Syndicate content

How is Development Economics Taught in Developing Countries? What we learned from looking at more than 200 courses

David McKenzie's picture

Also available in: Français | العربية

Earlier this year I blogged about a research project I was beginning with Anna Luisa Paffhausen, that had the aim of seeing how development economics is taught in developing countries. Thanks in part to the help of our readers, we were able to collect a combination of syllabi, surveys, and exams from 145 undergraduate courses in 54 developing countries, and 81 masters courses in 34 developing countries. We then benchmarked these against undergraduate development classes taught in the U.S. at top-20 and non-top-20 programs. We now have a preliminary draft paper which I thought I’d share and see if our readers have comments. I summarize some of the key findings and recommendations below:

What is taught as development economics?
  • Not a very standardized view of what is crucial in development: only four of the 22 topics we looked at are taught in at least half of all undergraduate courses in developing countries: growth theory, poverty and inequality, human capital, and institutions. Courses in top U.S. departments are much more likely to cover credit markets, risk and insurance, land markets, data analysis, and experimental methods than courses in developing countries, while the latter are more likely to cover dependency theory, the environment, and macroeconomic management.
  • Developing country courses are more dissimilar to one another than courses in the U.S.: Using similarity indices we show that both undergraduate and masters development economics courses in developing countries are less similar to each other than are undergraduate courses in the U.S. Moreover, the average developing country undergraduate course only overlaps with 23 percent of the course content with courses offered in top U.S. schools.
  • Several important topics are rarely taught at all: these include entrepreneurship, firm growth, and international migration.
How is development economics taught?
  • The data revolution has not yet come to development courses taught in developing countries:
    • In topic coverage: Developing country courses focus much more on concepts and models and less on empirical findings than U.S. courses.
    • In assessment: developing country classes largely rely on written exams, essays, and presentations, and very rarely require students to use statistical software or answer
    • Reflected in textbook choices: developing countries and non-top U.S. courses are more likely to be based around the textbook of Todaro and Smith (2014), first published in 1993 and now in its twelfth edition, and which relies heavily on case studies and stylized facts, while top programs are more likely to use the empirically-focused book of Banerjee and Duflo (2011), and the theory-based book of Ray (1998).
What explains these differences?
We measure the similarity between each developing country course and courses at top-20 U.S. schools and see what country and instructor characteristics are correlated with this degree of similarity. We find courses in poorer countries are less like those in top U.S. schools (see Figure below), as are courses in countries with a higher share of government involvement in the economy and countries with lower overall educational attainment. Instructors who are actively involved in research are more likely to have their courses closer to the frontier.

What could be done better?
The field of development economics has changed dramatically over the past twenty years. This is reflected in changes in the topics that command most research attention, and particularly in the rapid growth in data availability and empirical analysis. Our survey of how development economics is being taught in developing countries suggests that many classes have not kept pace with this change, and are not meeting key student learning goals of teaching students to be critical users and analyzers of data to answer economic questions. This is important since the next generation of policymakers responsible for implementing key development policies are likely to have their views of what policies they should pursue heavily influenced by what they have been taught.
We have several suggestions for how instructors can improve the teaching of development economics in developing countries (as well as in a number of developed country schools), but this is also where we would love reader feedback of good/innovative ways the subject is being taught:
  1. Take advantage of the availability of macro and micro data to have students do basic data analysis for their home countries. This could involve calculating poverty rates, plotting Lorenz curves, or estimating simple regressions that do not require complicated statistical knowledge.
  2. Have students work in teams to collect and analyze basic data about their home country. This could include, for example, visiting local marketplaces to attempt to understand how small shops compete with one another, collecting wage data from spot labor markets during different weeks of the class to see how stable wage labor is, or visiting firms to learn how they hedge against exchange rate risk (the latter was one of my most memorable assignments from my coursework in New Zealand, we each had to interview 5 firms about how they hedge, and then come back and share the experiences). More ambitiously, one could develop small panel surveys by having each new intake of classes return to the same settings year after year in order to build up rich knowledge of how the local economy works. Does anyone have any examples of undergraduate (or masters) classes in developing countries which actually do this?
  3. Don’t just rely on textbooks: Assign students readings from current newspapers, blogs, or magazines in order that they can apply what they are learning to current economic debate in their country.
  4. Consider ways to better share what is taught across instructors. Our analysis found very few syllabi were on the web (One noteworthy exception was a repository of Brazilian development economics syllabi collected by the Brazilian Network of Development Economics Education). Another potential role for regional associations here is continuing education, teaching instructors how to better incorporate a more empirical approach in their classrooms.
We would love to hear from any instructors or students who have comments or ideas on this. Thanks.


