Ciclo de Seminarios
de Economía 2013
Washington University in St. Louis
“The Dynamics of Immigrant Earnings”
"La dinámica de los ingresos de los inmigrantes"
Fecha: Jueves 20 de junio, 17:00 hs.
Lugar: Prudencio de
Pena 2544 - Salón B001 - Universidad de Montevideo
It is a well-established fact that
earnings of immigrants and natives are not equal even when they have
similar levels of schooling. Moreover, the evidence suggests that the
dynamics of earnings over the life cycle are different for natives and
migrants. We summarize the evidence in the following "facts":
Migrants vs. Stayers. Migrants and stayers with the same schooling
have different age-earnings profiles. Upon migration, migrants earn more
and their earnings grow faster than the earnings of stayers.
Migrants vs. Natives. Even after controlling for schooling,
migrants earn initially less than natives. However, they show steeper age
earnings profiles. The gap in initial earnings is negatively related to
the level of GDP per capita in the country of origin.
Option to Migrate. There is limited but suggestive evidence that
increases in the probability that individuals be allowed to migrate to a
high wage location has a positive (and large) effect on investments in
human capital of the whole eligible population.
In addition to those
"facts" there is some evidence that suggest that:
Scale: Migrants tend to be drawn from the ranks of the high
Positive Selection: The mix of skilled/unskilled (immigrants) in
the destination country is higher than in the sending country.
Positive Sorting: The higher the wage in the receiving location
the larger the ratio of skilled to unskilled immigrants.
This paper presents a version of the human capital model
originally developed by Ben Porath augmented
with the decision to migrate to understand the forces that can account
for these observations. Our work can be viewed as a dynamic version -with
endogenous accumulation of human capital- of the more standard Roy model
that is the workhorse of the migration related literature.
First, we explore the theoretical implications of the model in the
case of unanticipated migration. We show that, at least qualitatively,
the predictions of the theory match the evidence about the difference in
age earnings profiles between migrants and stayers. In this case the key
mechanism is that migrants and stayers face different prices and, hence,
they make different decisions in terms of on-the-job training efforts
that result in a steeper age-earnings profile for the individuals in the
high wage location.
We then explore the implications for the differences between
migrants and natives. A key insight is that in the model if two
individuals chose to acquire the same amount of schooling in different
economic environments, then they cannot be identical. Moreover, in a
model that allows education to be two dimensional ---with a quantity
dimension as measured by years of schooling, and a quality dimension that
we view as a better measure of human capital--- it is optimal for the
individual in the low wage location to choose a lower quality of
schooling. This, in turn, implies that he has lower human capital than a
native at the time of migration. These two results explain both the lower
initial earnings ---because initial human capital is lower - and the
steeper age-earnings profiles - because the migrant must have had higher
innate ability to choose the same years of schooling in a lower return
environment. Moreover, since the "quality" of human capital
holding years of schooling constant varies positively, the model also
accounts for the observation that migrants from poorer locations earn
less than migrants from higher income regions.
Our view implies that differences in earnings are not a good
measure of differences in TFP across countries. Thus, our approach
contrasts with the more standard approach, as exemplified by Hendricks
(2002), that views differences in earnings for migrants and natives with
the same schooling as direct measures of TFP. This view is silent about
the differences in age-earnings profiles.
We study the impact of facing individuals with a known probability
of migration in a context in which migration is costly in terms of goods
and human capital. Our main result suggests that whether there is
positive or negative selection depends in a somewhat complicated manner
on the "size" of the stock of human capital lost, the
differences in wage rates between the sending and receiving locations and
the age of the potential migrant.
We then use a calibrated version of the model to understand
whether human capital and selection can account for the evidence on the
earnings of immigrants in the U.S. Even though our exercise is very
preliminary, we find the results encouraging as the model seems
consistent with the data".
Rodolfo Manuelli es doctor
en Economía por la
Universidad de Minnesota, “James S. McDonnell Distinguished University
Professor” en Washington University in St. Louis e Investigador en el Federal Reserve Bank of St. Louis.
Se especializa en las áreas de
Crecimiento y Desarrollo
Económico, Macroeconomía y Economía Monetaria.
Por más información:
Seminarios de Economía 2013
Para ser removido de la presente lista o
cambiar su contacto, favor indicar respondiendo a este mail.
académica sin costo
Confirmar asistencia a:
firstname.lastname@example.org / 26042544