on the beneļ¬t nor
was any record of previous employment required. As a result of these two
features, the unemployment assistance beneļ¬t was predominantly used by
two groups: the long-term unemployed and ļ¬rst-time job seekers.
Since 1995 the Dutch unemployment assistance beneļ¬t has been lim-
ited to twenty-four months, and unemployed persons under the age of
twenty-one can receive unemployment assistance only if participating in
a Youth Work Guarantee ( Jeugdwerkgarantiewet, or JWG) activation
scheme. These changes signal a deterioration in unemployment beneļ¬ts
for young people relative to the quite generous years from 1964 through
the late 1980s, which has occurred for reasons discussed in the ļ¬nal chapter
of this volume. But despite the restriction of unemployment assistance ben-
eļ¬ts relative to the ā€œgolden ageā€ of the Dutch social spending, youth unem-
ployed in the Netherlands remain better covered by the welfare state than
in Italy. Youth unemployed in the Netherlands in the last resort still have
access to social assistance beneļ¬ts pegged to the minimum wage, whereas
Italian youth must rely for support on their families.

Unemployment Insurance Extensions for Older Workers The preva-
lence of long-term unemployment among older workers in the Netherlands
has prompted the government to provide extended unemployment beneļ¬ts
for older workers. In the Netherlands until 1999, unemployed workers over
the age of ļ¬fty-seven and a half could continue to receive insurance bene-
ļ¬ts past the two-and-a-half-year cutoff point for extended unemployment
insurance beneļ¬ts until they reached the age of sixty-ļ¬ve ā€“ at which point
they began receiving an old-age pension. This meant that older workers
received the earnings-related beneļ¬t rather than the means-tested social
minimum beneļ¬t that long-term unemployed under the age of ļ¬fty-seven
and a half received after their unemployment insurance beneļ¬ts ran out.
In Italy, extended unemployment insurance beneļ¬ts for older workers have
not been available. In practice, though, older CIG beneļ¬ciaries often con-
tinue to receive beneļ¬ts until they reach retirement (or early-retirement)
age, and the longest duration of beneļ¬ts (forty-eight months) for mobility
beneļ¬ts is reserved for older workers.

Early Retirement Provisions Early retirement has been used exten-
sively in both Italy and the Netherlands as a way of facilitating the exit
from the labor market of difļ¬cult-to-employ or less-productive older
Beneļ¬ts for the Unemployed

workers.3 In the Netherlands, the ļ¬rst experimentation with early retire-
ment schemes began in 1976. These schemes were expanded in the late
1970s in response to high levels of youth unemployment, on the prin-
ciple that facilitating the exit of older workers would make room for
young entrants (OECD 1995, 88). In Dutch law, early retirement contracts
(vervroegde uittreding, or VUT) are negotiated between employers and labor
representatives within an economic sector, and then applied to all ļ¬rms
in that sector. By 1979, around 80 percent of all private sector employ-
ees were covered by collective agreements granting early retirement pro-
visions (OECD 1995, 88). In most sectors the minimum retirement age
under VUT is sixty or sixty-one years, although in some cases it is as low
as ļ¬fty-ļ¬ve (OECD 1993, 71). Early retirees receive the equivalent of a full
state old-age pension and supplementary occupational pension: normally,
early retirement contracts require that employers make up any difference in
supplementary pension beneļ¬ts that would occur as a result of a shortened
contribution history.
In Italy, laws regulating early retirement in speciļ¬c sectors were enacted
in the 1970s. Legislation in 1981 responded to the employment crisis gen-
erated in the wake of the second oil shock, expanding early retirement
provisions to include workers in large ļ¬rms in most sectors of the econ-
omy (Gualmini 1998, 137). Early retirement laws normally allow workers
to retire with full pension beneļ¬ts up to ļ¬ve years early ā€“ although in some
cases early retirement up to ten years early may be compensated.

