Rohini Pande is Mohammed Kamal Professor of Public Policy at the Harvard Kennedy School, where she also co-directs their Evidence for Policy Design (EPoD) program. She has described her research as examining the economic costs and benefits of informal and formal institutions in the developing world and the role of public policy in changing these.
1. You have worked on a range of different topics – including rural banking and microfinance, governance, environmental regulation, son preference, and housing – but almost exclusively in one country, India. When you think about your broad research agenda, how to you think about the tradeoffs involved in focusing deeply on one country, vs exploring these topics in different places?
Starting with my PhD work on political reservations in India, I have been fascinated by the why and how of public policy in democracies and, in particular, how the political and social context shapes the choice of policy. I have also found that viewing problems of economic development through a political lens that engages with questions of power creates links across questions and topics that might before have seemed disparate.
Once you adopt this perspective, the advantage of focussing on a single country becomes apparent. Over time, one begins to understand how power structures operate and which policy lessons are generalizable and which remain specific to a location. The Indian economist Jean Drèze, who very much inspired my career choice to become a development economist, told me that he has never been to Africa. “Once I got to India,” he said, “there was more than enough for me to do for a lifetime.” His most recent book – Sense and Solidarity – provides a strong rationale for an action-research agenda that is focussed on a single country.
The Indian development trajectory also epitomizes for me a central problem that development economists are now grappling with – neither technocratic state capacity fixes nor ensuring free and fair elections have, in themselves, been sufficient to ensure that economic growth is the tide that lifts all boats. A striking fact of recent economic growth in the developing world is that now the majority of the world’s poor live in a handful of high-poverty, middle-income countries. Focussing on India has helped me gain a perspective on what it would take – beyond technocratic fixes – to create an effective state that works for the poor.
Put differently, I have found that (for me) it’s often a fool’s errand to try to isolate a policy problem from its environment, or understand a single human force assuming that it explains the issue at hand but somehow isn’t closely linked to context and politics.
Saudi Arabia has a huge unexploited demographic dividend, as in the other countries where EPoD works. Officials at the Ministry of Labor and Social Development (MLSD) had come up with a range of policies to draw Saudis – particularly women and youth – into the labour force. They were interested in working with researchers to identify the root problems and then applying rigorous analysis to see what works. Meanwhile, my colleagues at EPoD and I had developed a methodology we call Smart Policy Design and Implementation (SPDI) whereby researchers and policymakers collaborate, first on the essential step of identifying problems both relevant to the policymaker’s needs and feasible for research to help solve, and then on a cycle of steps to develop and continuously refine solutions. From the MLSD’s perspective, our methodology fit their needs. From our perspective, Saudi Arabia represented an unprecedented opportunity to work in a country with administrative databases that rival those in rich countries, but whose demographics and institutions are similar to emerging economies. Plus, we could try out the SPDI methodology in a new context where none of EPoD’s co-directors had ongoing research engagements.
So, over the past few years we have gone through three rounds of requests for proposals and commissioned over a dozen research projects on how to get Saudi people into the labour force in the right way. We would love for more researchers to join us – so please apply in the next call for proposals! Turning to my own work in Saudi, Rema Hanna and I are looking at the country’s unemployment insurance, where over 80% of recipients are women, and testing how best to integrate these women into workforce and what the socioeconomic impacts will be.
3. In your role as an instructor at HKS, you teach a lot of people who have government experience and who will be going back into positions of policy influence in different countries. What do you see as the biggest issues you face in getting them to appreciate the need for rigorous evidence in policy decision-making? Are there a set of questions these students have that development economics currently doesn’t have good answers for? What kind (or sets) of skills do they need to have to make good use of evidence?
For years, Dani Rodrik, Asim Khwaja, and I have co-taught the flagship MPA/ID development course at HKS. We often see students, especially those with experience in government and policymaking, arrive with a solution in hand – digital identification, or universal basic income, or whatever – and our work is to get them to back up and start thinking hard in terms of problem identification. If you can get them to do that – to be question-orientated rather than answer-orientated – then it’s an easy step to get them to think about analytical frameworks and the evidence needed.
But I should point out that people who come through HKS are a pretty select population, and to affect policy at scale, you need to go to where it’s being made. A big part of EPoD’s mission over the last four years has been to build an appreciation for analytical frameworks and a capacity for their use among policymakers in country. Here we see the same problem emerge: solutions dominate the conversation before problems have even been identified. This is often because a particular solution is being pushed by political bosses or superiors in the bureaucracy. But we are committed to engaging in shifting this discourse with in-country partners. EPoD has developed a blended online/classroom training program that is now being ingrained into India and Pakistan’s civil service training academies. This, in combination with pilot projects that show the value of evidence-based policymaking, and events where policy problems are discussed with colleagues and researchers, is beginning to shift the culture.
