Canadian universities have strong incentives to create PhD programs, and admit students into those programs. This is because provincial governments typically provide generous funds for each PhD student a university takes in. Also, PhD students are useful and cheap workers. Moreover, having a PhD program raises an academic unit's status, by signalling that it is a "research" unit rather than a "teaching" unit. For individual faculty members, having a stable of PhD students makes it easier to get research grant funding, partly because the faculty member can show that they are active in "training", and partly because PhD students can be used as co-authors.
This incentive structure means that it can be in a department's interest to create a PhD program, even if the job prospects for their PhD students are mediocre, or only half the students they admit graduate. It can even be in a department's interests to admit PhD students that no faculty member wants to supervise. For example, if the university rewards units for admitting some target number of PhD students, or if the university threatens to close the PhD program if not enough students are admitted, there are strong incentives to take a chance on a student with poor prospects for success.
It's cruel. Students suffer. They sacrifice years of their lives, and tens of thousands of dollars, in the pursuit of a PhD, and too often end up with shattered self confidence, no degree and/or no improvement in their job prospects. (I've not given you numbers are on PhD completion rates because, to the best of my knowledge, no published data is available.)
Others have noticed these problems and suggested solutions, for example, greater differentiation among universities, and tying funding to student outcomes. Yet there are serious issues with both of these models as applied to PhD programs.
Differentiation does not work because the PhD is primarily about the supervisor/supervisee relationship. Labeling some universities "doctoral institutions" and others "non-doctoral institutions" prevents students from working with smart people who would be good supervisors at so-called "non-doctoral institutions." For example, Scott Sumner has, for many years, taught at Bentley University, which has no doctoral program in Economics (as far as I can see from the university website). That has meant that PhD students have had limited access his expertise on market monetarism. In "doctoral institutions" a different problem arises. Students flock to a particular institution hoping to work with The Great Professor. However The Great Professor only has limited time for supervision, so cherry picks the best PhD students and agrees to supervise them – and then complains about his/her supervisory load and asks for course releases. The Almost Great Professor does the same thing with the almost best PhD students. The least attractive PhD students struggle to find supervisors who are a good match for them.
Tying funding to student outcomes does not work because the time between PhD admission and outcome realization is so long. Moreover, it is not obvious, when it comes to PhDs, how to measure success or failure. Outcome measures typically look at employment rates, or salary levels. However even if a PhD graduate has no better employment or salary prospects than an MA graduate – i.e. the return to a PhD is zero – that will still translate into a relatively high employment rate and salary level. Then there are the usual issues with controlling for student characteristics when measuring outcomes. For example, since immigrants typically have inferior labour market outcomes as compared to the native born, presumably one should control for the heritage of students when comparing the outcomes of programs, but it is not obvious how to do this.
I would like to propose an alternative solution: the PhD Voucher.
Here is how it would work. The provincial government – the major providers of PhD funding in Canada – would announce a PhD competition in each discipline. International students would also be able to apply. An adjudication committee would rank the candidates, and decide to whom they will award PhD vouchers. The value of a voucher for an international student would be lower than a domestic student's voucher value.
Each student would then take the vouchers, go to universities, and buy courses and supervisory services. Students would be able to mix and match – buying, for example, an econometrics supervisor from Carleton and an economics of education committee member from University of Toronto. Each university could decide how to split the voucher payment between supervisor, department, and university.
PhD programs typically involve some amount of specialized course work and examinations. The comprehensive examinations, like the admissions, would be centralized. I imagine universities competing to offer PhD micro and macro theory/comp-preparation courses. For example, each year three, say, or five micro theory courses would be held across the province, with students opting to take the course most convenient and attractive to them. For field-specific expertise, students could use their PhD vouchers to take MA level courses at any university.
One beauty of the PhD voucher scheme is that it would be possible to have a hard link between progress towards a PhD and continued funding. Too often, I have seen PhD students trapped in academic limbo: the student is not making progress, but the incentives to keep students in program are so strong that no one at the university is willing to pull the plug, and kick the student out of the program. An inter-university evaluation committee would have no such incentive to keep funding students who were not making progress. The threat of losing funding would give PhD students – and their supervisors – a real reason to get stuff done.
It's odd: school vouchers are frequently advocated for primary and secondary school students. Yet there are good reasons why they would not be expected to work well, especially for primary school, as the ones exercising the vouchers (the parents) are not the ones experiencing the education (the children). Also there are very large costs associated with switching schools, and especially going to a school outside your immediate neighbourhood, if you are six years old and not able to drive, or 12 years old and in a tight friend network. Hence the possibility for effective competition between schools is limited at the K to 12 level.
