In this paper we propose an alternative method for investigating the s
ources behind the behaviour of real wages and unemployment. The model
we study is a certain cointegrated VAR, a so-called common trends mode
l, which has become increasingly popular in the empirical growth/busin
ess cycle literature. This model leads us to emphasize the distinction
between long-run and short-run relations. Using quarterly Swedish dat
a (1965-90), we find only weak evidence of a short-run relation betwee
n real wages and unemployment as reported in traditional single-equati
on error correction models. There is even less evidence of a long-run
relation.