Disability Beneļ¬ts Both Italy and the Netherlands recorded exceedingly
high rates of disability among the working-age population in the 1980s and
1990s, a peculiarity that in both countries is widely recognized as a result
of the use of disability beneļ¬ts as a substitute for open unemployment.
In the Netherlands since the passage of the Disablement Insurance Act
(Wet op de Arbeidsongschickthheidsverzekering, or WAO) in 1967, and in
Italy since 1974, the criteria for disablement included not just a personā€™s
physical incapacity but the likelihood that he or she could ļ¬nd suitable
work given local labor market conditions. The highly discretionary nature
of this criterion meant that poor labor market conditions for particular
population groups (southerners in Italy, older workers in the Netherlands)
could be used as a reason to grant disability beneļ¬ts, even in the absence
of a readily identiļ¬able disability. Aarts and de Jong (1992, 345) estimate

3 Neither country, however, has ever had a blanket regulation permitting early retirement.

Age in the Welfare State

that in the 1980s between 33 and 52 percent of disability beneļ¬ts in the
Netherlands were really hidden unemployment beneļ¬ts; Franco and Mor-
caldo (1990, chapter 3) estimate a similar ļ¬gure for Italy. In both countries
the labor market criterion for disability was revised (in Italy in 1984 and in
the Netherlands in 1987 and again in 1993), due to the growing expense
of carrying such a large volume of ā€œhidden unemploymentā€ within the dis-
ability pension system. Still, most observers remain convinced that the dis-
ability systems continue to harbor large numbers of essentially employable
Disability beneļ¬ts are an appealing, and expensive, substitute for unem-
ployment beneļ¬ts in both Italy and the Netherlands because they pay more
and last longer. In the Dutch case, disability beneļ¬ts for employees are
related to previous income ā€“ not to the social minimum, as was the case for
unemployment assistance beneļ¬ts. So an unemployed person on disability
would continue to receive a beneļ¬t of 70 percent of prior earnings until the
age of sixty-ļ¬ve, while someone receiving unemployment beneļ¬ts would
have to switch over to the social minimum two and a half years after the
onset of unemployment, when unemployment insurance beneļ¬ts ran out.
In addition, unions and employers negotiated top-off beneļ¬ts for disability
in most sectors to bring the disability beneļ¬t up to 100 percent of the previ-
ous wage (de Vroom and Blomsma 1991, 104). So even for relatively short
spells of unemployment, disability would still be more generous than the
75 percent extended unemployment insurance beneļ¬t. Aarts and de Jong
(1992, 40) report that the after-tax replacement rate of disability beneļ¬ts
for a modal worker in 1980 was 87 percent.
In Italy, as in the Netherlands, disability beneļ¬ts have been an appealing
alternate pathway of exit from the labor market. In southern Italy, where
fewer people are covered under CIGS than in the more industrialized north,
and where long-term unemployment has reached epidemic levels, disability
beneļ¬ts work as a substitute for a regular unemployment insurance ben-
eļ¬t pegged far below the subsistence level. Disability insurance beneļ¬ts,
like old-age pensions in Italy, are calculated based on formulae combining
years of service and prevailing wage rates; there is no uniform beneļ¬t stan-
dard for disability claimants in different sectors. But even at the low rate
pegged to the minimum pension (see chapter 6), disability beneļ¬ts typically
provide a higher income than regular unemployment insurance beneļ¬ts,
since the minimum pension is indexed to wage growth and/or inļ¬‚ation,
while unemployment insurance is not. In 1998, the mean INPS disabil-
ity beneļ¬t was 410.5 euros per month, just above the subsidized minimum
Beneļ¬ts for the Unemployed

pension level (data from INPS 2003).4 At average production worker wages,
this amounted to 26 percent of gross pay.

