Large-Scale Collaborative Work

The largest and most intensive piece of work I have done in development was co-authoring the 2015 World Development Report, which focused on what the social and behavioral sciences had to offer development policy. Over the course of 2013-2015, I was part of small core team of authors led by Karla Hoff and Varun Gauri. The World Development Report (WDR) is the World Bank’s flagship report—a new report is released every year, and each year it is on a different topic important to development policy. The topic is largely determined by an internal competition, and the report is considered a staff report, meaning that the team authoring it has broad autonomy to develop the report as they see fit. There are multiple stages of peer review from both inside and outside the Bank, which includes a fairly massive outreach effort, covering aid organizations from many countries, many independent NGOs, academic experts, and others who provided a great deal of insight and criticism.

This was a very large project, to say the least. Unlike my initial work with UNICEF, which was confined to the summer months, this project continued for nearly two years, from when I was first brought on board to do very' preliminary research work, to the end of our promotional efforts around the world post-publication. During this time I maintained a research fellowship at Penn. The original plan was that I would spend approximately 60 percent of my time in my World Bank office and 40 percent of my time at my Penn office and, as such, treat each job as part-time. In reality, I had two full-time jobs, and had to perform at each as if the other did not exist. This was easily the most intensive period of labor that I have experienced.

At its core, the task of writing the Report requires that over the course of a year, a group of about eight people (with some assistance from outside authors) writes a book that offers an operationally useful analytic framework to make sense of the relevant academic knowledge together with the published outcomes of development initiatives in a particular topic area. Our task was especially challenging, as themes for Reports are rather technical—for example, in 2008 the Report was on agriculture, in 2010 it was on climate change, and in 2013 it was on jobs. Our assigned topic was the relevance of social and behavioral sciences to development policy, and thus was more methodological. This was in one sense quite exciting—it was work that was relevant to everyone, regardless of region or sector—but also challenging, in that there was not a natural scope to the project. Because the subject itself lacked a clear scope, our initial efforts revolved around determining what to include and what to exclude. We ultimately decided to divide the Report into three parts—the first offering a general analytic framework for thinking about social and behavioral issues in development, the second offering more sustained examinations of how these issues play out in certain policy areas, and the third turning the focus to development organizations themselves, examining how behavioral insights might inform better practices for development professionals. In my own view, our most important contribution was in that last section: it is easy to fall into the belief that experts know what’s best, and that it is only the targets of policy that are subject to biases or are in the thrall of an unhealthy norm. We demonstrated empirically that policy experts also make systematic errors of judgment. This led us to suggest a number of tools and reforms that could improve the policymaking process in light of these biases, heuristics, and norms.

Working on this project was highly collaborative. As a group we worked out the core areas to cover, developed a framework to help systematize the use of behavioral tools in development, and then co-wrote chapters and supplements. We also had an experimental component to this project, and so part of the team developed an extensive set of lab and field experiments that were conducted and discussed in the Report. This was a project with many moving parts, and it required a broad range of skills. This diverse range of skills also brought with it a diverse range of views about how to carry out our work, and so this required a fair amount of deliberation, negotiation, and consensus building. This experience cemented for me the importance of perspective taking and flexibility. While I was nominally brought on board to work on the more academic component on norms, I ended up working more on some of the applied topic areas, and the policy process component.

Working at the World Bank is different from academia in another important way: meetings are quite important and are often where much of the real work gets done. Academics are used to the idea that departmental meetings are tedious and pointless, but at the Bank, my experience was that meetings could be incredibly productive and were often how key decisions were made. This required skills in building consensus and seriously addressing people’s concerns. Simply putting one’s head down and working might be okay for a first draft, but it may be a draft that’s simply thrown away if there isn’t any support for it. This work was fundamentally a collective product, not just sections written by different people stitched together by a lead author. In part because it’s unusual to co-author papers in philosophy, these are not skills that are emphasized. But when one works with so many collaborators, it becomes essential to employ these skills.