I should note that all my answers are based on 12 years experiencing advising Berkeley graduate students. I'm not sure how well it translates to GMU students. Also, my info is 10 years old.
>1. Once you are almost ABD, what is the opportunity cost of prolonging your >graduation one more year, as opposed to just taking your chances at the job >market right now? Depends on how long you have been working on your PhD. It used to be the case that taking more than 4 years to finish an Economics PhD was seen as a bad sign and going on the market in your 4th year even if your PhD dissertation wasn't that far along was the normal advice. However, it is my understanding that the 4 year norm has been breaking down so that may not be an issue so much anymore. If not I'm sure there is still some marginal penalty in that people who finish early are considered better risks than people who take a long time to finish their dissertation. >2. What is the opportunity cost of trying the job market "prematurely", and, >worse comes to worse, just cancel the job search if nothing appealing comes >about and then try again the year after that? The cost of going on the academic market more than once is huge. If you go on one year and don't get a job you are known as damaged goods (someone no one wanted last time). Since testing the water and then pulling out is not the norm this hurts. Avoid doing it at all cost if you want a good academic job. If you are in some other market where people aren't watching who is coming out of where every year I doubt it would matter. >3. What is the benefit, at the margin, of going into the market with say >one, two, or zero publications? (For instance, perhaps the market is so >tight that the marginal benefit is not that big--after all, low-tier >institutions may not be that different.) Having publications before you go on the market can make a very big difference. At least 10 years ago very few people did and having published work was taken as a very strong signal. However, most hiring committees can probably do the math. Five years with two publications vs. four years with 1. But who are you kidding. You really don't have a choice. If you go on before your advisor tells you to our don't do it when your advisor tells you to you are going to get your advisor ticked off at you and then you will never get a job anyway. - - Bill Dickens