Drummond on Higher Education (An Ontario Post)

Since it seems that all policy decisions now have to be justified on economic grounds, I suppose it made sense for the Ontario government to turn to a former banker and current economist for advice on the future direction of the province. The 600-plus-page document Don Drummond produced in response has been extensively covered in the news since its release last Tuesday, and I can’t competently comment on many of its recommendations. However, an entire chapter of the report deals with post-secondary education in Ontario — and that’s a subject on which I have a thing or two to say.

The Good

My first reaction, and this is an indication of how low the bar is set these days, was relief: the report recommends slight funding increases for higher education (although it anticipates that enrolments will outpace the funding growth it recommends, so in effect, Drummond accepts decreasing funding levels). The per-student grant the Ontario government provides to universities has remained stagnant for years, not even keeping pace with inflation (which is to say, it has shrunk year for year in real terms), so any actual growth is good news of a sort. And in the context of a report that suggests some of the most brutal cut-backs this province has ever seen, it looks like the higher education sector got off more lightly than we might have expected.

As a humanities scholar, I also was heartened that Drummond insisted on linking the notion that education is a path to economic success to a set of intellectual or educational ideals many of us would consider core concerns of the humanities:

In this knowledge-based era, education and innovation will be the key for Ontarians to be prosperous. But for Ontario to succeed in the competitive global economy where many other jurisdictions are improving their educational systems, Ontario graduates will need much more than a simple handle on facts and figures. The value added by a PSE must increasingly be the ability to think critically, to express those thoughts clearly, and to adapt and apply knowledge to new areas and tasks.

While the language of “added value” remains alien to me, and I imagine to most other humanists, the idea that critical thinking, clearly expressed thoughts, and intellectual agility are central to the pedagogical enterprise could hardly be controversial in my field — and I would argue that these principles are more fully realized and of greater structural importance in the undergraduate teaching methods of the humanities than in those of most social or natural sciences. In other words, Drummond’s report, despite its overt focus on economic and fiscal issues, implicitly emphasizes the importance of subjects not commonly considered of the greatest economic value or significance. That’s a very welcome shift, and an attitude strikingly different from that recently taken by another former-capitalist-turned-policy-advisor, the UK’s Lord Browne.

The So-So

The report also calls for greater accountability:

The role of government should be to negotiate and articulate a clear set of expectations for each institution, informed by the university’s or college’s strategic plan. This plan should outline how the institution will achieve improved quality and establish a performance measure of that expectation; it should also modify funding based on the degree to which an institution succeeds in meeting the expectations.

In principle, it would be hard to argue against such a notion. Of course publicly funded institutions should be expected to demonstrate that they are putting that funding to good use — although what that means might well differ from university to university. Drummond takes such institutional differences into account:

Setting outcome targets based on the individual mandates of each institution is integral because it is unreasonable and potentially unproductive to expect all institutions to deliver the same results. For some institutions, government might bias the performance measures towards research output and productivity. For others, the performance matrix might be biased to excellence in undergraduate teaching. For still others, regional economic development would take on greater importance.

As the report’s rhetoric makes clear, however, the seeming equivalence of “negotiating” and “articulating” expectations is likely, in practice, to be replaced with an expression of governmental “bias” or preferences — individual institutions’ strategic plans may inform the government’s decisions, but ultimately, in the report’s vision, the provincial budget will determine the academic direction and purpose of Ontario’s universities and colleges. I am on the fence as to whether this would be a disaster or a boon, but then I am responding from the extremely privileged position of someone employed by a university whose status as a research-focused institution is unlikely to come under attack any time soon.

That said, parts of Drummond’s assessment of why there is a problem with the near-universal drive towards research ring fairly true to me:

As federal government support for research tripled between 1997 and 2003, universities responded universally. Most universities (and, more recently, some colleges) have elected to pursue the myriad federal and provincial research dollars available, all in the name of becoming “world-class research centres.” The reality is that few of Ontario’s research centres will become the best in Canada, never mind the world. Many, however, have gone so far as to apply cross-subsidies within their institutions, effectively taking money from undergraduate tuition revenues to further support research. Increasingly, institutions are allowing professors to sacrifice teaching commitments to conduct more research, as has been noted in books such as Academic Reform. Clearly, there must be a better balance; excellent research should not trump excellent teaching.

