Interview with Professor James Fenske
James Fenske is a Professor of Economics at the University of Warwick and is highly regarded for his innovative studies charting the economic history of developing countries. His work has been featured in reputable outlets such as the Journal of Development Economics, Economic History of Developing Regions, and the Review of Economics of the Household.
- Professor Fenske, you write extensively on Africa so what sparked your interest to study this continent?
In Africa, there is always another story. Take, for example, the book “Slavery in Africa” by Miers and Kopytoff. They bring together studies of how slavery worked in several African societies. The volume’s 78-page opening chapter starts by criticizing western scholars for their failure to appreciate the nuances of slavery in Africa relative to the plantations of the Southern United States and the Caribbean. That’s the first story. They then position African institutions of slavery against Western concepts, arguing that in land-abundant Africa people were scarce and African societies were happy to receive more people through any means, including slavery. The rights of an owner over a slave were not so different from other rights in persons, like those of a kin group over its members, and slaves were on the path of reincorporation from one society into another. That’s the second story. And then Robertson and Klein begin their book with the phrase “Most slaves in sub-Saharan Africa were women.” That’s the third story.
You can spend a lifetime reading about Africa, and still not know the full story.
- Africa is the poorest continent, according to the World Population Review. Could you explain why this is the case?
The World Development Indicators say that GDP per capita in sub-Saharan Africa in 2020 was $3,718.00 in PPP-adjusted 2018 international dollars. In the UK, the same number was $46,482.90. That’s a difference of about 1200%.
Now suppose, I naively do the following. I download Nathan Nunn’s data from his 2008 QJE paper from his website, and I regress log GDP per capita on log slaves exported per unit area. Then I compute the counterfactual log GDP per capita for each country implied by the fitted values had it exported zero slaves. The average country becomes 112% richer. The country that gains the most is Ghana, at 270%. But 270% is a long way from 1200%.
And this is only one of many examples. Over the last twenty years and longer, considerable evidence has accumulated that several measurable variables negatively affect African economic performance – the slave trades, constraints imposed by geography, the nature of pre-colonial institutions, the nature of post-colonial governance, the debt crises, problems with aid, problems with war, commodity prices, droughts, urban bias… Sometimes there is a set of right-hand side variables that, in the style of a 1990s cross-country regression, can push the so-called “Africa dummy” to levels of statistical insignificance. And sometimes it can’t. You can put together a set of factors, take the most credible estimates that exist of their treatment effects, and find a sum that reaches 1200%. But if you were to fix any one problem, the returns to solving others might fall. Or they might rise.
And then Ewout Frankema and Marlous Van Waijenburg tell me that wages were as high in Africa in 1900 as they were in Asia. So perhaps everything prior to 1900 is irrelevant in understanding Africa today.
Understanding economic growth is hard.
- Property rights are crucial to economic growth, but some contend that property rights in Africa are insecure due to land abundance. Is this an accurate assessment?
The clearest statement of this view comes from Gareth Austin, in his 2008 Economic History Review article. Though you can find versions of it in the work of John Iliffe and Anthony Hopkins. Gareth recently gave an excellent recapitulation of this perspective in a keynote speech at the Economic History Society annual conference. According to this view of the world, because land was abundant relative to people in Africa before colonial rule, it led to less investment in property rights over land. But the abundance of land also raised wages, spurring greater reliance on coercion. Related work by anthropologists and historians like Stanley Tambiah, Jack Goody, and Igor Kopytoff have tied Africa’s historically sparse populations to outcomes such as bride price and polygamy.
And yet the world has changed dramatically since the colonial period. According to the World Bank’s estimates and projections, the world’s population was a bit over 3 billion in 1960, with some 227 million in sub-Saharan Africa. Today those numbers stand at about 7.9 billion and 1.2 billion. For 2050, their projections are a world of 9.7 billion people with 2.2 billion in Africa. If these numbers are right, Africa’s population density was 9.5 per square kilometer in 1960, 50.3 today, and will be 92.2 per square kilometer in 2050. Herbst, in his “States and Power in Africa” puts African population density at 4.4 per square kilometer in 1900. But today’s 50.3 makes Africa’s population denser than either China or South Asia in 1900. Ninety persons per square km would give Africa the population density of the UK in the Victorian era.