Jobs Programs If extended unemployment insurance beneļ¬ts, early
retirement, and disability beneļ¬ts provide supplementary forms of unem-
ployment coverage primarily for people late in their working lives, jobs
programs have been used in Italy and the Netherlands to provide both work
experience and supplementary income for unemployed people. It is difļ¬-
cult to assess the impact of tax incentives and direct subsidies provided to
private employers to stimulate employment. However, both the Italian and
Dutch governments have also attempted to provide employment directly
through jobs programs.
The main jobs program for unemployed workers in Italy comes in
the form of socially useful jobs (lavori socialmente utili), which in 2000
employed approximately 200,000 long-term unemployed (Ministero del
Lavoro e delle Politiche Sociali 2001, 35). These jobs, which pay a monthly
allowance of 800,000 lire (Renga 1999), are reserved for workers who have
exhausted other forms of unemployment coverage, especially mobility pay-
ments. There are no public jobs for ļ¬rst-time job seekers, although a num-
ber of subsidies are in place to encourage the hiring of youth workers, loans
for setting up new businesses, and the like.
In the Netherlands, the Youth Work Guarantee Act of 1992 guaran-
teed a minimum-wage job for two years for unemployed people under
the age of twenty-one (twenty-seven if school-leavers). In 1998, the Youth
Work Guarantee was incorporated into the Job Seekers Employment Act
(Wet Inschakeling Werkzoekenden, or WIW) of 1998, which required local
authorities to assist in the placement of youth and long-term unemployed,
in so-called WIW jobs if necessary. In 1998 a total of 42,500 formerly
unemployed people were enrolled in WIW jobs and 5,000 in work-training
positions; one-quarter of these were unemployed youth under the age of
twenty-three. An additional 35,000 long-term unemployed were placed in
jobs under the Entry-Level Jobs and Follow-Up Jobs for the Long-Term
Unemployed (Ministerie van Sociale Zaken en Werkgelegenheid 1999, 33).

Summary Table 5.2 summarizes the beneļ¬ts available to unemployed
people under the variety of programs discussed in this section. The clearest

4 Disability beneļ¬ts paid out by INPS are representative of what most employees in industry,
agriculture, and commerce would receive, keeping in mind that beneļ¬ts for specialized
groups of workers were often higher.

Table 5.2 Beneļ¬ts for the unemployed in Italy and the Netherlands

Number of
Beneļ¬t Amount Duration Beneļ¬ciary pool beneļ¬ciaries
Insurance-type beneļ¬ts
Italy Regular unemployment 30% of previous gross 6 months Employees in most sectors 791,000
insurance wage
Special unemployment 80% of previous gross 6 months Employees in some sectors of 46,000
beneļ¬t wage, up to a ceiling industry and construction
Reduced unemployment 40%ā€“66% of wages 3 months Part-time workers in agriculture 383,000
beneļ¬ts in agriculture
CIG/CIGS (Short-time 80% of previous gross 12 months, extendable Employees in speciļ¬ed ļ¬rms in 84,000
earnings replacement wage industry and construction
Mobility 80% of previous gross 12ā€“48 months Employees in speciļ¬ed ļ¬rms in 93,500
wage industry and construction
Netherlands Regular unemployment 70% of previous gross 6 months All employees except civil servants ļ£“
insurance wage ļ£“
Extended unemployment 70% of previous gross 2 years after expiration All employees except civil servants ļ£“
insurance wage of regular
unemployment ļ£“
insurance ļ£“
Extended unemployment 70% of previous gross Until age 65 Unemployed aged 57.5 yrs or ļ£“
insurance for older wage older
Assistance beneļ¬ts
Italy Social assistance/minimum income ā€“ not available in all regions
Netherlands Unemployment assistance 70% of minimum wage 2 years Long-term unemployed, ļ¬rst-time 370,000
for single person job-seekers. As of 1996, youth
(min. wage is approx. under 21 must enroll in Youth
80% of average wage) Work Guarantee job.
Other beneļ¬ts with unemployment component
Italy Disability pensions Variable: Indeļ¬nite 1,053,000
ā€œMinimum pensionā€
is 26% of median net
wage in industry
Early retirement Variable: Until retirement age 165,000 (not
income-related including
Socially useful jobs 800,000 lire/month 12 months 145,000
Netherlands Work-related disability 70% of gross prior Until age 65 275,000
Early retirement 80%ā€“100% of gross Until age 65 178,900
prior earnings
Jobs schemes Minimum wage (approx. Varies 82,500
80% of average wage)

Note: Data are for 2000.
Sources: Ministero del Lavoro e delle Politiche Sociali 2001 and CBS 2000; 2003.
Netherlands ā€œEarly retirementā€ ļ¬gure is from CBS 2000.