If voucher schemes were to work at any level of education, they would be expected to work for PhDs, where the person exercising the voucher is highly informed and mobile. In any event, the downside risk of implementing PhD vouchers is negligible: it is hard to see how a voucher scheme could be substantially worse for students than the status quo.
Worthwhile Canadian Initiative 
“Students would be able to mix and match – buying, for example, an econometrics supervisor from Carleton and an economics of education committee member from University of Toronto. Each university could decide how to split the voucher payment between supervisor, department, and university.”
Interesting post, for sure. Not obvious to me how the split above would work: U of T and York (say), yes, but Uof T and Carleton? Transportation costs? Housing? Online courses? I understand that implementation is always tricky, but wonder if you have thoughts on this aspect?
Linda – Right now the provincial governments fund universities, who then give funds to students in the form of scholarships, which the students use to pay for housing, transportation, etc. What would happen under a voucher scheme is that the voucher would include funds to cover the cost of living.
Now right now some PhD students receive more funding than others – so a university will offer $XX to their top candidate and $X to a mid-ranked candidate. Under a voucher scheme, would every voucher be worth the same amount, or would top students get a bigger voucher than non-top students?
I think the greater practical worries are things like highly desirable universities/supervisors ratcheting up prices, and how students would be branded at the end of the day. Because a big part of what a PhD student is buying branding – the right to claim “I’m a graduate of …” How that would work with a voucher scheme I don’t know.
Perhaps a modified voucher scheme when students could choose which university to cash their voucher in at but otherwise the system worked not unlike the present system? Unfortunately (or fortunately, depending on how one looks at things), one strongly suspects that under such a scheme, students would swarm the top-ranked university.
“An adjudication committee would rank the candidates, and decide to whom they will award PhD vouchers. ”
~~~~~~~~
It’s easy to see that this is going to go horribly wrong. That people on the committee are going to have biases against some disciplines and against different people – conscious or unconscious. So it’s not hard to imagine what will happen – PhD candidates will be skewed while and male and in the “hard” disciplines.
@mpledger: I think that making the committee work is the hardest part of the system. It’s possible to avoid the bias you cite through points and quotas. Medical school in Canada is not overwhelmingly male. But it does mean that if the committee stops functioning then the whole scheme is toast.
I also worry about gaming of the system. If you can make UBC a choke-point (imagine a key course or something) then could it end up with all of the PhD students in a specific field? It becomes a virtuous circle — if all of the successful ethics PhD students go to UBC then how do you convince somebody to take their chances elsewhere. Preventing that requires something like the CIHR or NIH that independently works to make sure that the system isn’t undermined. So the committee feels like the crux of the whole plan.
mpledger – “It’s easy to see that this is going to go horribly wrong.”
There would have to be a two-step allocation process – allocating PhD vouchers to disciplines, perhaps on the basis of, say, demand for PhDs in that discipline, and then allocating vouchers to students. Though of course the allocation of SSHRC doctoral fellowships or insight grants across disciplines does not give one cause for optimism!
Joseph – ” If you can make UBC a choke-point (imagine a key course or something)” “It becomes a virtuous cycle”
From a pedagogical point of view, perhaps the virtuous cycle is indeed virtuous – perhaps it makes a lot of sense to have all of the ethics PhD students in one place. Let me ask you: do you think the current system of sprinkling PhD students across universities is the best possible one from the PhD students’ point of view? I’m not convinced it is. Whose interests are served by the current system?
The key to making a voucher system work is the student/supervisor relationship. The problem with the current system is that universities have big incentives to admit marginal PhD students but individual faculty members have little incentive to supervise marginal phD students, so the students end up in academic limbo – no one will flunk them out, no one will get them through their degree.
Perhaps imagine that students have to name three or four potential supervisors when they apply for a voucher and someone has to say “I am willing to supervise this student’s work” before a voucher is granted. In economics there’s a hard limit to the number of students that any one prof will take on, and that would prevent swarming.
Reading half part into it, it reads like some “dating scenes” (except for the “stable of students” part – though I’m not even sure about that).
Aside from what others pointed out, one potential problem with the voucher scheme, but that already exists with the existing funding schemes, is that an institution may have an incentive to milk students for part of the voucher value and then drop them. It could be a “viable business model”. What will ensure making an effort towards completion? Maybe “ratings” and completion statistics “reputation value”?
Which brings me to the point “tying funding to outcomes” – the primary measurable outcome is completion of the degree. So funds (or a target level of the funds) can only be collected when the degree is awarded? I can’t see anything that could go wrong with that. (Also compare: students rating the “effectiveness” of their lecturers and TAs, i.e. how happy they are with their grades.)
