80
Karl Ove Moene and Michael Wallerstein
1990b) estimates that a similar decline
occurred during the 1960s as well, implying
that the variance of log hourly wages in 1983
was only one quarter of what it was in I960.
Three decades of solidaristic bargaining in
Sweden resulted in one of the most egalitari-
an wage structures of any industrialized soci-
ety (Freeman 1988). Similar longitudinal
data is not available for Norway but the pat-
tern appears the same. According to survey
data reported by Kalleberg and Colbjornsen
(1990: 360), wage inequality in 1980 as
measured by the coefficient of variation
(standard deviation divided by the mean) of
log earnings was even lower in Norway than
in Sweden, ɪ
Solidaristic bargaining also had
important institutional effects. Many of the
distinctive institutions and policies that have
characterized Scandinavian social democracy
in the postwar period cannot be properly
understood without reference to the imple-
mentation of egalitarian wage policies. The
centralization of wage setting at the national
level, for example, was endorsed by the uni-
on leadership as a necessary prerequisite for
the reduction of wage differentials on a nati-
onal scale. The active labor market policies
providing retraining and mobility subsidies
to unemployed workers that were adopted in
the 1960s (in Sweden) and 1970s (in
Norway), to cite another widely studied (and
widely admired) innovation, were originally
advocated as a response to the plant closings
caused by a wage policy that kept wages from
falling in inefficient plants (Swenson 1989).
More recently, conflicts over the
appropriate degree of wage equality, both
among the blue-collar unions and between
the blue-collar, white-collar and professional
unions have undermined support for centra-
lized bargaining and the associated corpora-
tist arrangements in both countries (Moene
and Wallerstein 1993, Hernes 1991). Blue-
collar unions that supported solidaristic bar-
gaining began to lose members to white-col-
lar unions that didn't (Lash 1985). Too
much solidaristic bargaining and too little
connection between work performed and
wages received became a prominent and per-
sistent complaint of Swedish employers
(Myrdal 1991). In Sweden, both the centrali-
zed bargaining system and solidaristic bargai-
ning goals were abandoned in the early
1980s. Immediately afterwards, wage diffe-
rentials began to rise. From 1983 to 1990,
the variance of the log of hourly earnings
among Swedish blue-collar workers increased
by 60 per cent (Hibbs and Locking 1991). In
Norway, the system of centralized bargaining
remains intact, but support for egalitarian
wage policies appears increasingly fragile
(Hernes 1991, Hogsnes and Hanisch 1988).
In a related paper (Moene and
Wallerstein 1995), we examine the impact of
solidaristic bargaining on productivity and
employment, a question that is at the center
of current debates over the economic desira-
bility of centralized bargaining systems.
Local wage setting generally results in une-
qual pay for equal work. One of the effects of
industrial unions is to impose equal pay for
equal work at the industry level. Centralized
bargaining implements equal pay for equal
work at the national level. As we discuss
below, there are circumstances under which
the wage equalization associated with centra-
lized wage setting increases economic effici-
ency.
The central focus of this paper,
however, is the politics of solidaristic bargai-
ning. In particular, we investigate who gai-
ned and who lost from three decades of wage
equalization in order to better understand
how such a policy could be implemented,
and why support for greater wage equality