Age in the Welfare State

differences that emerge from this comparison between Italy and the Nether-
lands are the striking dissimilarity in the replacement rates of regular unem-
ployment insurance beneļ¬ts in the two countries; the lack of coverage in
Italy for long-term and youth unemployed; and the importance in Italy of
the CIG, CIGS, and mobility programs that provide superior beneļ¬ts to a
restricted group of employees.

The Age Orientation of Unemployment Beneļ¬ts
The comparison of beneļ¬ts between a variety of more-or-less directly com-
parable programs ā€“ regular unemployment insurance, unemployment assis-
tance, early retirement, disability insurance, and so on ā€“ is necessary to
comprehend the basic institutional similarities and differences between the
Dutch and Italian systems of unemployment protection. But the overlay of
differentiated beneļ¬ts makes it difļ¬cult to grasp immediately the effects of
the system as a whole on speciļ¬c age groups. A cross-national comparison
of the age orientation of entire systems of unemployment beneļ¬ts programs
should take into account both the relative generosity of single programs and
the extent to which these beneļ¬ts are actually available to workers of dif-
ferent ages.
Standard unemployment beneļ¬ts do not tell the whole story. In both Italy
and the Netherlands, welfare state ā€œeffortā€ on unemployment coverage is
seriously underestimated when the hidden unemployment component of
disability beneļ¬ts is not counted. Further, in Italy CIGS (and sometimes
mobility) beneļ¬ciaries are not counted as unemployed, but many aggre-
gate measures include CIGS and mobility payments in the unemployment
expenditures category. Per capita spending measures that use registered
unemployed in the denominator and include CIGS payments in the numer-
ator overestimate the average unemployment beneļ¬t payment in Italy by
50 to 100 percent, depending on the year.
Standard unemployment insurance beneļ¬ts cover widely varying per-
centages of the working population across countries and over time. For
example, Italian unemployment insurance legislation in 1975 applied to
only 51 percent of the working population, as compared with 61 percent in
France, 80 percent in the United Kingdom, and 93 percent in Germany
(Mittelstadt 1975, 5). In the mid-1980s, the actual percentage of unem-
ployed persons with insurance ranged from less than 10 percent in Italy,
Greece, and Portugal to over 85 percent in Belgium (Schmid and Reissert
1996, 4). Figure 5.1 shows changes over time in the percentage of registered








1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990

Figure 5.1 Percentage of unemployed receiving unemployment insurance beneļ¬ts, Italy. Source: Franco 1993, 161ā€“2.
Age in the Welfare State

unemployed receiving unemployment insurance beneļ¬ts in Italy. This ļ¬g-
ure peaks in 1965 with insurance coverage for about 35 percent of Italyā€™s
unemployed, dipping to just over 10 percent in 1981. The new deļ¬nitions
introduced by Eurostat in 1992 result in even lower reported coverage
rates, hovering around 5 percent for the 1990s (Eurostat, Labour Force Sur-
vey Results, various years).
Despite the fact that CIG beneļ¬ts in Italy provide for a higher replace-
ment rate than unemployment insurance beneļ¬ts in the Netherlands, and
despite the fact that as many people receive the generous CIG beneļ¬ts as
received the meager Italian unemployment insurance beneļ¬t, the Italian
unemployment protection system taken as a whole provides remarkably
little to the modal unemployed person, who is a ļ¬rst-time job seeker. The
Dutch unemployment beneļ¬ts system is far more generous in terms of both
beneļ¬t levels and the access that unemployed people of all age groups have
to these beneļ¬ts.
Measuring aggregate spending on programs for the unemployed rela-
tive to other kinds of social spending tells us how the welfare state treats
the unemployed, in general, relative to other population groups such as
the elderly. But we have also seen that different groups of unemployed peo-
ple receive widely differing amounts of income in unemployment, depend-
ing on which type of beneļ¬t program they fall under. To the extent that
different types of beneļ¬ts are more likely to accrue to persons of one age
group than to another, we can also say that the unemployment system itself
may have a built-in age orientation.
The Italian unemployment system has a very clear bias toward older
members of the work force and against younger ones. In 1996, some 60 per-
cent of all unemployed Italians were under age thirty. The rate of unemploy-
ment was 18.1 percent for twenty-ļ¬ve- to twenty-nine-year-olds, and only
4.8 percent for forty-ļ¬ve- to forty-nine-year-olds (Eurostat 1998, Table
008). Yet coverage under Italyā€™s bewildering array of unemployment beneļ¬ts
is clearly concentrated on the older age groups. By 1993, only 4 percent of
registered unemployed people under age thirty in Italy received unemploy-
ment beneļ¬ts of any kind, while 20 percent of those over ļ¬fty did (Schmid
and Reissert 1996, 248). The predominance of younger people in the ranks
of the unemployed means that, overall, Italy provided unemployment
beneļ¬ts for only 6 percent of the unemployed (Eurostat 1998, Table 093).
The under-thirty category in Italy overlaps with two categories of unem-
ployed that are ineligible for insurance beneļ¬ts. First-time job seekers,
who have no right to unemployment insurance beneļ¬ts in Italy, made up
Beneļ¬ts for the Unemployed

