Andrew M. Jones 573bias. The estimates allow for heterogeneous treatment effects and show an impact
on labor market outcomes for low birth weight, but not for high. The implica-
tions of the US results for worldwide health inequalities are explored at the end of
the paper.
Blacket al.(2007) use Norwegian registry data and, like Behrman and Rosenzweig
(2004), they use twins to investigate the impact of low birth weight on long-term
socioeconomic outcomes rather than just short-run outcomes. Within-twins fixed
effects estimates are shown to be significant and similar to standard least squares
estimates for long-run outcomes, such as height, IQ , earnings and education, while
the estimates for short-run outcomes are smaller for the twins data, as suggested
by Almondet al.(2005). The analysis is made possible by the richness of the data,
which use personal identifiers to link all Norwegian births between 1967 and 1997,
as recorded in the birth registry, with other registry data for those aged 16–74
in the period 1986–2002. The register data is augmented with military records
and a survey of twins that identifies zygocity. Within-twin variation is used to
capture unobservable socioeconomic and genetic factors that may confound the
causal effect of birth weight. This means that identification stems from differences
in nutrition in utero (resulting from different placentas for fraternal twins and
different positioning on the placenta for monozygotic). Birth order is included as a
control. The robustness of the findings is assessed by separate analyses for mothers
who have more than one singleton birth, allowing for mother fixed effects rather
than pregnancy fixed effects. To assess the role of zygocity the sample is restricted to
same-sex twins. The sub-sample where there is survey data on zygocity is also used.
The findings are robust, but reveal interesting evidence that those who participate
in twins studies are a self-selected sample. Also it should be borne in mind that
selection into the sample of registry data for long-run outcomes may be affected
by infant mortality. Finally, there are substantial differences between twins and
singletons in terms of factors, such as gestation and the age of their mothers, and
twins usually appear in the lower part of the distribution of birth weights.
12.2.3.3 Communities
Many studies use variation within groups, communities or geographic areas to
control for unobservable factors that are common to all those within the commu-
nity or locality. For example, Wagstaff (2007) controls for village effects in a study
of the impact of health shocks, such as the death of a working-age member of
the household, on incomes of urban and rural households in Vietnam based on
the Vietnam Living Standards Survey (VLSS). Arcidiacono and Nicholson (2005)
find that adding fixed effects for individual medical schools eliminates the posi-
tive peer effects that appear to exist when selection bias is not taken into account.
The inclusion of school effects means that the impact of peer effects on a student’s
achievements and on their choice of specialty are identified by variations over time
within schools in the ability and preferences of students. The aim is to separate cor-
related effects from exogenous peer effects. The study relies on data for graduates
from US Medical Schools over a relatively short period, 1996–98, so identification
may be limited by a lack of variation over time. Currie and Neidell (2005) use