|
daageep, you are on the right track. TA responsibilities take time away from research, so unless you have outside funding or attend a school that makes minimal demands on its graduate students (Princeton is in this category) you will be slowed somewhat by your teaching. Also, not everyone is done with classes by the end of second year. Michigan's program is structured in a way that makes it virtually impossible to be done with classes that soon. In addition to the regular micro/macro/econometrics sequence, there are two other required courses: an advanced quantitative methods class and a cognate (in another, related department -- stats is common). Also, most students take an applied econometrics course in addition to the two semesters of econometrics core.
Having a topic by the beginning of third year isn't trivial either. An even bigger problem for anyone wanting to do empirical work can be having data by the beginning of third year. No problem if you want to work with a big public data set like public use Census, but if you need restricted use data or specialized material, it can take a while to obtain.
And even at schools that aim to get people out in four years, it is not at all uncommon for it to actually take five. A quick look around Princeton's job market candidate page showed that most people on the market have been there for five years, not four.
If you come in with a masters and pass one or both of the qualifying exams at the start of your first year, that will speed things along. But I don't necessarily think that's the way to go. You'd spend two years getting a masters, which you are probably paying for, to avoid one year in a PhD, which you are not paying for. I think the single best thing to do to get yourself through the PhD quickly is to obtain funding that doesn't have strings/obligations attached, so that you can spend most of your time doing research.
As for whether or not it is desirable, I think it's better to come out in five years with a very strong job market paper and a couple of other things in the works, than to come out in four with a mediocre paper or even a good paper but NO other work in progress. Schools that interview you will ask about your plans for future work, and you need to have something to tell them. Also, as an assistant professor, you need to publish to get tenure, and it helps to have things underway before you start. Being an assistant professor puts a LOT of extra demands on your time (teaching!) that you don't have as a graduate student. Why not get more work done before the tenure clock starts? There comes a point when taking too long sends a bad signal, and of course it sucks to get a grad student's salary rather than an assistant professor's. But I'm not sure that rushing to be done in five years is the strictly dominant strategy.
|