What's Involved In Getting a Ph.D. In Finance?

This seems to be take your readers to work week -- in my last few pieces, I've discussed the teaching and research sides of a finance professor's job and where data for research projects comes from. So, this time, I thought I'd take a step back and talk about what is involved in getting a Ph.D. in finance.

The biggest misconception is that the Ph.D. is something akin to a "super MBA". In reality, the MBA and Ph.D. programs are almost totally different animals. The MBA is geared towards practitioners. In contrast, the Ph.D. is primarily intended as training to be a researcher. So, the natures of the programs (and therefore their approaches) are distinctly different from each other.

Some doctoral programs require an MBA before entry, but quite a few don't (like the Unknown Alma Mater). In fact, the best preparation for getting a finance Ph.D. isn't an MBA - an MS in Finance, an MS in Econ, or a masters in math, engineering, or physics (there's a LOT of math involved at this level) would probably prepare you better.

In terms of admittance standards, you'll need a GMAT score in the mid-to-upper 700s to be even considered for admittance at top schools, and even lower ranked schools will probably look for a GMAT well above 600. As a benchmark, I went to a solid "2nd tier" program, and my classmates all had GMATs of between 700 and 760. Of the two component parts of the GMAT, the math score is the more important than the verbal one, and most schools look for someone at the 90th percentile (for lower ranked schools) or better. At top schools, almost everyone has GMAT math scores in the top 1 or 2%.

Probably the best way to describe a Ph.D. level curriculum is to start at the end and work backwards. The final step in getting a Ph.D. is to write a dissertation, which is an original piece of research. In finance, dissertations usually run between 65 pages (the shortest I've seen) and 150 pages. The dissertation is supervised by one person (called the chair of the dissertation committee), and must also be approved by a committee of 3-4 other faculty members. There's a love-hate relationship between Ph.D. students and their dissertation chairs, because a good chair constantly asks the student for "more" - more analysis, better writing, more literature review, etc... Because of this, the dissertation is probably the most exhaustingly thorough piece of research most Ph.D.s will do in their lifes.

Working backwards, to be able to do original research in a dissertation, you must be familiar with what's already been done on the research question you're asking. The main way a student gets this familiarity is through "seminars", which are the backbone of a doctoral program. Seminars are much more self-directed that a typical textbook/instructional class. In a seminar, you may read anywhere between 25 and 100 journal articles during the course of the semester (one of my professors covered almost 120 articles in a 10 week quarter, which was a brutal pace). For example, in a typical 3 hour seminar week, you typically cover anywhere from 3-8 articles. If the article is a theory piece, it will have a good deal of math (calculus, partial differential equations, real analysis, or linear algebra), and if it's an empirical piece, there will be a good deal of statistics (more math).

In an undergrad or MBA course, the professor usually presents the material. However, in a doctoral seminar, the papers get presented by the students. Each week, one or more students walks the rest of the group through the article (or articles) to be covered, and the professor asks questions (usually from the sidelines). In many of the seminars, you're also expected to produce a small research piece as part of the seminar. In addition to giving you hands-on experience doing research, one of these small seminar research projects often becomes the basis for an eventual dissertation.

There are a number of different ways to organize the seminar sequence, but typical seminar topics might include Finance Theory, Corporate Finance, Investments, Derivatives, Markets and Institutions, and Empirical Methods.

Most students don't have the skill set necessary to handle the seminars right off the bat, so the first year (or in my program, about the first 1 1/2 years) is devoted to "foundational" classes. In the case of the program I attended, this involved classes in microeconomic theory (much of finance is nothing more than applied microeconomics), statistics, linear algebra, real analysis, and econometrics.

Since I worked you through the sequence backwards, here's how it looks going forward (with approximate time ranges). Again, this is based on my program, and others might differ:
  • Foundational classes - 1-1/2 years
  • Seminars - 1-2 years
  • Dissertation - variable, but usually 1-2 years.
Adding it up, the typical time to complete the doctorate is about 4-5 years. The fastest I've seen it done at my school was a little over 3 years, and the slowest is about 8 years.

Note: If you're new to Financial Rounds , welcome. I hope you look around a bit -- if you want to find out more about the blog, check out the Frequently Asked Questions (FAQ) page. And if your want to subscribe to our RSS feed, there are links on the sidebar.