With the rise of development economics has also come a rise in a microeconomic focus towards development. Whereas tradition institutions such as the World Bank and IMF have been hotly contested for their macroeconomic policies – especially in regards to the 80s and 90s structural adjustment period (SAP) – there is much more consensus for microeconomic policies. Macroeconomic policy is much greyer, and is not helped by the fact that the structural adjustment period was on average marked by negative economic growth for Sub-Saharan Africa. For instance, one big question is whether it is ethically correct to place conditions on loans to developing countries. There are arguments on both sides. Others could argue that the incentive structures within the Bretton Woods Institutions are set up to encourage movement of money, rather than spending money well. From here we could slip into the messy crossfire of the big debate between Sachs and Easterly as to whether there even is a place for foreign aid in international development. And we have not even touched on the influence of political institutions on the implementation of macro policy. With the million different variables involved in the theory and implementation of macroeconomic development, it is no wonder that many economists have turned to a micro approach towards development. Today, institutions such as Jameel Poverty Action Lab (J-PAL) at MIT, Innovations for Poverty Action (IPA), Center of Evaluation for Global Action (CEGA), and the Institute for Financial Management and Research (IFMR) all focus on various micro level development initiatives.
Because micro level development initiatives are much more easily testable (can often account for OVB and endogeneity), there has been a surge in interest for impact evaluations. The idea is that development initiatives should only be undertaken (with some exceptions) when there is rigorous research to support the claim that it is an effective and useful development initiative. Indeed, much of the development economics community, regardless of their particular beliefs, are united by a common rally against evidence-free development. This sounds fair enough, and until recently I too believed in only supporting well-evaluated initiatives. However, while discussing product strategy with a staff member of an appropriate technology company based out of Bangalore, I realized that my approach towards development might have been naïve. The employee stated that while it was important to the company to build a safe and effective product, the extent to which there would be product impact evaluation is primarily based upon the desires of various stakeholders such as hospitals and governments. He said that his company was confident (for various reasons I won’t go into here) that the product works. While it would be possible to run a number of impact evaluations to figure out exactly how effective the product was and maybe even make specific calculations regarding the returns to investing in the product, he saw this as overkill. If we know that a product is good, then why spend the enormous amount of time and money to conduct rigorous evaluations just so we can quantifiably measure the effectiveness of the product? Essentially, what is the opportunity cost of conducting rigorous evaluations? This is the question that I have been trying to grapple with.
Let’s take an example. My development economics professor at Cal, Dr. Ted Miguel, is has a famous deworming project. He was able to conduct unique evaluations to measure the effectiveness of deworming on children’s health and educational outcomes. Additionally, he was able to make some estimation as to the returns on investment that governments would receive for spending money on deworming tablets. While this is all very important and certainly fascinating work, was such analysis really necessary? If you ask any layman whether deworming tablets would be useful in settings where worms are rampant, the common sense answer is yes such tablets would be useful. Thus, to what extent is it necessary to quantify the usefulness when even an ordinary person could see that deworming tablets are important and what is the opportunity cost of doing such rigorous evaluation?
After discussing with several economists, the most common rationale I get for quantifiable results is based on the concept that resources are scarce. We only have a limited amount of time and money. Therefore, if a government only has enough money to enact one policy, should it be to provide deworming tablets or classroom textbooks? Such rigorous evaluations help answer this question.
Maybe through evaluation we are able to figure out that deworming tablets are better than textbooks, but what about scholarship programs, anti-retroviral, and hygiene campaigns? These are three more initiatives that we did not have the chance to evaluate. In this case, do we say that we cannot go ahead and roll out a program supporting deworming until we have also evaluated all of these other initiatives and definitively proven that deworming is the best use of resources? At the limit, this process of evaluation would go on forever because there is no end to the number of initiatives we could suggest. And at the limit, the resources it would take to conduct such evaluations would be infinite and no actual policy would ever come out of initiative debate. Perhaps, at some point we should just accept that what we are doing is good enough.
This seems to be the logic of the appropriate technology company I discussed earlier, and to me the logic seems pretty sound. Maybe it’s not always necessary to make sure that there is a ton of evidence. Maybe sometimes you just have to accept that while you have compelling evidence, it would not be enough for a publication, and that’s okay! And perhaps it’s these kinds of risks that move us from theory to action: the point where we actually get something done.
Tags: No Comments.