Syndicate content

IO and Development Part 3: Where are some opportunities for work intersecting these areas?

David McKenzie's picture

The first two posts on this topic this week have looked at the gap in the use of IO in development, and some possible reasons why IO tools might not be used as much. Today, the final post in my Q&A with Dan Keniston [DK] and Katja Seim [KS], looks at where there might be low-hanging fruit from better use of methods from IO in development.

Field experiments are sometimes cheaper and easier to pull off in developing countries than in the U.S. – at the very least having many more countries to work in should open up more opportunities. My sense is that there have been very few field experiments in the IO literature. If people are thinking about policies to evaluate or experiments to investigate some theory in IO, where do you see some of the most productive areas where experiments could be employed?
[DK] Many IO questions do not lend themselves to experiments as naturally as fields of economics focusing on households or individuals. For example, designing field experiments to understand regulation, contracting, or collusion by firms would be challenging even in a developing country context. Perhaps the best opportunities to use field experiments are to understand demand. It might be much easier to experimentally manipulate prices or purchases in developing countries, since the total amount of money required to subsidize the shopkeepers or purchase the goods would be much less.
As I mentioned briefly in yesterday’s post, the microfinance RCT literature could be seen as a field experiment in market entry, which would put it into an IO framework.

What are the questions where they are the greatest potential synergies between IO and development?"
[DK] My own answer is still evolving, but I think it would at least include the following points:
 

  • Understanding demand for health products, in particular the high elasticity of demand around 0. This is one of the few topics that development economists can agree is a major question in the field, and also one to which IO and marketing may have a lot to contribute.
  • A sub-topic here is the welfare effect of new goods. This was once an active (and bitterly controversial) topic in IO, and seems all the more relevant in developing countries where new products are getting introduced to new markets very quickly.
  • Understanding the effects of competition on productivity. Productivity dispersion in general is a link between IO and development, and I think the question of why increased competition does not cause increases in productivity is a major outstanding question (from your work and others') that had very strong links to the IO literature.

[KS] A second area that empirical IO economists have devoted significant efforts to – both as an input in investigating questions of competition, but also in its own right – is demand estimation.  These techniques seem very applicable to a host of questions on technology adoption, such as Tarozzi, Mahajan, Yoong, and Blackburn’s work on mosquito nets or Suri’s work on the introduction of hybrid maize.  An advantage to conducting such a study in a developing country context is the more ready ability to combine observational data with experimental evidence to generate additional sources of variation that alleviate (some of) the need of relying on strong model assumptions in the context of estimation.  This is rare in the larger demand estimation literature, presumably in large part because of the cost of primary data collection across a heterogeneous consumer population in developed countries.

Thanks again to both Katja and Dan for their participation in this discussion. This is an area where I hope to see more interesting work, and we would love to hear from people who have either attempted to incorporate more IO into their development work and run against unexpected challenges, as well as see more examples where we learned something we wouldn't otherwise have been able to.

Comments

Submitted by Ron on

Its an interesting parallel between the rise of New Empirical IO (NEIO), the term coined by Bresnahan, to distinguish it from the old IO/ SCP paradigm, and the new Development Economics, especially in the context of effectiveness or "What works". The concerns over mis-identification/ endogeneity associated cross-sectional analysis meant that both needed to deploy the advances from the econometrics to get a better handle on internal validity issues. The difference is that unlike say Development Effectiveness field that has in the main adopted the Rubins-Holland causal effects framework in applied work, the NEIO has not.

The other challenge is that the Health Education Nutrition dominates the results agenda both within development agencies, donor community and development economists. Case in point UNIDO that actually computes and compiles TFP stats has budget of 100 million and losing support of big name donors.

Add new comment