But doesn’t this go back to the “real problem” discussed a few articles back – that in large part it has become about credentialism, i.e. the mission of educational institutions drifting from education/research/etc. to issuing degrees for labor market signaling purposes. Of course for a reasonable fee – no free lunch.
But after this load of cynicism a countervailing perspective from a world where academic degrees were not (as much) of a commercial endeavor – i.e. cost of living/housing was a much larger part of students’ cost structure than tuition (if there would even be a tuition charge at all) – which would still be a considerable financial burden for years to come. Actually two worlds.
In those worlds, doctorate degrees were still used for signaling, but not for getting run of the mill “subject matter expert/contributor” jobs, but as a requirement for advancement to higher technical management (in technical lines of work), more generally higher management, or higher government positions (usually doctorate degree in non “technical” field). Then some people like to put a Dr. in their name (but usually also because of socioeconomic incentives). Academic careers are in reality a minority phenomenon – most people want the degree for outside use. And frankly a number of people got a PhD because it allowed them a few more years in school working on interesting stuff before joining the drab world of work.
In Germany, academic titles like Prof. or Dr. actually become part of your legal name, and individuals and institutions are obliged to address you with your title (and I guess you are obliged in turn to announce your title so you can be properly addressed). That’s why German websites have “title” fields in forms for personal information where you have to select/enter “Prof/Dr/Mr/Mrs” etc.
In the past few years, several government figures include the ministers of defense and education were busted for having obtained their titles with plagiarized or not properly cited material in their dissertations, their titles were retracted, and they had to (or anyway decided to) step down. As part of that it was considered to end the practice (and incentive structures) of attaching academic titles to the name, making it a social marker. But it seems grass has grown over it, and nothing happened. Also the scandals subsided.
In “real life”, people who insist on very prominently announcing or emphasizing their title, or being addressed with their title in informal communication (formal esp. written is another story) are pretty much despised and ridiculed behind their back. The line is kind of fuzzy. If somebody introduces themselves as Prof./Dr. X, nobody will blink an eye. But if there is a pattern of correcting people “actually it’s Dr. X” in unofficial contexts, that’s considered overstepping some line, even though there is a legal entitlement.
cm: “Which brings me to the point “tying funding to outcomes” – the primary measurable outcome is completion of the degree.
Right now it’s possible for a student in a second or third or even a fourth year of a PhD program not to have a viable thesis topic, not to have a supervisor, not to have a good draft of the first chapter of their thesis. The way the funding formula works right now, there’s no incentive for a university to kick a student in that situation out of the program. With a voucher system, the people awarding the vouchers might have an incentive to say to that student “nope, time to move on.” And if the supervisor and the supervisor’s university risked losing cash if the student failed to make progress, there might be a bit more effort made to, e.g., give the student rapid and effective feedback on their research, guidance in finding a topic, etc.
Frances, yes, I realized that point. It would definitely increase the capacity of the dept to take on more “viable” students, and probably “increase quality” and perhaps even quantity of graduating PhDs in the long run.
But it is not much different from tying grants/continuation of student loans to minimal grade achievement, and there have been stories (as part of the discussing the US student loan debacle) of how students were curve-graded out of the program, or out of their funding, and ended up with debt, or money spent for nothing.
That is probably more an issue with curve grading than anything else, but the same thing can happen with vouchers – an institution may still take on “unviable” students (though nobody will know upfront who is viable or not), and let them coast for 2 years while taking the fee, then replace them with the next piece of cattle with voucher attached (you said “stable” first), etc. That a replacement candidate exists will be assured as long as there is external pressure (or “incentive”) to have a PhD degree, and the institution still graduates enough “quality” PhDs to maintain a certain reputation and (signaling?) quality of degree.
Tangentially related, I know an anecdote where a department was offered a relatively prestigious third party project+funding, but with the under the table stipulation that the project leader’s fraternity protege (who worked in the partner organization) gets an easy ride to a PhD. They took the deal. Even not-for-profit institutions will be ready to compromise their supposed mission to get additional funding, of course because the primary funding is not sufficient to carry out the mission adequately – most stuff anywhere always costs more to do well than the decision makers are willing to pay for. In this case, like all other third party funding, I heard a lot of talk how the uni cuts out a sizeable percentage of the money to fund general operations.
The same will happen with PhD funding – a significant part of the money will be used to prop up the general funds of the dept/uni – whether the student completes or not.
In fewer words, this is not an argument against vouchers, but that vouchers may not solve the “problem” of students not completing the program, as they will still (have to) pay for efforts made, for a while. It is like paying by the hour vs. by the piece of completed work – it does change the incentive structures, but most people and institutions would probably prefer to be paid by the hour, the more so when completion or effort per work units is uncertain, unless they have no other choice.