4. You have been on the Committee on the Status of Women in the Economic Profession (CSWEP). Recently there has been increased attention on some of the ways women are treated in the profession and the constraints faced by women. Do you think there are specific issues that affect women working in development economics that differ from these general issues in other fields of economics?
I remember when Esther Duflo and I were just getting started and we co-wrote a paper on dams and their impact on development. A senior male World Bank economist wrote to our senior male colleagues at MIT and Yale asking that they review our work and correct our mistakes. These days, I don’t get undermined that way within the profession, but still – and this is a problem for female development economists, in general – I will meet with a policymaker in South Asia, and he will be looking over my shoulder wondering where my male superior is. For those of us who are now in positions of seniority in the profession, I think it is essential that we promote young women in both academic and policy settings and not undermine their intellectual authority in any way.
I got involved in CSWEP to look at issues for women who didn’t do their undergraduate education in the US but come here for graduate school or for a job. There has been a lot of discussion on pipeline issues for women in economics PhD programs, but I continue to believe that we don’t focus enough on, first, improving the low numbers of foreign female applicants for PhD programs (relative to their male counterparts) and, second, identifying the specific hurdles for women who did not start (or complete) their education in the US and, as a result, lack access to the networks that their US-educated peers have. We know a lot now about how networks build self-confidence and pave the way to positions and promotions.
I would guess that these issues hit development hard(er) since gender stereotyping is often stronger in developing countries and those educated outside the US are over-represented in development economics relative to other fields (just based on the countries development economics focuses on).
5. It seems recently that there has been a growth in work on gender in development economics. Do you see a trend here? What do you think is causing it? And what are the big questions you think we still need to tackle?
I think it’s a matter of righting a wrong. Researchers are coming to see that this isn’t a matter of feminism or activism, but of accuracy and rigor: research that doesn’t view development through a gender lens makes errors. I’ve seen this many times in my own work. Microfinance is one example: it’s easy to look at microfinance through the lens of profits and poverty-reduction, and say it doesn’t work. But if you frame the question differently, if you say, Have these programs successfully provided credit to poor women – some of the most isolated and disempowered people on earth – and made a profit at the same time? Then you get a very different outcome. It becomes a success story.
A related example is about returns to capital. In work with Arielle Bernhardt, Erica Field and Natalia Rigol we show that the estimated returns to providing female entrepreneurs more liquidity are very different when returns are measured at the household- rather than individual-level. And the reason is straightforward: women (and men) invest in the highest-return enterprises within the household. The facts that female entrepreneurs are more likely to be married to male entrepreneurs (relative to the converse) and that men often gain access to higher-return activities within the household imply that women often optimize by investing in their husband’s business. And as an interesting aside, a parallel phenomenon is present in evaluations of economic research as well – a study showed that when women co-author with men, the man tends to get the credit and, more often, the tenure.
I’d say the big question that research still skirts around relates to the power structures that underlie gender – and that underlie other social constructs. It’s easy to create and introduce an indicator variable for gender in regressions, much harder to create a sensible indicator for the power structures underlying gender inequities.
6. In the last year you have become a co-editor at ReStat. Thinking about your experience so far, are there things you wish development economists would stop doing in their papers, or trends in recent research that you find particularly exciting?
The first thing I would love is for more development economists to submit their papers to us! We’ve been lucky to publish some excellent work lately and we have dramatically improved our response time. We’re happy to engage with young economists at the outset and tell them whether ReStat is the right outlet for a paper and maybe give them suggestions for the clarity of thought that is required for publication.
I would also love to see the central engagement of development economists being with the research question they pose and not the tools used to answer it. The last decade has seen what I consider a phony war between supporters and opponents of RCTs. I think, as a profession we agree that RCTs are a valuable method for identification in some settings but the analytical toolkit available to development economists is, and will remain, broader than just RCTs. As an editor, I would like to publish papers that help the development community to refocus research and policy on the root problems that prevent countries and individuals from reaching their full potential.
As far as trends, I see it as a very positive move that empirical journals in economics are publishing more meta-analyses on topics, and work based on existing data sets. At Restat, we’re thinking about how to create space for shorter, fact-driven papers that usually find homes in science journals. And finally, we’re thinking about how can we maintain the rigor of peer-review process and yet publish research closer to real-time in a way that will be more relevant to policymakers and make the discipline more dynamic. We’d love to hear ideas on all of these from the development research community.