A while back Shantanu posted a link to an article on the question of whether our universities are producing too many PhDs. I don’t have a yes or no answer to that question, but I have some rambling thoughts on the subject. Bear with me, the punch line is way down at the end.

I was a undergrad in the late 70s and a grad student for most of the 80s. (I’m not that one guy who skews all the statistics on graduate degree completion times, but I took a while. That’s another story, though.) I wanted to be a university professor, and for most of that time, university jobs for physicists were very tight. Most universities expanded a lot in the early 70s, then didn’t hire anyone for a long time. Industry jobs got harder to find, too. When I was an undergrad, my profs told me, “Don’t worry, when you finish your PhD in five years the job market will really open up.” When I was an MSc student, they said, “Don’t worry, in five years the job market will really open up.” When I was a PhD student, they said “Don’t worry, in five years the job market will really open up.” What I learned from this is that Physics profs don’t know very much about job markets, or about economics in general. Nonetheless I was fortunate to get a faculty position in 1989, after a short postdoc in the States. There were six faculty positions in physics advertised in Canada that year. If I remember correctly, two of them were 3-year positions and the other four were tenure-track. I know three of the other people who got hired that year, and I have to say that I am in good company.

In the early 90s, students figured out that jobs for physicists were hard to find. Enrollment in university physics programs all across the continent started to drop, and kept dropping for all of the 90s. I don’t have the figures for Canada, but in 1999 the number of BSc and PhD degrees in physics hit a minimum. I think around this time UWO’s honors physics class had all of two students. In the meanwhile, though,  something really did change in the job market. In the early 90s things started to pick up. Not just in the universities – industry started to hire more physicists, too. Opto-electronics, semiconductors, and “high-tech” in general exploded. Investment banks discovered that the problem solving, math, and computer skills that physicists knew were useful in the world of finance, and hired a significant number of physicists at salaries on the order of $300,000 per year. As the world’s economy became more and more computer-based, information based, knowledge-based, whatever you want to call it, people like physicists who were good at dealing with computers, and information, and knowledge became more in demand.

So, eventually students figured out that there actually were jobs for physicists out there. Through the 00s, physics enrollments steadily went back up. In 2010, the number of BSc and PhD degrees awarded in the States was the highest it’s ever been! Again the numbers for Canada are harder to find, but the trends here have been the same, and Western’s honors class has been in the high teens to low 20s for the last few years.

The point is this: If there are physics jobs available, students will take physics. If there are not, they won’t. There tends to be a bit of a time lag between employment conditions and enrollments, so the system is not perfect (actually this kind of feedback system can be mathematically unstable, but again that’s another story) but it is more-or-less self-regulating.

In recent years, both the federal and provincial governments have decided that Canada needs more “highly qualified personnel” to support the “new knowledge-based economy” … blah blah blah. To help meet this need, the governments have dangled money in front of the universities, rewarding them for increasing their graduate enrollments. Of course the universities love this, because they get more money, and more students, and more research papers, and more degrees, and more alumni who when they are rich will give more money to the universities…. But do programs like this really help the economy? What is the effect of these programs on the more-or-less self-regulation of enrollments that I mentioned above? Most importantly, will all of the “extra” grad students were are training find jobs when they graduate?

I said that physics profs aren’t very good at economics, and that applies to me, too. I can’t answer my own questions. I assume that somebody, somewhere, thought about these questions before the governments decided that we needed more scientifically-trained people, but I’ve never heard the answers.

I will end on an optimistic note. According to the American Institute of Physics, the unemployment rate among newly-minted PhD physicists has never been more than about 6% for as long as they have been keeping statistics. It’s typically around 2%. Their graph is quite interesting; you can see it here: http://www.aip.org/statistics/trends/highlite/emp3/figure1a.htm

You can track the strength of the economy by following the curves for “Potentially Permanent Position” and “Postdoc.” Their graph shows a small hump in unemployment in the 90s that peaks at 6% in 1993, but since then unemployment has been lower. Here the message is simply that essentially all PhD physicists have been able to get jobs when they graduate, even when there were supposedly no jobs for physicists out there! I don’t understand economics, but I do understand this: the bottom line is that the skills that physicists have are useful, so people hire them. I am hopeful that this will continue even as the supply of physicists increases.

While I was in Brazil this fall, I visited Raimundo Costa Filho at the University of Ceara in Fortaleza. Raimundo was a prof at Western for a couple of years. Fortaleza is only three degrees south of the equator, and the weather there is the same all year round: sunny, temperature in the low to mid 30s, with a nice breeze blowing off the ocean to keep things from getting too hot. Lovely! And the beaches around Fortaleza are absolutely spectacular.

While he was at Western, Raimundo and I talked about some work I was doing with my PhD student Simon de Vet on the formation of impact craters in granular materials. Since then, Raimundo and one of his former students have been working on numerical simulations of the impact process. I’m looking forward to seeing their paper.

I gave a talk about our crater work in Fortaleza. The experiments Simon did are quite simple – he dropped a ball into a bucket of tiny glass beads, and looked at the crater that results.  Our granular craters look just like craters on the moon: Small ones are smooth bowls with a raised rim, and bigger ones have a peak in the middle. Simon and I were interested in the extent to which the process we were studying really mimicked the formation of craters by a meteor impact. The energy scale in our case was some 20 orders of magnitude smaller, and many of the details were dramatically different, but nonetheless the end product was very similar. There were also some theoretical reasons to think that our experiments really did have something to say about “real” impact craters.  Simon’s work focused on the granular flows that took place after the impact. We think that the fluid dynamics is essentially the same as it is in a meteor impact, and that’s why the shape and morphology of the craters is the same. On the other hand, Simon found that the (appropriately scaled) energetics of the crater-formation process were quite different in the two cases. In big lunar craters, essentially all of the kinetic energy of the projectile is used to excavate the crater, while in our granular impacts it’s about 0.1%. Where does all the energy go in the granular case? Friction, and inelastic collisions between the grains. I spent part of my time in Brazil writing a paper about some of Simon’s work.

I won’t write about Brazil any more, becasue next month we are going to Hong Kong.