Colonial Investments and Long-Term Development in Africa: Evidence from Ghanaian Railroads (with
Remi Jedwab) Web Appendices
Abstract: What is the impact of colonial infrastructure investments on long-term development? We investigate this issue by looking at the effects of railroad construction on economic development in Ghana. Two railroad lines were built by the British to link the coast to mining areas and the hinterland city of Kumasi. Using panel data at a fine spatial level over one century (11x11 km grid cells in 1891-2000), we find strong effects of rail connectivity on the production of cocoa, the country's main export commodity, and development, which we proxy by population and urban growth. First, we exploit various strategies to ensure our effects are causal: we show that pre-railroad transport costs were prohibitively high, we provide evidence that line placement was exogenous, we find no effect for a set of placebo lines, and results are robust to instrumentation and matching. Second, transportation infrastructure investments had large welfare effects for Ghanaians during the colonial period. Colonization meant both extraction and development in this context. Third, colonial railroads had a persistent impact: railroad cells are more developed today despite a complete displacement of rail by other transportation means. Physical capital accumulation and the demographic transition account for path dependence. Colonial railroads thus shaped the economic geography of Ghana.
that Divide: Education and Religion in Ghana and Togo since Colonial Times (with
Abstract: When European powers partitioned Africa, individuals of otherwise homogeneous communities were divided and found themselves randomly assigned to one coloniser.
This provides for a natural experiment: applying a border discontinuity analysis
to Ghana and Togo, we test what impact coloniser’s policies really made. Using
a new data set of men recruited to the Ghana colonial army 1908-1955, we find
literacy and religious beliefs to diverge between British and French mandated part
of Togoland as early as in the 1920s. We attribute this to the different policies
towards missionary schools. The British administration pursued a ”grant-in-aid”
policy of missionary schools, whereas the French restricted missionary activities.
The divergence is only visible in the Southern part. In the North, as well as at the
border between Ghana and Burkina Faso (former French Upper Volta), educational
and evangelization efforts were weak on both sides and hence, did not produce any marked differences. Using contemporary survey data we find that border effects
originated at colonial times still persist today.
Referral and Job Performance: Evidence from the Ghana Colonial Army (with Marcel Fafchamps)
compiled from army archives, we test
whether the referral system in use in the British colonial army in
Ghana served to
improve the unobserved quality of new recruits. We find that it did
anything, referred recruits were more likely to desert and be dismissed
or unfit. We find instead evidence of referee opportunism. Army
have been aware of this problem by insisting that referred recruits
Heights and Development in a Cash-Crop Colony: Living Standards in Ghana, 1880-1980
(with Gareth Austin and Joerg Baten)
While Ghana is a classic case of economic growth in an agricultural-export colony, scholars have queried whether it was sustained, and how far its benefits were widely distributed, socially and regionally. Using height as a measure of human well-being we explore the evolution of living standards and regional inequality in Ghana from 1870 to 1980. Our findings suggest that, overall, living standards improved during colonial times and that a trend reversal occurred during the economic crisis in the 1973-83. In a regression analysis we test several covariates reflecting the major economic and social changes that took place in early twentieth-century Ghana including railway construction, cocoa production, missionary activities, and urbanization. We find significant height gains in cocoa producing areas, whereas heights decreased with urbanization.
Inequality in Net Nutritional Status
Anthropometry provides one of the
best tools to assess nutritional and health status. Nutrition and
health influence bodily growth positively and hence, body stature can
shed light not only on the average endowment of nutritional and health
inputs but also on an unequal consumption thereof. Comparing mean
heights of social groups is a popular approach for measuring height
inequality. Studying height differences between social groups, however,
ignores inequality within those groups. Consequently, conclusions on
total inequality can be misleading. In this paper, I address the
measurement of height and health inequality within populations. Based
on empirical findings, I develop a theoretical model, from which I
infer how the height distribution responds to increasing inequality.
Unequal societies tend to have a larger standard deviation in heights
as the input-induced variance adds to the biological variance of
heights. This property can be used for a measure of height inequality
within populations. Finally, I draw attention to potential pitfalls in