So crazy it just might workPosted: April 18, 2013
Summary: There are no bad ideas in brain storming! Let’s make grad students pay for their PhDs.
A post by Dr. Zen yesterday morning prompted a Twitter discussion about what’s wrong with the science training system. I’ve been an advocate of limiting intake into PhD programs by banning spending federal research funds on trainee salaries. This would put control over the number of trainees into the hands of funders (either through things like NIH F mechs or by earmarking separate, limited funds for trainees with research grants, e.g. 2 trainees per R01). I would further take the NIH out of the business of choosing WHO gets to do a PhD by allowing them to allocate the funds but have no say in who the PI hires or the program accepts. The main problem with this approach is that it is predicated on the assumption that NIH is competent to do anything differently and gives a fuck about the training pipeline problem, which 1) They barely acknowledge and have shown no willingness to address, 2) They probably don’t see as a problem.
What prompted me to rethink this is the question of whether taxpayer money should be used to fund science PhD training at all. Currently, the number of PhDs who will go on to use their training to conduct publicly-funded research is somewhere under 20%. The largest share of the rest will go to industry, and many more will not become scientists of any kind. I see no compelling reason the taxpayer should foot the bill to train a labor force that will fill the coffers of drug companies by selling the fruits of their labor back to the taxpayer who funded their education.
My proposal is that we make students pay grad student tuition and fees, like other professional higher degrees. If, like medical school or law school or architecture school, doing a science PhD put you in substantial debt, this would solve two problems: 1. It would limit the number of people seeking a PhD by making it a harder choice, rather than a default for people who couldn’t think of anything else to do, thus relieving some of the job market pressures PhDs face. 2. It puts the cost of training PhDs on those who will benefit from their work.
This sounds evil. PhD students already struggle in high CoL areas. But consider this… law students and med students incur large debt because they can expect careers that will allow them to pay it off. One reason medicine and law have such high starting salaries is the recognition that their new employees are normally starting their careers in substantial debt. In both these fields, what this means in economic terms is that hospitals and clinics pay for training their doctors and law firms pay for the training lawyers.
Right now, you’d be crazy to take that risk doing a PhD. But this would change… fast. The number of people doing PhDs would drop quickly to near the number who can find employment as scientists afterward. PIs can spend their grant money on research instead of on salaries. Best case scenario: postdocs become rare to non-existent.
There are a lot of possible objections to this.
1. Everywhere that there are people, we need doctors and lawyers. Less true of scientists, the market for which is more volatile and patchily distributed. Someone from a third-tier law school can still practice some kind of law somewhere. On the other hand, the medical and legal professional societies make some effort to regulate the numbers of trainees in their fields, and scientists could – very theoretically – do the same.
2. Many people with science PhDs do important, “non-traditional” work that might not support a high level of initial debt. Well, this is true of MDs and JDs, too. There could be debt reduction or forgiveness plans for working in the public or non-profit sectors, etc.
3. Who loses? There is a trade off for PIs and departments. Now YOU will have to compete for the best students, who will suddenly be much rarer, rather than them competing for you. You will be judged by career-outcome metrics for your trainees at the lab and department level. Yes, this could greatly exacerbate bullshit like pedigree, glam, and rich-get-richer problems as labor is concentrated in the most “desirable” areas, though I’m not sure these forces could be stronger than the are now. On the other hand, you will get federal student loan money coming in and not waste grant money paying people. But mostly, fuck you because you’ve been benefitting enormously from a shitty system for decades. As I’ve noted before, every attempt at labor reform ever – minimum wage, unions, parental leave, weekends — has been met with the same teeth-gnashing, hair-rending cries that it will reduce competitiveness and efficiency and woe betide us because nothing will ever be the same again with the peasants in the castle. On the other hand, maybe faculty salaries in the sciences would start to look more like those in medicine and law. That shut you up quick, eh?
4. Who else loses? A lot of people who would have done PhDs will not. Maybe we will get the “wrong” people. I don’t know, this doesn’t resonate for me. There is a danger that economically disadvantaged students would be more averse to taking on debt for grad school. However, these issues are addressed in some ways in medicine and law. Another argument is that having well-educated scientists out and about in the world is a good thing. It is, but I’m not sure the numbers were talking about here matter against the general population. If we want a scientifically literate population, the place to start is not having more people get PhDs. That’s like saying if we want incomes to rise we should have more bankers.