Last week, an email from David McKenzie alerted us Development Impact bloggers to the following: “Christian Meyer received the Vilfredo Pareto prize for best economics thesis and part of the announcement reads “
How did the job market go? Where will you be next year? Was it affected by the pandemic?
Before getting into the former candidates’ answers, I should start by saying the following: of the 16 people I emailed, we actually hired three of them at the World Bank: two into the Development Research Group and one into DIME. The rest of the responses suggested a satisfying, albeit sometimes stressful experience through the job market. Most candidates were lucky to have their flyouts finished early enough that they got back home just before lockdowns, etc. and generally unscathed by the pandemic. Fortunately, I have not heard from anyone that their offer was later retracted due to the pandemic, as it seems to have happened in at least some institutions. The candidates had good jobs: Wharton (UPenn), University of Maryland, University of Hong Kong, USDA’s ERS, Paris School of Economics, and the University of St. Andrews.
Did you feel that preparing the post for
Here, there were two types of responses: both positive. Some candidates had already prepared their pitches for the job market and for them, the preparation of the blog post was a matter of editing that for their submission to us. In such cases, where the candidate was already prepared before Thanksgiving and had perfected their pitch, the back and forth with their editor/reviewer at Development Impact also ended up being mostly editorial and not hugely substantive. For those candidates, the main benefit would come from dissemination.
However, another, perhaps larger, group suggested that the exercise of writing the blog post was very helpful for them – mainly because it helped them understand what parts of their pitch was “too much into the weeds” and which parts absolutely pertinent. The comments from David, Florence, Kathleen, Markus, and I helped them understand what arguments were unclear, what seem important to highlight, address questions about methodology succinctly, and so on. In that sense, the ‘light editorial’ approach that we take to this exercise seemed to have helped – even if a little bit. I can attest to this: sometimes you read the initial draft of a post and take a look at the paper. You have a sense that you really like the paper or the idea, but it’s not shining through the 1,200-word post. In such cases, sometimes we do have a round or two (or even three) of back and forth with the candidate to try to improve the write-up.
Candidates also mentioned the value of having to put their work in non-technical language, adjusting their pitch to crowds with different levels of knowledge about the background, etc. These are all things that can come from a diverse set of eyes reading your work, which is hard to do at your own institution only.
Did you get any sense, implicit or explicit, that people had seen your post and this helped with your interviews, etc.?
Again, there were two typologies of answers to this question: most of the candidates who responded, being the good economists that they are, mentioned that they could not attach causality to the interest in their paper being from their blog appearing in Development Impact.
[Over the years, people have, every year, suggested to us that we do some experiment and measure this and we resisted the urge – mainly on ethical grounds, but also for lack of statistical power. The least problematic design would be doing something akin to David’s business plan competitions work in Nigeria (see the paper here), where, we would expect the top submissions, reject the lowest-ranked ones, and randomize the group in the middle, for whom we are on the fence. Just announcing the experiment would change the composition of submissions on the extensive margin, already altering the purpose of this exercise. In addition, at the current volume of submissions, it would be an experiment with 30 or so blog posts in that middle group, a quarter of which would be randomized into appearing in Development Impact. Not worth it…]
The first group was not sure, other than mentions from a few peers, etc. that the people who invited them for flyouts or made them job offers had seen the post. No one explicitly told them, although they were fairly certain that the post had increased their visibility. The second group was more emphatic: a number of people explicitly told them that they had seen the post in Development Impact. Such awareness of the candidates’ work made some of the one-on-one meetings at the flyouts easier to start the discussion of their work, especially prior to the seminar – as they could skip the ‘spiel.’ Others received explicit invitations to give talks at various institutions, such as the Inter-American Development Bank, governments of the countries about which their JM paper was, Oxford University’s Blavatnik School of Government, a World Bank country office and so on. Others received invitations to write about their papers elsewhere, many of whom republished the post in Development Impact with permission and proper attribution.
And, consistent with Christian Meyer’s experience at the start of this post, some people mentioned the ‘signaling’ value of having their blog post appear in Development Impact: it does seem like there may be some effect of the job market paper having been screened and selected by us on decisions of some institutions on the margin. People did also mention the number of visitors to their personal website, number of downloads of their papers, and number of Twitter followers increasing on the day their blog post was published here. One of the candidates had their post reposted by LSE and IFS and featured in the media in Peru.
Would you encourage your own students on the JM to submit to Development Impact for the same thing in the future?
The answer here from all respondents was a clear ‘yes.’ The reasons cited included: the difficulty of disseminating one’s work as a graduate student and the value of reaching so many people in the field; being able to share the link with policymakers or potential future collaborators, which makes it easier than sharing the paper, by including it on their website; connecting to more people on Twitter; the whole exercise improving their ‘elevator pitch;’ and the appearance in Development Impact being a positive signal to the market about the quality of the paper. I’d like to thank one anonymous respondent, who said that getting their JM blog post into Development Impact was one of their dreams as a graduate student. We also heard from a candidate that some students may have been discouraged from preparing a submission, because it was deemed too competitive. I can tell you that getting into Development Impact is not as competitive as trying to get into a typical field journal: we have been publishing around one in three submissions, which is a much higher percentage than those in the single digits for journals. And, as one candidate said, “I think it's definitely worth the shot. Even if the JMP is not selected, the exercise is extremely useful.”
Is there anything we can do to improve the process?
To be clear, this was not one of the questions that I sent the bloggers, although in hindsight I should have. But, I was very grateful for two of the candidates bringing up the issue of the timing of the posts. As you know, our tradition has been to have a deadline around mid-November, and then start the string of posts the week of Thanksgiving, continuing well into December. This has generally been influenced by the fact that most students have (used to have?) their packages ready by mid-November, so it seemed appropriate to ask them to write about their papers then. And, the ASSA meetings are not until early January…
However, the counterargument, as the students put it eloquently, is that a lot of the flyout decisions are made over Thanksgiving in the US. The ones that did not get made then are mostly made by early December, especially because the European job market is earlier than the U.S. in late December. Given that we go almost all the way up to X-mas with the posts (we usually do three the week of Thanksgiving and 12-15 afterwards), many of these posts come out too late for visibility among people who are making flyout decisions. While the primary intent of starting this tradition was not for this purpose, a slight adjustment in the timing to move the submission deadline and the posts earlier could help.
My current thinking is that we might consider moving the deadline to the beginning of November and have 15-18 blogs appear all the until the day before Thanksgiving. If some people miss that window, we could have a small number of ‘late breakers’ blog posts appear in December. Of course, these are just my quick thoughts, a decision on this would be made by all five of us later this Fall. Your comments are, of course, welcome.
So, there it is. Overall, not bad. I certainly enjoyed catching up with the bloggers, learning where they ended up, pleased that they were not affected too much by the pandemic with respect to the outcome of the JM, and hearing about their experiences and suggestions. I hope there is something in it for you and for your present and future students as well.
I thank Sarah Baird for encouraging me to do this follow-up with the JM bloggers and discussing it with me before posting.