Table 5.3 Age distribution of long-term unemployed

Age Italy Netherlands
Under 25 62 22
25ā€“54 35 73
55+ 3 5
Note: Data are for 1985. ā€œLong-term unemploymentā€
is deļ¬ned as greater than one year.
Source: Calculated from Sexton 1988.

Table 5.4 Working-age disability beneļ¬ciaries,
by age

Netherlands Italy
Younga 3
Prime-ageb 29 25
Olderc 68 74
a Netherlands: age 15ā€“25; Italy: age 15ā€“29.
b Netherlands: age 25ā€“45; Italy: age 30ā€“49.
c Netherlands: age 45ā€“65; Italy: age 50ā€“59.

Sources: CBS 2003; INPS 2003.

54 percent of the unemployed in 1996 (Eurostat 1998, Table 093). The
long-term unemployed, another category of unemployed persons with no
right to insurance beneļ¬ts under Italian law, are also concentrated in the
younger age groups of the population (see Table 5.3). Even including the
hidden ā€œunemployment componentā€ of disability beneļ¬ts does not boost
the coverage rates for younger workers, since the age proļ¬le of disability
recipients is similarly skewed toward older workers (see Table 5.4).
The Italian unemployment systemā€™s lack of coverage for ļ¬rst-time job
seekers and the long-term unemployed and its concentration of existing
beneļ¬ts on older members of the work force results in an unemployment
system with two overlapping types of age bias. On the one hand, the system
as a whole is limited in scope as compared with other areas of the welfare
state targeted exclusively at the elderly. On the other hand, even within
the limited conļ¬nes of the unemployment protection system, coverage is
skewed overwhelmingly toward people over the age of forty-ļ¬ve, and away
from workers at earlier stages of family formation, skills formation, and
wealth accumulation.
Age in the Welfare State

