An Update of Nunn’s “Slave Trades”

Recent studies on African economic history have emphasized the structural impediments to African growth, such as adverse geographical conditions and extractive colonial institutions. The evidence is mainly drawn from cross-country regressions on late 20th century income levels, assuming persistent effects of historical causes over time. But to which extent has African poverty been a persistent phenomenon? Our study sheds light on this question by providing new evidence on long-term African growth-trajectories. We show that slave trade regressions are not robust for pre-1970s GDP per capita levels, or for pre-1973 and post-1995 growth rates. We calculate urban unskilled real wages of African workers in nine British African countries 1880-1965, adopting Allen’s (2009) subsistence basket methodology. We find that real wages were above subsistence level, rose significantly over time and were, in major parts of British Africa, considerably higher than real wages in Asian cities up to, at least the 1930s. We explain the intra-African variation in real wage levels by varying colonial institutions concerning land alienation, taxation and immigration.

….slave export intensity is highly and significantly (at the 1% level) correlated to GDP per capita in 2000, but not to income levels in 1950 or 1960. In 1970 the effect is significant at the 10% level, but the coefficient is much smaller than in 2000. Column 4 to 6 shows the regression on growth rates including initial GDP per capita (ln). A regression of slave exports on per capita GDP growth is only statistically significant for the period 1973-1995, which explains why the regression on GDP per capita in 2000 is so robust. However, for the periods 1950- 1973 or 1995-2008 the correlation is insignificant and after 1995 the coefficient turns positive. Hence, the claim that Africa’s slave trades affect current economic performance is multi-interpretable.

That is Northwestern’s Marlous van Waijenberg and Utrecht’s Ewout Frankema in an interesting paper on the issue of structural impediments to economic growth in Africa.

I also found this paragraph interesting:

The welfare ratios of urban unskilled workers in pre-modern London and Amsterdam were obviously higher than in late nineteenth century British Africa. However, the average annual growth rates in Accra between 1880 and 1965 (1.17%) were comparable to the average growth rates in London (1.14%, 1840-1900). Welfare growth rates in some other countries were even higher, although it has to be said that these growth rates were affected by very low starting points. In Mauritius we observed the highest long-term growth rate (1.58%), which suggests that the Mauritian ‘Miracle’ is not just a post-colonial phenomenon.31 In sum, we find little evidence that suggests that four generations of African urban wage workers in the colonial period were trapped into persistent poverty. Welfare improvements were certainly not confined to very specific regions in British Africa or brief periods of time (such as 1945-1960). In fact, the whole idea that Africa has been the poorest and most slowly growing region since the Industrial Revolution is based on a backward extrapolation of post-1960 growth experiences without a historical empirical foundation.

The paper is not about the Nunn thesis per se, but investigates more generally whether historical moments (like the slave trade period) produced structural impediments that have made Africa perennially the slowest growing region of the world (hence the need for explaining the “Africa” dummy in popular research and thought).

I wish development economists read more history, especially economic history.

Africans Covered 98% of the Cost of Administering Colonial French West Africa (AOF)

Elise Huillery writes in the Journal of Economic History:

What share of French expenditure was allocated to West Africa? What share of West Africa’s revenue was provided by France? These two questions are crucial since scholars and politicians who claim colonization had a “positive role” make essentially the two arguments that the colonies benefited from imperial public investments and that mainland taxpayers sacrificed local investments for investments in the colonies.

I find that the costs of AOF’s colonization for the metropolis were low. From 1844 to 1957 France devoted on average 0.29 percent of its public expenditures to AOF’s colonization. Colonization of French West Africa was profitable for France to the extent that the impact on cumulative domestic production exceeded 3.2 billion 1914 francs. The military cost of conquest and pacification accounts for the vast majority (80 percent) of the average annual cost. The cost of central administration in Paris accounts for another 4 percent. So subsidies to AOF account for only 16 percent of the average annual cost, meaning that less than 0.05 percent of annual total metropolis public expenditures were devoted to AOF’s development.

For French West African taxpayers, French contribution was not as beneficial as has been argued. From 1907 to 19578 the metropolis provided about 2 percent of French West Africa’s public revenue. Local taxes thus accounted for nearly all of French West Africa’s revenue. These resources supported the cost of French civil servants whose salaries were disproportionally high compared to the limited financial capacity of the local population. Administrators, teachers, doctors, engineers, lawyers, and so on, were paid French salaries and got an additional allowance for being abroad. Thus, in the colonial public finance system, most revenues were collected on an African basis while being spent on a French basis. To illustrate this point, I show that colonial executives (eight governors and their cabinets) and district administrators (about 120 French civil servants) together accounted for more than 13 percent of local public expenditures.

The rest of this very fascinating paper is here.

Besides the headline finding, also interesting in the paper are: (i) the extent to which Paris subsidized private firms involved in the colonial enterprise; and (ii) the structure of the public finance system that allowed the AOF administration to borrow directly from French banks with the full backing of Paris (which allowed for lower rates). This might explain the persistence of the monetary relationship between former AOF territories and Paris in the form of the CFA and a common central bank (BCEAO).Screen Shot 2015-07-05 at 12.05.23 AM

As I keep saying, Economic History is hot again. And sooner rather than later it’s going to become more apparent to more people that African political and economic history did not begin in 1960, or for that matter in 1884-5. And neither was it just about the unimaginably catastrophic Atlantic experience.

Education and Human Capital Externalities (in Benin)

Wantchekon, Klasnja and Novta have a really cool paper (forthcoming in QJE) investigating the relationship between human capital and development:

Using a unique dataset on students from the first regional schools in colonial Benin, we investigate the effect of education on living standards, occupation and political participation. Since both school locations and student cohorts were selected with very little information, treatment and control groups are balanced on observables. We can therefore estimate the effect of education by comparing the treated to the untreated living in the same village, as well as those living in villages where no schools were set up. We find a significant positive treatment effect of education for the first generation of students, as well as their descendants: they have higher living standards, are less likely to be farmers, and are more likely to be politically active. We find large village-level externalities – descendants of the uneducated in villages with schools do better than those in control villages. We also find extended family externalities – nephews and nieces directly benefit from their uncle’s education – and we show that this represents a “family-tax,” as educated uncles transfer resources to the extended family.

The amazing finding is that having just one educated person in an extended family makes a significant difference, not only for the educated person’s offspring, but also for their nieces and nephews:

These descendants have better education at all levels than descendants (either children on nieces and nephews) in families where no progenitor was educated. These effects are statistically significant and substantial  – such descendants are 20% more likely to have primary school education, 19% more likely to have secondary school education and 11% more likely to go to university..

The main takeaway of the paper is that investment in human capital has a positive effect on long-term development that is independent of (colonial) institutions.