Job Market Debrief

Q&A

Q: If the interviewers are all theorists in your field, would you recommend saying something about your general research briefly and then jumping directly into the problem you study?

Kevin: It is good to begin with a few sentences saying these are general questions that I consider. You might say, "I focus on questions in information and learning, and my job market paper is about blah." That part is good. I think you could start with a general statement about the area you study before launching your job market paper. Nevertheless, even though they're all theorists, I think you can't expect them to be all theorists in your area. There may be decision theorists, or someone who works on mechanism design, or something very different than what you work on. So you can be a bit less of a general audience than if you're talking to a macroeconomist. But I still think you would want to motivate the question. You still want to say why this is an important question.

Xiao: It is helpful to motivate something specific to your topic but not obvious to other theorists working on different projects. For example, suppose you work on robust mechanism design. You might not have to motivate mechanism design, but you’d better emphasize why we should care about robustness.

Ji Hee: In the first two minutes, as both said, it's almost impossible that you just get an interviewer within your field, so you need to have an initial motivation as before. But I think you can shorten it a little bit. For example, my interview started as follows: “I'm working on market design. I want to think about how the change in the trading mechanisms or the transparency design or decentralized market would affect the trader's behavior, equilibrium outcomes, and efficiency“ as a general version. If I meet a finance faculty, I will just say I'm working on market microstructure and then go into the job market paper faster. So you can change the speed a little bit. But I agree that you cannot skip the motivation and jump into the model. You still need to say your question and your motivation in at least three or four sentences. But potentially, you can use some jargon in your field.

Q: Is impressing the theorists the necessary but not sufficient condition for a good interview?

Xiao: Getting the support from theorists is definitely crucial. But to what extent you need to convince all the people present could depend on the department and the position (e.g. it is a general slot or a theory-specific slot.)

Ji Hee: In my experience, which may not apply to every case, if there's one theorist in the room, probably, that's the person who invited you. So it is relatively fair to assume that this person already liked you enough. So I think your target needs to be broader than that person. He needs to have some support to invite you to fly-out. So I think you need actually to impress not just him. Of course, that person is a necessary condition, but I do not think that person needs to be your main target. It is too narrow and too risky. So broadening a little bit is potentially beneficial. For example, if your research is closer to IO, potentially a person in IO would be the next who can like your research. Think about who can be your support potentially. Of course, you may not have enough time to tailor anything right before every interview. So be creative about imagining who can be your support.

Kevin: I think at some schools with a lot of theorists, it may also just be the case that a couple of theorists like your research ending up giving you the interview. In that case, slightly broadening a bit might impress other theorists.

Q: Do you have any advice for early-year graduate students? What do you think you can improve in your graduate years?

Ji Hee: I think it would have been better if I talked to the seminar speaker early on because that's the experience you will have during the interviews anyway. Participating in the interaction with seminar speakers, conference attendees, and student groups is something you can start earlier if there is a chance. The most important thing is still, as Xiao said, your job market paper. So what I said is conditional on the fact that you have a good idea and a clear plan for your job market paper.

Kevin: It's helpful to start writing papers or just start thinking of ideas. It's true that you only need one job market paper at the end, but it may be difficult to generate that one idea in half a year. So even for theory, you start thinking about generating ideas in the third or fourth year. It's fine that your first idea is not good. Many projects will be dead ends. The nature of research is that perhaps 90% of projects are dead ends. So given that 90% of projects are dead ends, in expectation, you need to generate ten ideas to have one that works out. So you should start early and start getting feedback from faculty, internal lunches, and stuff like that. It's very helpful.

Xiao: I would add that it's very beneficial to talk with your peer students, theorists or non-theorists. Being exposed to broader topics helps you understand what general economists are interested in. And also you might be able to write papers together. I feel nowadays, co-authored papers are penalized less and less over time because people realize at the end of the day, the nature of science is cooperation. If your job market paper is co-authored with a junior faculty or your classmate, I don't think you will get punished at all. It's a lot more fun and efficient to work with a co-author than to work alone.

Q: You may have a lot of ideas during your graduate years. Is there anything to help you find out that a particular idea should be your job market paper?

Ji Hee: I guess it depends on which stage the project you're considering is. If you have projects basically close to being done, the best paper must be the job market paper. If it's in the early stage of the project and you're selecting which idea you want to pursue, I think the best person who can give you that advice is probably your advisor or people in your field. They will better understand whether this idea is feasible, exciting, and contributes to the literature. You need to get feedback earlier, even if you feel it is too early, and you may be embarrassed. Just go to the advisors.

Xiao: It is really hard to say as researchers and papers are so heterogeneous. I think in many cases JMPs are signaling devices, and candidates use them to signal their creativity, clarity, technical ability, understanding of the field, ability to talk with other groups, etc. So when choosing a job market paper, this might be an aspect for consideration.