Submitted by Ambrish on

This is quite interesting. Can you check whether faculty background matters. For example, a faculty who has obtained ph.d. from US/UK is more likely to introduce similar approach, techniques when he/ she is teaching.

This was something we hoped to do, but we weren't able to collect this information for enough faculty - many of the faculty didn't have webpages or the place of their PhD listed, and although we tried to administer a survey, response rates were low.

Submitted by Justin Sandefur on

Seems very non-ideological. Oddly, I guess I mean that as a criticism.

The contrast between anecdotes and narrative versus theory and rigorous empirics is worth documenting. But I clicked on the post expecting to read about how developing-country syllabi are much more concerned with the detrimental effects of rich-country policies, dependency theory, the horrors of structural adjustment and WB/IMF policies of the 80s/90s, etc. (In contrast to rich-world development economics which often implicitly takes the viewpoint of a benevolent aid bureaucrat as social planner.) But I didn't see much of that here.

Perhaps developing country instructors eschew newer textbooks using 'modern' methods because those texts neglect the questions they care about? Not sure I really believe that, just speculating.

It looks like the paper touches on all this a bit more than a short blog post can, but I did feel like the last 3 points shifted pretty abruptly from positive to normative commentary. If the courses differ in content not just methods, I'd be reluctant to take it as axiomatic that rich-country syllabi are the ideal to emulate.

Overall, an extremely creative and cool project, per usual. Thanks for sharing.

Thanks Justin. While we did come across a bit of dependency theory, one of the perhaps surprising things was how non-ideological the content seems to be. Perhaps economics courses are the wrong place to be looking for existential debate here, but really the thing that struck me was how boring these classes are in a lot of countries: week 1, read textbook chapter 1 and learn what a poverty index is, week 2, read textbook chapter 2 and learn what the idea behind the Solow growth model is, etc. This is where the lack of actually looking at data and trying to apply the ideas to see what is going on in one's own country strikes me as a missed opportunity.

Submitted by Justin Sandefur on

That made me laugh. Shame if my image of developing world professors resisting the NEUDC version of development economics so they can teach Hirschman and Prebisch is actually just underpaid lecturers being lazy about updating the syllabus.

Submitted by angus on

All professors, in rich lands and in poor lands should come together to resist the NEUDC version of development economics!!

Submitted by Mike Toman on

VERY interesting comparisons. I hope one byproduct of the paper and blog is more attention to the environment in the course content in developed countries.

Submitted by Manuel Fernandez on

Why speculate about motivations by comparing syllabi, when you can easily interview 5 professors about how they teach Development Economics?

Submitted by Asif Dowla on

I was wondering if you noticed more emphasis on agriculture in developing country courses. I had to take a separate course on agricultural economics as an undergraduate. My economic theory and statistics professor was Muhammad Yunus. It is hard to believe! He used to teach heavy dose of utility theory. He was freshly minted Ph.D from Vanderbilt.

Submitted by Lee Crawfurd on

Seems like there is an opportunity here for someone to develop and support some kind of network along the lines of or but focused on supporting teaching at universities, perhaps with space for collection / curation / discussion of existing open source course materials (such as and and I call it lecturr

Submitted by Max Connerie on

I find it fascinating that you seem to equate 'the top US schools' with best practice. Couldn't it be that course designers in some developing countries choose to cover the topics that seem most applicable there?

