Institutions rule: COCD's path to the Nobel in economics this year

Why Colonial Origins of Comparative Development is Nobel-worthy

When I first read the “Colonial Origins of Comparative Development,” (COCD) by Daron Acemoglu, Simon Johnson and James Robinson (AJR) more than 20 years ago, I was smitten. So much so that I told several colleagues at the International Monetary Fund that the paper would one day win the Nobel Prize. I say this not to trumpet my predictive abilities (mostly poor) but to illustrate that Keatsian sense of being awestruck: “Then felt I like some watcher of the skies, When a new planet swims into his ken.”

At that time, pioneered by Robert Barro, an entire sub-discipline of economics was striving to explain the differential economic growth performance of countries in the post-World War II period. Around that time too, Jared Diamond had produced his tome, “Guns, Germs and Steel”—breathtakingly magisterial and multi-disciplinary in method — and yet able to advance a mono-causal explanation for the most fundamental development question: Why are some countries rich and others poor? Diamond’s answer was: It’s all about geography. 

COCD came along to offer a counter-explanation, but did so differently. COCD drew upon history, unearthed imaginative historical data, deployed it in a simple, comparative and quantitative, economic setting and produced an answer that was striking and plausible. It is hard to convey how audaciously creative and parsimonious COCD was in that combination of setting, method, data and narrative.

AJR’s answer to the big question was doubly (constructively and destructively) mono-causal: It’s all about institutions and not about geography. And the mechanism was colonisation and specifically about “Those who Leave and Those Who Stay” to draw upon Elena Ferrante. The critical difference between good and bad development outcomes was whether the colonisers stayed to build good institutions or always came with the intention of leaving after rapaciously exploiting the colonies’ natural resources, leaving behind awful and damaging institutions. The mortality of settlers (not overall mortality) was their clever-conceit-cum-imaginative-data to tease out in some causal fashion where they stayed and where they did not.

COCD in particular and the broader AJR oeuvre are controversial. A non-exhaustive list of critiques would be something as follows. AJR’s idea that institutions matter is as at least as old as Douglas North if not Adam Smith; their insight that colonisation shaped the evolution of institutions is neither novel nor historically textured nor even accurate; their strategy of teasing causation out of the natural experiment of colonisation is debatable because it cannot distinguish between the places that colonisers went to and the human capital they brought along with them; the data for their key  variable, namely “settler mortality,” is flawed and selectively chosen; and finally, for some of these reasons, the empirical findings in COCD are shaky and non-robust. And there is some truth in each of them.

So why is COCD, on which I want to focus — and which certainly changed my professional life — Nobel-worthy? 

First, this paper has had a massive influence, and on several disciplines, including economic development, political science, economic history and institutional studies. This is reflected not just in the 19,000 citations on Google Scholar (among the highest for economics) but also in the number of economists and papers that have flattered through imitation the approach and sensibility of CO. Nearly 25 years on, development is full of PhD theses even today arguing for the persistent effects of some historical event or intervention.

But is influence the only measure of Nobel-worthiness? How does it hold up against some more intrinsic standard — typically novelty of idea or methodology — of “scientific” achievement? I would argue that AJR is a package of contributions, not all without flaws and individually not necessarily totally novel but together amounting to something Big.

The package is really a coming together of the use of history in economics, its quantification and methodological application, and the imaginative specific idea/data (“settler mortality”) all deployed in the service of testing or advancing an important idea relating to long-run economic development.

It advances significantly our understanding of Big Development (with a capital B and capital D) and does so with parsimony and simplicity that should be the hallmark of any social science. As a permanent bonus, economics — perhaps also other disciplines — has become richly, even if somewhat reductively, inter-disciplinary. Compare that achievement with the “development” (small d) that is all the rage in the Academy today: With some exaggeration, it is a nano-ized vision that purports to incontrovertibly demonstrate that in some village somewhere in the world some households were helped by some small magnitude because of some micro-intervention.  

It is natural with the benefit of hindsight to quibble with the methodology and data of AJR. But in real time, as I have discussed earlier, COCD was daring in using the cross-country approach to answer the biggest development and historical question. Allegations about the non-robustness of CO’s findings reveal a certain narrowness of the critics who are perhaps unable or unwilling to sift the plausible, deeper truth from the possibly shaky results. Macro-development as a discipline was no longer the same after the AJR work.

My one reservation about the award relates more to the body of work of Acemoglu and Robinson on democracy and political institutions. In the contributions of AJR that I have singled out, the emphasis is on the chain of causation runs from economic institutions to development outcomes. But Acemoglu and Robinson (for example, their best-selling book, Why Nations Fail) take that chain of causation one additional step back in privileging political institutions as the deeper determinant of economic institutions and hence of long-run development. 

But, as I argued in these pages more than a decade ago, that step is problematic not least because it is unable to explain or account for the development experience of the two largest countries in the world, comprising nearly 40 per cent of humanity: China and India (see Figure). China has done too well economically given its political institutions and India has not done economically well enough given its political institutions.

A related, and perhaps biased, gripe is of the Acemoglu and Robinson book on “Democracy and Development”, which in its hundreds of pages does not have a single word to say about the most unique and successful democratic experiment in the last 100 years, namely that of India. Cavilling aside, there is little doubt that our understanding of economic development over the last 700-1,000 years will have been immeasurably less rich without the AJR oeuvre. Even if specific aspects of that are interred, human knowledge will have advanced from that funeral. As economic theorists would say, all models are wrong but a select few are profoundly illuminating. That is what makes the  AJR Nobel Prize  richly deserved.

Arvind Subramanian is Senior Fellow, Peterson Institute for International Economics. Full disclosure: Simon Johnson is Subramanian’s good friend and co-author, and readers should accordingly feel free to apply their discount factor to this piece. The comments of Shoumitro Chatterjee, Rohit Lamba, Dev Patel and Justin Sandefur considerably improved this piece