On Mandarins and Bandits

26 October 2026 · on substack

Mike McGovern’s article “Popular Development Economics—An Anthropologist among the Mandarins”¹ was a striking and important read, evoking doubt in my pursuit of the field of Political Economy, combined with a sense of urgency to protect what my original intentions.

The article is a response to two books by Paul Collier, The Bottom Billion: Why the poorest countries are failing, and what can be done about it and Wars, Guns, and Votes: Democracy in dangerous places. Collier argues that development parameters for the poorest billion in population have stagnated over the last 40-50 years², and that these are caused by four traps. Namely, these are - “conflict, reliance on natural resources, being landlocked with bad neighbours, and bad governance”. Collier poses some policy solutions such as more restrictive development aid, more peacekeeping missions, revised international laws and trade policy favoring the poorer nations. His most extreme solution is an agreement between a rich country and a poor country’s government, wherein the poorer country agrees to following democratic practices, and the richer country uses its resources to keep at bay bad actors and coup d’états. If the poorer country fails along its democratic axes such as by rigging elections, the richer country would turn a blind eye to any attempts by other groups to take power, henceforth binding them accordingly.

McGovern proceeds to then critique Collier’s work (not necessarily of his findings, which he points out that he somewhat agrees with in some cases), followed by critiquing the field of economics, and some quite important factual inaccuracies in Collier’s description of events in Côte d’Ivoire.

I found myself agreeing with a lot of McGovern’s critiques, especially the necessity of ethnographic nuance. He states it as requirement to be able to meaningfully engage with a peoples that you have not really been around long enough to stop your impressions of them being mere projections of your worldview on to them.

As a researcher from India, I believe this puts into words why I want to quite strongly (almost solely) engage with India within my research, as I believe I have some “true” impressions of life and the people there, having lived there for twenty-five years. As someone currently living in Chicago, it also brings to light why I feel quite protective of research on India that arrives at premises and assumptions that are clearly under-informed.

My organisation and colleagues in India, had the goal of creating research products and knowledge that was of the Global South, for the Global South, and this article highlights why something like that might even be necessary, as a lot of research is done from the outside looking in. I want to say, we were not immune to this, having certainly done a lot of research on peoples in India for whom we were said outsiders looking in.

As someone working in Chicago, it also makes me cautious of how under-informed I am, given my short history with the city. As an example, I work a fair bit on property taxation in and around Chicago, and I believe it is impossible to do a good job of it, without being informed of the history of red-lining and segregation, that deeply colours any and all present realities of the city today.

I am fortunate enough to work with a team that helps open these blind spots, but the large question that this raises is, how much is enough?

How much time, and/or how much understanding must one have of a place and a people to be able to meaningfully study it? After all, are we not ostensibly trained in the methods of economics and political science to give us to tools to study any of these, provided we use them well? And could these well-intentioned guidelines not raise barriers, and make the world a harder place to understand?

Two of McGovern’s final sentences in the article resonate with me. The first as an approach to good research -

... first identifying perplexing correlations and then patiently teasing out explanations of the causal dynamics involved.

And the following as somewhat of a caution, but also as my strongest takeaway from 8 weeks of this course so far.

In the move to explain causality, the common sense that allows us to decipher correlations in our own societies may betray us if we attempt to apply it where people have a very different sense of what is common.

My reflections of this paper are inextricably linked with my personal feelings about the field of development economics, and political economics, and my place within it. While I do think the work we do (and strive to do), remains important, this article to me underscores the criticality of ethnographic nuance and doing research “around you”, not “outside in”.

Raúl Sánchez De La Sierra’s paper on Stationary Bandits3 is in my opinion, an example of this work. It is a stellar piece of research, asking questions that are only possible due to a “true” understanding of the environment. It furthers places those questions quite well in the broad economic research. It is quite interesting to see such dichotomous results of state formation depending on whether it is easier to tax extraction/production of goods, or tax the consumption. While I still think there are some leaps in logic from armed bandits to new state formation, and the identification can be a little weak, I still think it is a better example of economic research than most other papers where I read them and go “that’s great, so what?”. This paper is a result of trying to ask important, big questions, and is only possible due to said ethnographic nuance that comes from taking time and understanding your surroundings.

To end, McGovern’s critique of Collier was quite rooted in pointing out that Collier did not know what he did not know, and as much as I think the Stationary Bandits paper makes a lot of sense, I still do not know what I don’t know, and I am unsure of how to even grapple with that. Could I in the future even be a referee for a paper like this? And if I am, should my pushback be to include more ethnographic understanding and grounding to the paper?