Maybe the kind of economics practiced in the US needs to change to fit the needs of poor countries? Das et al. show in "U.S. and them: The Geography of Academic Research" (J. of Dev. Econ. Nov. 2013) that of 76,000 papers published between 1985 and 2005 more more papers were written about the US than on Europe, Asia, Latin America, the Middle East and Africa combined.

I think this is a more valid point for the question of WHAT is taught than HOW it is taught, arguing that being able to be understand how data can help distinguish between competing ideas should be a critical part of a development economics education in every country. In terms of what is taught, as noted there are key topics that seem undertaught in both the US and in developing countries

Submitted by Martha on

Great ideas! I don't teach devt econ, but your ideas can definitely be applied to other courses as well.

One thing you might want to consider in your research is the age of the instructors. In some countries (especially Asian countries), the foundational econ classes (intro micro, intro macro, devt econ, etc) are assigned to senior members of the faculty, and the more applied classes (econometrics, statistics, advance micro, etc) to the younger. I think this influences a lot the teaching styles, textbooks used, etc.

Submitted by Djamester Simarmata on

Smart comments. I was one of the users of Ray's textbook, but before I used also Todaro's textbook. One of the interesting things from Ray is his treatment of rural credit market, which happened in either rural areas as well in urban areas in Indonesia. This urged me to say to myself that the book is appropriate for Indonesian situasi. But another issue, since long I was dissatisfied with the assumption of constant return to scale in the macroeconomic context as well as in developmental economics. Due to that I have written an article on this issue, in conjunction with the issue of the decreasing return to scale for small farmers. This was originated from Sen, but previously it was from a Russian agronomist, and it was taken by Ray in his book. I am a son of a farmer, so I felt it was reflecting the reality of the farmers in Indonesia. I agree with most of your critics on the way DE is taught in DC. Thanks

Submitted by Ron on

Axel Leijonhufvud's 1973 satirical ethnographic "study" of the "Econ" tribe comes to mind -

"The priestly caste (the Math-Econ), for example, is a higher "field" than either Micro or Macro, while the Develops just as definitely rank lower... . The low rank of the Develops is due to the fact that this caste, in recent times, has not strictly enforced the taboos against association with the Polscis, Sociogs and other tribes. Other Econ look upon this with consider- able apprehension as endangering the moral fiber of the tribe and suspect the Develops even of relinquishing modl-making."

Of course Leijonhufvud ends on a somewhat depressing note with the Econ tribe being progressively deprived of their sense of history. This is arguably the problem today. History of economic thought pretty much MIA in most US departments.

Submitted by Manoel on

Thanks for doing this, very interesting and relevant stuff.
I have been teaching growth (and development) for third-years in South Africa since 2007, and I use the textbook by Jones and Vollrath (with Easterly's Elusive at the background).
The students are fairly positive about the book, mostly because it is a nice linear way of introducing the canonical models. I think making sure that undergraduates know the canonical models is very important in developing countries, some of these kids will land top jobs in government all over Africa, and knowing the basics is paramount so that the usual mistakes can be avoided (as a kid I had to go through the Brazilian hyper that could have been avoided if our economists at the time knew the basics). The students also enjoy using the PWT to plot data and get some correlations and even regressions.
Oh, yes, I was educated at Warwick and Bristol, and I try to follow what I learned there, without trying to reinvent the wheel. Are there colleagues trying to reinvent the wheel in sSA? Plenty!

Submitted by Tara Prasad Bhusal on

Thanks for sharing the survey report. It's great job and the result is as expected.It shows the existing situation and problems of developing countries regarding the teaching of development economics in general.The only problem is that all of us know the problems but we don't pay attentions to get rid of the problems.It will be more valuable if we can suggest some specific text books, methods, materials, manuals(if any). If not, our effort should be devoted to prepare those. I hope, the World Bank will take the initiation to prepare required text books, manuals, methods, and more, based on the specific conditions of developing countries to prepare skilled , civilized and cultured citizens and make the world a place to live in harmony and peace.

