is at the symmetric equilibrium. It is easy to see that if labour allocation between regions is
identical, that is S1 = S 2 = U1 = U2 = 1/2, the nominal wages are the same in the two regions,
w1=w2. Assume now that as a consequence of a shock to the symmetric equilibrium L1 > L2,
that is some workers are allocated from region 2 to region 1. If this change in stocks
positively affects real wage in region 1 relative to region 2, it will encourage migration as
initial symmetric equilibrium is unstable. The opposite is true if the real wage differential
becomes negative.
The initial movement of a single worker has four effects. 15
The first is the price-index effect, working in favour of divergence. As mentioned
above, the cost of living will be lower in the country with the larger manufacturing sector,
because a smaller proportion of trade costs will be paid for by the same bundle of
manufactured goods.
A second agglomerative force, the home-market effect, reinforces this effect. From
(26), nominal wages in a region will tend to be higher if income in the region is high. The
reason is that firms can afford to pay higher wages if they have good access to a larger market
As a consequence, the large market retains a more than proportional share of manufacturing
sector. The idea is not a new one but finds its root in the literature on market potential (see
Harris, 1954).
Third is the competition effect. The presence of more firms in the local market will
increase competition to serve the regional immobile consumers thus tending to reduce local
profits and so encouraging the stability of the symmetric outcome.
Finally, the migration of a skilled worker has a positive effect on the productivity of
skilled workers in the host region, and therefore a positive effect on their nominal wages. We
call this last effect which is peculiar to high-skilled workers migration, skill premium effect.
22