Figure 6: Change in unemployment [on the vertical axis] as a function of trade costs and
the shape parameter of the Pareto distribution for a given change of b1 from 0.4 to 0.8.
(1980) model is combined with search-and-matching unemployment. In the last step of
our numerical analysis, we therefore work with homogeneous firms. We assume that all
firms export, implying that the free entry condition given in equation (23) holds for all
firms.27 Without selection, there are many more firms left that export, as every firm that
enters the market also serves the customers abroad. We find the following result.
Result 6a [Spill-overs in the Krugman economy]
If firms are homogeneous and external economies of scale are important (ν = 1), then an
increase of unemployment benefits in country 1 leads to increases of unemployment in all
countries.
Figure 7 shows the effects of changes of unemployment benefits in country 1 on un-
employment for countries 1 and 2 when firms are homogeneous and ν = 1. We see that
the changes in unemployment are positive for both countries. The mechanism underlying
the spill-over is the change in the market size which results form higher unemployment
in country 1. Hence, the Krugman model highlights the market potential effect in the
absence of firm selection.
It is interesting to contrast this result with those obtained when firm heterogeneity
persisted but was assumed to gradually vanish. There, firms become more equal as γ
becomes larger. The spill-over there becomes smaller with higher γ as only few firms are
productive enough to export. If we focus on the Krugman economy, every firm that enters
the market also exports. In other words, the home and foreign market in the Krugman
model are not separated, rather firms maximize joint profits over these two markets.28
27In this setup, either all firms export or no firm does so. If no firm exports, there are no spill-overs at
all.
28 Or one may argue that firms indeed maximize profits for the home and foreign market separately,
27