Thanks. This is where I think peer-learning would also be very useful - it would be great if instructors in developing countries could better share some of this content among themselves - as well as if the writers of textbooks could pay more attention to accompanying data exercises.

Submitted by Bucekuderhwa Celestin on

Very interesting and relevant result. I began to share and initiate student on empirical findings. However, the big task of our student is how to access statistical software or answer.
If the world bank could help instructors and student in developing countries, it can be a good new for them.

I am not sure if the issue is a hardware or software access problem. So long as students have access to computers, there are a variety of different ways they can do data analysis even without access to Stata/SPSS/SAS. These include:
1. R, which is a well-known and free software package for statistics.

2. Using the built-in basic regression package in Excel:

or in the free Google docs package:
or if you search, YouTube has a bunch of videos showing how to do regressions with Google Docs.


Submitted by Henry on

Great blog post. Out of interest, where the course guides/syllabus difficult to get ahold of? It always strikes me as odd they aren't more easily accessible. And a follow up, were they provided on a condition of anonymity or are you able to stash them somewhere for others to have a look at?

Cheers, Henry

Submitted by Aditya Bhattacharjea on

I recently came across the World Bank Working Paper and read it with great interest. I have some comments on your approach:

1. You conflate undergraduate and master level development courses in developing countries and compare them to US undergrad courses. Your finding that there is little difference in topic coverage between undergrad and master’s courses in developing countries probably conceals significant differences in the level of sophistication. At least in the best Indian and British departments that I am familiar with, the standard of the master’s program is higher than that of a U.S. undergraduate major. The relevant comparison should be only between undergrad programs.
2. Even undergrad courses can be taught at very different levels in different universities, as they are even in the U.S. You note that the coverage of courses outside “non top-20” U.S. departments is closer to that in developing countries, but you do not attempt to explain this divergence from the top-20 as you do for the developing-country courses. Obviously, the same explanatory variables would show much less variation within the U.S. (and the fact that, unlike in most countries, American students go to college far from their home towns would make the location variables irrelevant anyway). But there is something to be explained which is unrelated to the ‘backwardness’ of developing countries. You found that data on instructor’s background and research activity were hard to come by for developing countries, but you did not pursue this line of inquiry for the U.S. departments where it would be easily available. In any case, the top-20 U.S. departments by Repec research rankings should not be the benchmark for undergraduate courses. That would be unfair to the faculty of many liberal arts colleges who provide an excellent undergraduate training without being research-oriented.
3. Whether in the US or other countries, surely it matters whether the development course is part of an economics major or minor sequence, and therefore has prerequisite courses in micro, macro, and statistics, if not mathematics, or whether it is a stand-alone course for students majoring in, say, history or literature. This would determine the topic coverage and textbook used. Your arguments for more empirical work, and suggestions regarding free statistical software, are well taken, but the latter category of students may find even simple applications too hard.
4. Relating course material to what students read in the media and in blogs is all very well, but the discussions in the media are usually about macro-economic management, stock markets, environment, international trade and payments issues and institutions, multinationals, and large-scale government spending programs. Such topics, as you notice, are absent or given very limited coverage in the Ray and Banerjee-Duflo books that you seem to regard as representing “best practice” in development teaching, brushing aside the criticisms of Ravallion and Rosenzweig on this count. On such topics, students are more likely to find a resonance with books like Todaro and Smith.

Submitted by Olofin, O.P. on

You have done a great work. I think one of the ways you can help these poor countries is by providing training forum for the teachers involved in this area annually. This is better than giving cash-aid to their countries in a way. These are the people that influence behaviour of uncountable number of future generation. I think two or three weeks summer training in this area is not too much considering the importance of this findings, even if they are going to pay at discount rate. Thanks.

Add new comment