the bargaining power of labor. In contrast, relatively export-oriented states should
experience more strike activity, as the increasing demand for labor strengthens labor's
bargaining position.
This research also leads us to expect an indirect effect of globalization on strike
activity through its impact on union density, which is generally thought to be a major
predictor of strike activity. Most scholars argue that international competitive pressures
should reduce union density through its effects on the bargaining position of labor. If so,
we expect that relatively import-sensitive states should have lower union density.
Because it is possible that export orientation may have the opposite effect, strengthening
the bargaining position of labor and increasing union density, we will test that hypothesis
as well.
The second set of hypotheses comes from Tsebelis and Lange (1995), who argue
that the bargaining position of labor and capital should affect the demands each makes
rather than the probability of a strike. They emphasize instead sources of uncertainty that
might lead to a failure in the bargaining process, resulting in a strike. Rapid
globalization, as indicated by changes in import sensitivity or export orientation, should
create uncertainty of this sort and thus make strikes more likely. Because this line of
argument suggests that both positive and negative changes in these conditions should
both increase uncertainty, we will use the absolute value of the change in these indices to
test this hypothesis.
The third set of hypotheses stems from the relative deprivation literature. The
principal implication of this research is that strikes, as a form of protest, should be more
likely when conditions fall below workers' expectations. Different assumptions about
19
Convention