The question may not be whether African property rights are as they are because the population is scarce, but how they will adapt to the continent’s rising population.
- Many argue that the legacy of the transatlantic slave trade still affects Africa. Is this claim verified by empirical research?
Absolutely! Nathan Nunn’s work on GDP and his work with Leonard Wantchekon on trust paved the way for an entire generation of scholars to examine the influence of the slave trade on a wide range of other factors. Nonso Obikili has shown that the slave trade increased political fragmentation and reduced literacy. John Dalton and Tommy Leung have shown the Atlantic slave trade raised polygamy, while Edoardo Teso has shown that it raised women’s labor force participation. In forthcoming work, Warren Whatley has put together evidence that the slave trades spread an entire package of institutions, including slavery, polygamy, and centralized aristocratic forms of power.
I think there is still a lot of work that can be done by economists on this topic, not only in Africa, but looking at the African diaspora and its connections with Africa. I have also seen some exciting new work presented recently at conferences on the role of the slave trade in European development – hopefully, these will be circulated widely as working papers soon.
- Political theorists submit that warfare improved state capacity in Europe, but what are the findings for Africa?
The key counterpoint here is the work of Charles Tilly, who argued that in Europe “war made the state and the state made war.” For Europe, Tilly argued that fiscal pressures brought on by the military revolution led to the development of state capacity. Later writers like Nicola Gennaioli, Joachim Voth, and Mark Dincecco have provided evidence that economists would recognize in support of Tilly’s view, and they have traced the long-run consequences for urbanization, fiscal centralization, limited government, and ultimately, development in Europe.
But this may not be a story that holds elsewhere. Several historians and political scientists have considered this story and concluded that it cannot apply to Africa. Jeffrey Herbst, for example, argues that because of Africa’s relatively low population densities, war was over people and not over land. Richard Reid and Martin Klein have pointed out that raiding wars of the sort that dominated war in sub-Saharan Africa, in contrast to the campaigning wars of Europe, left more room for the prolongation of wars. Further, Robert Bates has argued that the colonial peace cut short the military revolution in Africa. Together, these helped undermine the links between historic war and later development in Africa. It is no surprise, then, that Timothy Besley and Marta Querol-Reynal have found that, at the grid cell level, parts of Africa that experienced more conflict before colonial rule experience more violence today.
- Though you are best known for covering Africa, your research also explores India. Could you comment on the relationship between pre-colonial warfare and state capacity in India?
The environment of pre-colonial warfare between states in pre-colonial South Asia resembled the inter-state military competition that existed in Europe after the military revolution as described by Charles Tilly. Madhabi Roy has made a convincing case that fiscal resources mattered for military success. And, unlike sub-Saharan Africa, South Asia before 1757 was relatively densely populated. And so, one might expect Tilly’s description of Europe to have explanatory power in South Asia that it does not in Africa. Writers like John Richards, Andrew de la Garza, and Pratyay Nath have indeed described state-building efforts under the Mughals that were responses to military pressures. The empire gave land to military officials to extract agricultural surplus, developed a pyramidal treasury system, and implemented a land revenue tax.
Some of the most interesting empirical work on this has been contained in PhD theses. Roberto Foa finds there is greater state effectiveness today, even looking within modern Indian states when comparing districts just within the boundaries of the subcontinent’s eighteenth-century challenger states to those just outside of these boundaries. Safya Morshed recently presented a fascinating paper in the Economic History of Developing Regions seminar in which she connects the threat of warfare faced by the Mughals to their appointments of government officials and the salaries they paid them.
- India is a diverse country with multiple languages and diversity is usually portrayed as a barrier to economic growth. Based on your research is linguistic diversity adversely affecting development in India?
