First of all, I should explain why I am posting here instead of under the "continuation thread" that twofish created. twofish titled the continuation thread "Value of a Ph.D." However, jksaid's question is not so much about the value of a Ph.D. but the value of an ABD. In other words, he is asking about the opportunity cost of a Ph.D.Here's his question again, for reference:"If I choose not to complete the PhD, what opportunities are open to me?"My guess is that jksaid was aware of the value of a PhD when he applied to his PhD program.That said, let's take a moment to look at the value of a PhD before we go on to jksaid's question. Let's think of the PhD as a project that one can invest in. Let's approximate the PhD as a project that pays CF1 per year for five years then CF2 for 40 years. (Or, more realistically, CF2 could grow over time, or even grow, reach a peak, then decline) To calculate the NPV, we also need to know how to discount the cash flows. These cash flows are not risk-free - let's choose some rate r that is appropriate for the level of risk as the discount rate. (Alternatively, we could imagine a futures market where people could wager on jksaid's cash flows, and the prices of these contracts would provide us with risk-neutral probabilities. Then, we could find the expectation of these cash flows under risk-neutral measure, and discount at the risk-free rate. However, the problems of asymmetric information and moral hazard would be severe for this futures contract - for example, if jksaid completely hedged himself in this market, he would have no incentive to work.)So now we have enough information to calculate the NPV. Do we have enough information to decide whether to do the project? In this case, the answer is no because haven't considered competing projects. (Also, jksaid cannot clone himself, so he cannot take on all positive NPV projects.)This is also ignoring another aspect of the project - there is a real option to terminate the project. This option, of course, enhances the value of the project.What I wrote above would be appropriate for deciding whether or not to start a PhD, but jksaid has already completed his coursework and passed his orals. The opportunity cost of the time that he has already spent is, of course, a sunk cost, so we should exclude it from our analysis. So, starting from where jksaid is today, should he continue the project or should he exercise his option to terminate it?A key consideration in valuing the termination path is whether he will have a Master's degree on this path. It is my understanding (correct me if I'm wrong) that in most PhD programs, if you quit when you are ABD, you can get a Master's degree as a consolation prize. jksaid, is this true for you?Let's explore the "risk" of a PhD in greater detail. Part of the risk is that the cash flows upon completion of the PhD project are uncertain, but part of the risk is that not everyone finishes a Ph.D. in five years. I know someone who took more than 8 years to finish his degree in a quantitative discipline. I know someone who couldn't complete his Ph.D. and had to leave after 5 years with only a Master's degree. These are atypical cases, but it underscores the fact that the PhD is not risk-free. And if you dislike your research, the probability of never finishing is not negligible. jksaid, what do you estimate to be the probability that you can finish the PhD in one more year? two more years? three more years? etc.Implicit in the above analysis is that jksaid seeks to maximize a NPV stated in terms of dollars. But, people are more like utility-maximizers rather than profit-maximizers (though money does play a big role in most people's utility functions). If jksaid doesn't enjoy the PhD program, this should be accounted for as well. Of course, he also might not enjoy the jobs he can get without the PhD. Then again, maybe he would - this is uncertain. What does seem to be certain is that he doesn't enjoy the research he is doing now. twofish also had some remarks about the thinking skills one develops in a PhD program. We can include the value of these developed thinking skills (beyond their ability to generate cash) as "utility."Also, I have a comment on twofish's statement "I think that jksaid making the mistaken assumption that the job searching system is rational" and his associated analysis. I recommend the book "Game Theory for Applied Economists." (but be sure you finish Hull before you start this one) pp. 190-205 cover job market signaling - in particular, the signaling role of education. It's true that the job market seems to place undue emphasis on credentials like an academic degree, but does this mean it is irrational? If it really is irrational, we ought to be able to devise a better system, but as twofish stated, "The trouble with the system is that if you try to come up with a better one, it's really hard." If employers could quickly and costlessly determine all of the attributes of each job seeker, then it would be irrational to rely on academic credentials when a superior measure of ability would be available. However, it takes time to read through resumes (and it is easy to lie on resumes) and it takes time to interview people, and given these costs, it is rational to look for some kind of short-cut (academic credentials) to narrow down the pool of applicants to a group of interview candidates. Academic credentials are a good "short-cut" because they are relatively costly (in terms of effort - especially expensive in terms of effort for people with low ability) for job seekers to obtain and easily verified. twofish, maybe you meant to say that the job market system is not perfect rather than not rational. Whether something is rational or not has a special meaning in economics.
Last edited by propertyrights
on September 18th, 2005, 10:00 pm, edited 1 time in total.