Last updated on January 3, 2025
(Over the last few years, I have explained to a number of colleagues and graduate students that I was not interested in doing development anymore. I am writing this post to avoid having the same conversation over and over, probably against all hope since no one reads anything anymore anyway. Other mid- to late-career folks might find something useful in this post, having perhaps witnessed the same changes I have over the last 25 years.)
“The mediocre teacher tells; the good teacher explains; the superior teacher demonstrates; the great teacher inspires,” said William Arthur Ward.
So it was that in the late 1990s, after switching majors from philosophy to economics during my undergraduate years at the Université de Montréal, I took a development economics class with Jean-Louis Arcand, and that class laid the cornerstone of my research interests and inspired me to go to graduate school to study development economics.
Back then, development economics was a field of economics whose focus was the study of how multiple market failures led to persistent poverty and chronic underdevelopment. It was also a field from which economics had learned a lot, and could (and would) learn some more. In Stiglitz’s words in his introductory chapter to Bardhan’s (1989) Economic Theory of Agrarian Institutions:
A study of least developed countries is to economics what the study of pathology is to medicine: by understanding what happens when things do not work well, we gain insight into how they work when they do function as designed. The difference is that in economics, pathology is the rule: less than a quarter of mankind lives in the developed economies.
The 1999 textbook by Bardhan and Udry is probably the definitive statement of the the idea that development was about how market failures enable persistent poverty, if not explicitly at least implicitly.
After a few decades during which development had been seen as a fringe field of economics—one seen as being of lesser worth than other fields of economics—this was a welcome development.
But sometime around the mid-2000s, something happened that somehow turned development from a field of economics studying how multiple market failures lead to persistent poverty into an area of research (i) using empirical methods aimed at causal inference with (ii) data from low- and middle-income countries (LMICs). For a time, development economics seemingly became almost exclusively about impact evaluation.
Then in 2019, the Nobel prize for economics was awarded for a “new experiment-based approach [which] transformed development economics.”
With that kind of recognition stemming from a change in methodology, and in what felt at the time like an obsession with getting on the poverty-reduction bandwagon before there were no longer any poor people left to help, many graduate students in economics and related disciplines decided that they, too, wanted to do development economics.1
But between the late 1990s and the late 2010s, something else happened that changed the fabric of development economics. In a keynote address I gave a few years ago, I wrote:
In fact, it can be argued that, since the mid-2000s, development economists began answering all sorts of questions from other fields (e.g., agricultural, health, or labor economics) using LMIC data and that, very often, those questions do not explicitly involve market failures or tradeoffs.
In other words, development economics went from a field with a clear focus (i.e., how multiple market failures constrain economic development and lead to persistent poverty) to a field about anything and everything… as long as it’s in LMICs, and preferably with a specific focus on causal inference. (Even then, it isn’t entirely clear that it’s always about LMICs; more on this later).
But to paraphrase GK Chesterton, when a field of inquiry becomes about anything and everything, the problem is not that it’s about anything and everything, the problem is that it really becomes about nothing.
One example: a paper on the effects of chess training (randomized, naturally, and in an LMIC, obviously) on academic and non-cognitive outcomes.
Another example: a paper on the effects of sending text messages to parents of preschool-aged kids (at random, obviously, and in a country whose GDP per capita is somewhere between that of Bulgaria and Romania, which I guess makes in an LMIC if you squint?) about how to get their kids to do educational activities at home on those kids’ cognitive skills.
Both papers were published in what is widely agreed to be the top journal in development economics.
We live in an era where Americans people will adopt the most literal and obtuse reading of anything written, so I (should not) have to (but must) clarify two things.
First, I am not saying the topics studied in the two papers just mentioned are not important. I’d be a hypocrite if I did: my wife and I (i) enrolled our child in an afterschool chess program a few years ago, and (ii) have our child engage in many educational activities outside of school. It’s just that I do not see how those two topics qualify as “development economics.”2
Second, I am not faulting the authors of the two papers just mentioned. Everyone is doing what’s best for themselves and publish in the best journal possible. Those authors are responding to incentives, and thus publishing their work in the best outlet that will have it.
Third, I too have at times played the game.
Until I chose not to. A few years ago I realized I no longer recognized development economics, i.e., one of the two fields I fell in love with and decided to go to grad school for as a twentysomething.3
This did not occur overnight. These past few years, after nearly two decades of being told that I was not really a development economist to begin with anyway,4 5 I realized that my training in development economics was not a life sentence, and I could do something else—something else more rewarding to me: a mix of applied econometrics and agricultural economics and policy.
This is not to say that I will never touch development topics or submit to development economics journals ever again. If development is about anything and everything, then it stands to reason that just about any applied work I do will end up somehow being about development, right? Less facetiously, some of the things I am interested in still overlap with the “old” definition of development economics. Many of the institutions arising in agricultural value chains, for instance, are a distinct response to market failures more likely to be found in rural areas of LMICs than in the Upper Midwest.
What this is about instead, more than me no longer recognizing what I came for, is about how the field has changed over the past few decades, and how other mid-career development economists might find themselves in the same predicament, choosing instead to solve for their own x in “development-and-x.”
In a previous post, I promised a return to long-form writing in 2025. I am hoping to have one post every week. In my next post, I will discuss whether, in light of what I discuss in this post, there is still a place for development economics in agricultural economics departments.
- From casual empiricism in my role as one of the graduate placement coordinator for our department, this also explains why the development job-market is saturated right now, with even the best students struggling to get academic jobs. Given how long it takes to do a PhD, this is a problem that will probably get worse before it gets better. ↩︎
- This issue arises for three reasons. First, it sometimes isn’t clear that something is “development” (say, because it is studied somewhere like Chile, China, or Mexico, all countries that strike me as no longer being “developing”; heck, from the point of view of infrastructure development, China is arguably more developed than the US). Second, it sometimes isn’t clear that something is “economics” (say, because there is no explicit relative price or market involved, and no tradeoff beyond the hackneyed “Well, there isn’t a lot of resources for development policy and they could be spending those resources on some other policy!” Yawn.) Third, it sometimes just isn’t clear that something is “development economics” (because it isn’t about how market failures constrain development). ↩︎
- The other field I fell in love with as a twentysomething was econometrics, but I did not go to grad school for it, and Cornell was certainly not the place for it: Almost all of the econometrics I know I have learned either as a Master’s student at the Université de Montréal or after getting my PhD. ↩︎
- If you have been reading this post and thinking “sour grapes,” I say: “Totally. But there’s more to my opinion than just sour grapes, if you care to see it.” ↩︎
- Inevitably, someone whose education comes from the Internet more than it does from books is bound to tell me that I am “gatekeeping.” If that is you, please re-read the sentence to which this footnote is appended. ↩︎