3028 migration and the labor force

as the pivotal factors often were in place years
before the migration began. A case in point is
the US black migration out of the South. Given
that the higher incomes of the North from
industrialization predate 1940, why did the
pace of the black Southern exodus quicken so
much after 1940? Perhaps a certain threshold
must be passed in terms of the characteristics
of the group in the original location and the
fortitude of the network in the new location
before a sizable group of potential migrants is
ready to migrate. In both internal and interna
tional migration, it is not the poorest poor who


The second focal point in studies of migration
and the labor force is how migration affects the
labor force of the host region or country. Social
scientists are far from resolving whether immi
grants hurt, help, or have no significant effect
on native born employment and wages. Angrist
and Kugler (2002) find a negative immigration
effect on employment in countries with restric
tive institutions. Hartog and Zorlu (2002) find
very small immigration effects on natives wages
in their analysis of the Netherlands, Britain,
and Norway. Using aggregate time series data
for Australia, Pope and Withers (1994) find
increases in immigrant labor positively affect
natives wages. Hercowitz and Yashiv (2002)
estimate negative employment effects in a study
of immigration from the former Soviet Union
to Israel, while an Italian study (Venturini &
Villosio 2002) finds no negative effect on
natives employment.
Several interrelated problems contribute to
this diversity of results. These include difficul
ties with measuring immigration, disentangling
immigrations effect on natives from the effect
of economic conditions on immigration, the
clustering of immigrants within a country, the
migration of natives in response to immigration,
reconciling statistical evidence with anecdotal
evidence and theoretical expectations, and fail
ure to consider how industries may change their
production practices or develop in response to
Immigrants may move to areas with better
than average wages and employment opportu
nities, thereby obfuscating any potential adverse
immigration effect on the economic status of
the native born in cross area analyses. Even if
immigrants do not locate in response to the
economic conditions of areas, their presence
may still correlate with economic conditions.
Natives wages in areas of high immigration
may be lower than they would have been in
the absence of immigration, even though the
across area snapshot reveals a positive associa
tion between percent immigrant and native
born wages and employment. This problem also
afflicts time series; historical analyses have
consistently shown an inverse relationship
between immigration and the host countrys
unemployment rate. Natural experiments may
help elucidate migrants impacts on natives (see,
for instance, Hunt 1992).
Another problem that may contribute to
the variety of estimated effects is that immi
grants tend to cluster in a few places. In the
US, immigrants are concentrated in six states.
Within those states, they are concentrated in
the largest cities or SMSAs. One time cross
state estimates of the effect of immigration on
native born employment and wages (or the
effect of changes in immigration) may be sen
sitive to economic circumstances (or changes in
economic circumstances) in any of the principal
immigration states. The problem of isolating
the effect of immigration on native born eco
nomic status from the effect of perturbations in
these states economies can only be overcome
by using time series information in combination
with cross sectional information.
In gauging the effect of immigration on
natives economic outcomes, domestic and
international migration research intertwine.
Several scholars argue that analysts find little
or no immigration effect on native born wages
and employment in cross area comparisons
because natives move in response to immigrant
inflows: immigrants reduce the wages and
employment opportunities of the native born
and, in response, natives leave. No wage or
employment effect is observed in analyses com
paring areas of high and low immigration since
the native born labor supply in high immigra
tion areas has decreased with the out migration
of natives from these areas. This migration