I had the opportunity this evening to hear Esther Duflo, MIT development economics professor and the founder of the Poverty Action Lab, speak tonight in London. It was great to put a face to the words that I have been reading in her recent book, "Poor Economics." Here is a rough transcript of her speech tonite:
Big world events seem to have more to do with politics than with policies; in the West, enough Aid was given to Egypt every year to give every child in Africa $20; the aid in Egypt was not about reducing poverty; in the South, we need to look at questions like what encourages mothers to immunize their children in the DRC? This comes down to politics, to the governments in disarray.
A growing number of academics are saying that we are wasting time in designing policies, as we are missing the elephant in the room, what determines whether policies will be implemented or not, is politics;
Institutionalist View: the main question of development is to figure out how to solve the political process; from a good political process will emerge good policies; without the political process, policy will not be implemented.
INSTITUTIONS are the main driver of success in a country (Acemoglu-Robinson); unfortunately, they are also very hard to change; they have a large shadow of history; in the areas in West Africa where the colonial powers spent more on education, there is still better education today; in India, places where more egalitarian tax collection are still doing better today.
Does this mean that we are stuck where we are? Institutional change cannot be easily engineered or imposed from the outside (see: Collier vs. Easterly-we cannot import freedom by force, this is an impossibility); Thus, can anything be done? Institutions, and hence history, clearly matter and define the broad constraints; To what extent is there slack for better policy? Is there a chance of improving policies can improve politics?
Progress with bad Institutions
One rarely sees wholesale institutional change, and these are very hard to predict or provoke; but incremental democratic changes do happen at the margin, even with fairly autocratic regimes, such as Indonesia, Brazil, China, Mexico; looking at fixing corruption at the margins is also seen as fruitless; however, this is not always the case; in Indonesia, the threat of audits on road construction projects was enough to decrease the theft level from 27 cents to 18 cents on the dollar; this shows that something can be done, that change at the margins is possible in imperfect systems
Even good institutions are not a guarantee for a good functioning of the institutions;
Many people believe that even democracy is bound to fail in many African countries because of the importance of ethnic voting; in Benin, a study was randomized in which the voters were given different messages in different areas of the country, with different degrees of ethnic appeal and clientism vs: messages of nationalism and national peace; the result was that the client/ethnic message won hands down in the nation, by over 20%. In Uttar Pradesh, Banarjee ran a randomized trial where an NGO went village to village and told the villagers to not vote on caste, but on the issues, and ethnic voting went down from 25% to 18%; these trials show that voters may simply not know enough to vote for competence providing that information matters.
Ideology, Ignorance, and Inertia
There can be good politics in even in bad environments; see: education in Suharto’s Indonesia; moreover, there is plenty of bad policies within generally good policy environment; corruption and inefficiencies are more likely to be due to the lack of understanding and attention to details then to a conspiracy against the poor; Ideology, Ignorance, Inertia (the three I problem)
Example of the Three I’s: school committers in Uttar Pradesh (poor should be involved in the public service)-this was a product of World Bank consensus; thus, the Indian government decided that every village would have a village education committee; after surveying, only 8% even knew these existed, 2% knew what they were supposed to do, and 25% of the members didn’t know they were members. Efforts to re-invigorate these has proved fruitless. This defunct scheme is a pure 3i example. Its not that people don’t want to do something , its just that the system was defunct-the implementation was completely disastrous.
From Good Policies to Good Politics
Voters adjust their views based on what they see on the ground; we can look at the attitude towards women in India; there has been a quota system in the village level Panchayat elections-women tend to be at least as good leaders as men. Randomized trials were held, in which respondents listened to speeches by either male or female; the villagers were then asked questions about the same speech; in the villages where there had been quotas for women, they fared much better in the results than in villages where there had been no quotas; the quotas showed to reduce prejudice in these villagers and respondents, and added to the chances of women being elected in the future.
Another example is in Benin; there was an expert conference held in Benin on the problems on the country, and the experts came up with policies and platforms based on it. In some villagers, the traditional method of campaigning was replaced by town meetings based on these concrete policies that had been formulated. In the villagers where there had been the meetings, the experimental candidate got a much higher level percent of the vote than in the control villages. This shows if you want good politics you must have good policies, to give people something to discuss in making their decisions.
Conclusion
Many in the West are pessimistic about the institutions of the developing world. Political constraints exist-and hard to predict political events have important impacts; however, there is a lot of scope for better policies, perhaps particularly in regimes that are not completely locked or at war; there are a lot of very bad politics for no good reason.
Nothing will come out of just one RCT; but at the same time, it is ridiculous to say that these experiments do not matter when used to establish patterns, to trace a story about what is going on in a particular domain. What matters at the end of the day is not the multilateral organizations, but the policies of the countries themselves-this is where most of the money is, not in the aid dollars, which are quite marginal. Once we recognize that aid money is marginal, this can liberate, and create the understanding of leverage for the money in the most effective way. Thus, some of the money should be used to find out what works best, through trials and experimentation. There is room for this with development policies. This will maintain relevance in a way that is more useful than today in the development world. Lack of change is more due to inertia than to resistance. Thus, change is often identifying opportunities.
