Unemployment in an Interdependent World

order of magnitude bigger. These two extreme scenarios open the interval in which the
‘true’ size of spill-overs is likely to lie.

In order to measure the empirical importance of spill-overs, we include trade-weighted
foreign variables into otherwise standard cross-country unemployment regressions run
on panel data for 20 rich OECD countries. The empirical evidence is in line with our
theoretical findings. Instrumenting the average foreign unemployment rate by foreign
exogenous variables and their time lags, and controlling for business cycle effects and own
labor market variables, we find a strong positive correlation between unemployment rates
of countries. Regressing the domestic unemployment rate directly on foreign institutions
confirms this finding. Moreover, we document that the importance of the foreign variables
for domestic outcomes is larger, the less domestic product markets are protected and the
more open the domestic economy is. We do not find any support for the prediction of
multi-sector Heckscher-Ohlin-type models, namely, that the correlation between domestic
labor market outcomes and foreign labor market regulation depends on the capital-labor
ratio. Hence, our empirical results confirm the qualitative predictions of our theoretical
model. On average, our results suggest that the effect of foreign institutions on domestic
unemployment is about 10% as strong as the effect of domestic institutions. We show
that, in order to replicate this finding, the Mortensen-Pissarides search-and-matching
model requires a stronger degree of real wage rigidity than the one provided by the usual
individual bargaining approach.

45

<<   44   45   46   47   48   49   50   >>

More intriguing information