My research has been narrower. A few years ago, Namrata Kala and I published a paper in the Journal of Economic History in which we showed that linguistic distance predicts less market integration in colonial India. That is, if I take a pair of markets in districts that speak dissimilar sets of languages, the prices of staple goods like wheat and rice correlate less strongly across that pair of markets than across other pairs that speak more similar languages, conditional on other measures of how far apart these markets are.
One of my favorite papers on linguistic diversity and development in India was written by Tarun Jain a few years ago and came out in the Journal of Economic History. He shows that districts within colonial provinces that did not share the official language of the province lagged in education. But in 1956 Indian states were reorganized on linguistic lines. As a result, the formerly mismatched districts began to catch up.
If there is a dimension along which diversity is likely to matter, it is caste. Latika Chaudhary has done important work showing how caste fragmentation limited access to education in colonial India. Suanna Oh’s job market paper showed that caste identity is a major constraint on the jobs workers will accept and Kaivan Munshi has written a thorough review for the Journal of Economic Literature. David Reich, in his book “Who we are and how we got here,” shows just how strong the pressures of endogamy are within castes. For the Vysya of southern India, for example, there is strong evidence of a genetic bottleneck dating back two to three thousand years that would have been erased had there been even a 1% influx from other castes per generation. In his words, India does not have a large population – rather, “India is composed of a large number of small populations.”
- The Colonial era in India is often maligned, however is there evidence to suggest that Indians benefited from Infrastructural developments?
There is! Many writers have charged that the railroads had low productivity, charged too much for freight, and contributed to the drain out of the country by guaranteeing returns to investors. But many economists have shown that the colonial railroads did have impacts that would benefit the average Indian. Tahir Andrabi and Michael Kuehlwein have shown that they reduced price gaps over space. Dave Donaldson has shown that railroads increased rural incomes. He and Robin Burgess have shown that railroads helped break the link between drought and famine.
And this is not a story limited to railroads. Aaditya Dar has a recent working paper in which he shows that districts that received colonial canals were more successful in adopting the new crop varieties of the Green Revolution – though today they are now struggling with greater depletion of groundwater.
It is striking though, how much of this literature is based on comparisons across locations in India. Did a city connected to a railroad grow faster than one that didn’t? Did a district connected to a railroad see less famine afterward compared to one that did not? Some of the biggest questions in the literature on colonialism – “were railroads part of the drain?” – may be beyond the toolkit of applied econometrics. And so, there is considerable scope for other economists who specialize in topics like finance or macroeconomics to make important contributions here.
- India’s Green Revolution has been heralded as a watershed in the country’s history, so how did it improve social indicators?
This is a story much broader than India. Douglas Gollin, Casper Worm Hansen, and Asger Mose Wingender recently published a paper in the Journal of Political Economy in which they demonstrate that the new crop varieties introduced by the Green Revolution raised incomes in the developing world. Had the Green Revolution come a decade later, incomes would be lower by more than 15%. And they show that the Green Revolution reduced birth rates and raised life expectancy. Jan von der Goltz and his coauthors, in a recent paper in the Journal of Health Economics, use data on more than half a million births from around the developing world to show that the diffusion of new crop varieties reduced infant mortality. And of course, there is classic work by Andrew Foster and Mark Rosenzweig on how, by increasing the returns to schooling, the Green Revolution spurred greater schooling in India.
There may, of course, have been costs as well. Sheetal Sekhri and Gauri Kartini Shastry have a recent working paper that suggests the greater abundance of wheat and rice in India may have led, in later life, to more diabetes among men. Other writers have focused on the potential downsides of increased fertilizer use. There is, I think, a lot more for economists to contribute here.
- Thanks for your participation. Do you have recommendations to improve development in poor countries?
Borrow over at long maturities at a fixed interest rate. And in your own currency if they’ll let you. Don’t do anything you wouldn’t want your god to see