The Dutch unemployment system until quite recently was in this regard
very different from the Italian, despite a legislative framework that is explic-
itly biased against younger workers in important ways. The Dutch system
has since its inception graduated beneļ¬ts according to the age and family
status of recipients. Younger unemployed, especially those living at home,
as well as single beneļ¬ciaries with no family obligations, have both lesser
beneļ¬ts and stricter eligibility conditions than mature household heads. In
addition to this legal bias against the young, extended insurance beneļ¬ts
for unemployed workers close to the retirement age explicitly favors older
workers. But this elderly-oriented legislative framework belies a de facto
situation that is much more favorable to younger workers than in Italy.
The situation of younger unemployed workers is less dire in the Nether-
lands than in Italy for several reasons. First, long-terms unemployment is
far less concentrated among the young in the Netherlands than in Italy
(see Table 5.3), which means that more younger unemployed people in
the Netherlands have access to the relatively generous but time-limited
unemployment insurance beneļ¬t. But even those who are not able to claim
regular unemployment beneļ¬ts are better off in the Netherlands than in
Italy. First-time job seekers make up half of the unemployed in Italy, versus
one-quarter in the Netherlands (Eurostat 1998, Table 093). And while ļ¬rst-
time job seekers have no beneļ¬t entitlements in Italy, through the 1970s
and 1980s all ļ¬rst-time job seekers in the Netherlands had a right to claim
unemployment assistance beneļ¬ts. In the 1990s tightened eligibility stan-
dards for unemployment assistance left many more ļ¬rst-time job seekers to
rely on the even less generous welfare (social assistance) beneļ¬t. But new
investments in active labor market policies and jobs programs for young
people mean that joblessness among the young remains much less com-
mon in the Netherlands, where the unemployment rate among twenty- to
twenty-four-year-olds is 6 percent, than in Italy, where 33 percent of this
age group are without work (Eurostat 1998, Table 008).
Rates of coverage under unemployment beneļ¬ts of any kind vary much
less dramatically between age groups in the Netherlands than they do
in Italy: approximately 42 percent of Dutch youth under thirty, versus
approximately 48 percent of Dutch workers over ļ¬fty, are covered (Schmid
and Reissert 1996, Figure 8.3). This is in large part because the risks most
common to young and older workers both receive some form of protection.
Disability insurance, the main alternative to formal unemployment subsi-
dies in the Netherlands, displayed a similarly ļ¬‚at curve of coverage rates for
different age groups relative to Italy (see Table 5.4). In sum, not only does
Beneļ¬ts for the Unemployed

the Netherlands offer on average better protection to unemployed people
than does Italy, but this protection is less skewed toward older groups in
the work force.
As unemployment expanded radically beginning in the mid-1970s, exist-
ing unemployment policies took on distinctive age proļ¬les in Italy and the
Netherlands. Highly protected labor market insiders in Italy man a rapidly
aging ā€œfortressā€ labor market, with younger entrants lacking access to both
jobs and preferred forms of unemployment coverage. In the Netherlands,
as in Italy, disability beneļ¬ts and early retirement buffer the early exit of
older workers from the labor market. But in the Netherlands, universal
unemployment assistance beneļ¬ts until quite recently covered even ļ¬rst-
time job seekers, so the youth segment of the labor market still had access
to some unemployment beneļ¬ts. In addition, job-creation targets linked
to early exit mechanisms and subsidies are more strictly enforced in the
Netherlands than in Italy, so that Dutch beneļ¬ts targeted at older workers
have more closely approximated the desired result of increasing employ-
ment for younger workers. This combines with the more successful Dutch
vocational training system to lessen the concentration of unemployment
within the youngest sectors of the labor market.
In Italy, the absence of coverage for young people, and the very low reg-
ular unemployment insurance beneļ¬ts that are the only form of coverage
for the noncore labor force, have combined with high rates of youth unem-
ployment and an aging of the core work force to skew unemployment ben-
eļ¬ts toward the elderly. The combination of legislated coverage for youth
unemployed and higher beneļ¬t replacement rates for regular unemploy-
ment insurance in the Netherlands has resulted in a less elderly-oriented
welfare state as a whole, as well as a less elderly-oriented unemployment
policy regime, as unemployment has grown since the mid-1970s. We turn
next to an exploration of the origins and trajectories of these policies that
have resulted in diverging age orientations of the unemployment systems
in Italy and the Netherlands.

Particularism, Universalism, and the Response
to Partisan Competition
As with family allowance policies, the response of politicians to partisan
competition, whether a particularistic or programmatic response, has a cru-
cial impact on the eventual age orientation of unemployment policies. How
has the mode of political competition affected the development of the two
Age in the Welfare State

crucial policy features that result in different age orientations of unemploy-
ment beneļ¬ts in Italy and the Netherlands: the replacement rate of regular
unemployment insurance beneļ¬ts and coverage for youth unemployed?
Discussions of Southern European welfare states in general (see Ferrera
1996c; Gough 1996; Rhodes 1997) and Italy in particular (see Paci 1994;
Ferrera 1997) have argued that clientelist partyā€“society linkages are respon-
sible for the fragmentation, particularism, and inefļ¬ciency of the Italian
welfare state. Indeed, the parties most implicated in clientelist practices,
the DC and the PSI, can claim partial credit, or take partial blame, for
the unemployment policies that create such a strong elderly orientation
in Italy. The occupationalism and fragmentation of the labor market pol-
icy system, which provides crucial resources for clientelist politicians, is
itself a response to Italyā€™s predominantly particularistic mode of political
However, the particularistic competitive strategies of center-right parties
also affected the policy preferences and behavior of left parties and labor
unions. Nowhere is this crucial, but rarely noted, consequence of clientelism
more visible than in the area of labor market policies. In an attempt to
mitigate the political effects of the DCā€™s and PSIā€™s clientelist capture of
the system of public administration, the PCI/(P)DS and unions in Italy
unwittingly backed policy solutions that reinforced the elderly orientation
of unemployment policies in Italy.
Of course, the Dutch system of high replacement rates and universal
youth beneļ¬ts is also a result of politiciansā€™ responses to electoral pressures.
But in the Netherlands, these responses were more programmatic in nature
and resulted in a system of more generous and more universal entitlements
against the risk of unemployment for workers of all ages. A more detailed
look at the genesis of these program features will help to explicate how
politiciansā€™ responses to competition in Italy and the Netherlands structured
the age orientation of the unemployment protection systems.

Unemployment Insurance Beneļ¬ts
Why did even the earliest unemployment insurance schemes in the Nether-
lands offer such a high (80 percent) replacement rate relative to schemes in
other countries? It may be that, as union leaders at the time feared (Kuijpers
and Schrage 1997, 91), the compulsory unemployment schemes adopted in
countries other than the Netherlands drove down beneļ¬ts by taking con-
trol of the schemes out of the hands of unions. But the voluntary schemes
Beneļ¬ts for the Unemployed

existing in Italy at the time received much less state ā€œinterferenceā€ than in
the Netherlands, and still offered much lower beneļ¬ts. A more compelling
explanation for the generous beneļ¬t in the Netherlands resides in the fact
that since union funds were subsidized by municipal and later state moneys
(van Leeuwen 1997), they could afford to offer higher replacement rates.
This program structure set the terms for the obligatory unemployment
insurance program adopted after the war. By the late 1930s unions and
employers had reached a consensus that a new, universal unemployment
beneļ¬ts scheme should cover the costs of ā€œnormalā€ unemployment with
contributions from employers and employees, but the ā€œcosts of abnormal
unemployment should be met by government,ā€ as under the old voluntary
system (Kuijpers and Schrage 1997, 93).
If the 80 percent replacement rate was facilitated by rather extensive
state underwriting of the costs of unemployment insurance coverage under
the old voluntary system, how was this generous beneļ¬t maintained when
the new obligatory system was set in place in 1949? The path of least
resistance for legislators was to keep the replacement rate where it was,
while universalizing coverage. Certainly, the unions and their allies in the
PvdA would have been reluctant to turn over control of the voluntary system
they had set up if the beneļ¬t level under the new law were reduced. But the
near absence of debate in Parliament over the new rate (van Kersbergen
and Becker 1988, 485) signals the eagerness of elected ofļ¬cials from Left,
Center, and Right to accept this solution.5 The unexpected popularity of the
Drees emergency pension provisions, which provided a tremendous boost
for the Labor Party in the 1946 elections, had convinced legislators of the
electoral value of universal social insurance. At the same time, avoiding
blame for cuts in replacement rates served politiciansā€™ interests as well as
those of the unions, while the joint employeeā€“employer management of
the system in the Industrial Insurance Boards satisļ¬ed employer demands.
The low replacement rate for Italian unemployment beneļ¬ts after the
war also is related to the level of beneļ¬ts in the prewar system. In Italy,
though, the prewar system offered some of the lowest beneļ¬ts in Europe
(ILO 1922). Even so, Italyā€™s unemployment insurance beneļ¬t continued
to decrease in real terms after the war because beneļ¬ts were not indexed
to either wages or prices and were only sporadically upgraded. How can
we account for this exceptional decrease in social beneļ¬ts at a time when

5 Only a small group of Communist Party legislators dissented, instead advocating a 100 per-
cent replacement rate.


ńņš. 4
(āńåćī 7)