On the one hand, turning to “research” as the category through which to define an institution’s identity and status makes sense. For one thing, research excellence is much easier to assess than pedagogical strength; it is a goal for which metrics can be developed more readily, and relative success or failure can be documented more credibly than in teaching. How exactly, after all, could any university demonstrate that its pedagogical achievements trumped those of another? Retention rates might be one significant measure, but beyond that, neither student satisfaction surveys nor grade-point averages seem particularly reliable (is a satisfied student necessarily one who has been taught well? Is a student with good grades within the context of University A necessarily “better” than a student with worse grades within the different context of University B?). Contrast the relative clarity of research records: grants received; articles and books published; citations; appearances at national and international conferences — all of these can be documented, compared, and ranked (even if such assessments still remain crude, distorted, or inherently biased). That doesn’t excuse, of course, the hyperventilating discourse now commonly associated with research goals. Drummond’s skepticism about the extraordinarily capacious category of world-class excellence is refreshing — especially given the inflationary use of such ideals in recent government efforts to justify university funding elsewhere (the situation in Britain comes readily to mind: Simon Head’s analysis of the fetishization of “world-class” research and its supposed impact remains as devastating now as it was when it was published last year).

All the same, the report’s analysis is hardly flawless. Its potted historical narrative of how research became the universal measure of a university’s status is inadequate, too narrowly focused, and probably wrong. More troublingly, its interpretation of how university funding works is disturbingly off the mark: “Many, however, have gone so far as to apply cross-subsidies within their institutions, effectively taking money from undergraduate tuition revenues to further support research.” In the context of a system almost entirely composed of public universities, one might wonder where exactly Don Drummond expects funds for research to come from. Ontario institutions rely predominantly on two sources of revenue: the government’s per-student operating grants and individual students’ tuition payments. Income from endowments varies widely from school to school, and has been severely depressed in recent years. Given the near-negligible research support on the provincial level, especially as of this year, and the extremely limited federal dollars in support of research (outside those distributed through the Tri-Council system), universities hardly have a choice but to redistribute the revenues generated through undergraduate and graduate enrolments.

Nor is there anything unusual or untoward about this practice. Even without any research activity, universities would shift funds between units and departments; cross-subsidies are simply an unavoidable consequence of regulated fees. Students in an English course are much cheaper to teach than students in any class that requires a lab — the space, equipment, and materials required for the latter as well as the higher personnel costs associated with running labs make this inevitable, even if faculty salaries in the humanities were at the exact same average level as in the sciences (they’re not — Canadian humanities professors make on average $10,000 to $12,000 less than their colleagues in the natural and social sciences). Since undergraduate pay the same fees per course across most academic disciplines, however, those in “cheaper” courses necessarily overpay, while those in more expensive classes are subsidized. As scholars such as Christopher Newfield have shown in great detail, the humanities and the social sciences almost always operate at a profit, allowing universities to cross-subsidize costlier fields.

This is even more the case when it comes to research. Research in the humanities, more often than not, is incredibly cheap. It may involve travel costs, sometimes for a couple of research assistants as well as the professor; it may involve modest equipment costs (a few thousand dollars for new laptops, software, and fancy scanners and cameras perhaps?); it may require modest expenses for individual book purchases, and equally modest costs for library acquisitions (we are always complaining, and rightly, about the cost of many monographs — those $100-$150 books from Cambridge or Oxford seem awfully expensive. But perhaps we should make more of the fact that one of those books still costs less than 3% of the annual subscription to a single Chemistry journal. And that the average annual cost of a single science journal, at $2263, would cover the subscription to over seven humanities journals, which cost an average of $299 a year. Yes, monographs and periodicals are more expensive than they used to be. But they’re a bargain compared to the cost of publications in other disciplines). The priciest aspect of research in the arts is time: nothing is as precious to researchers in these fields as the freedom to sit in an archive for months, to sit in a reading room, or to sit at their desk, reading, thinking, writing — or to run off to the occasional conference, to listen, think, and talk. And even that luxury isn’t all that expensive: at most, the cost would be one’s annual salary and the expenses for a temporary replacement.

However, in Canada time is extraordinarily difficult to come by now. The major federal funding body, the Social Sciences and Humanities Research Council, hardly finances release time anymore, so most professors depend on periodical leaves (at reduced pay) or external fellowships, all highly competitive and impossible to plan for. Thus, in reality, our research costs the universities almost nothing.

One might argue that the customary split of a professor’s time between teaching, research, and service allows for teaching to subsidize research — and that’s undoubtedly true. But it’s less true, again, in the humanities, where the average teaching load is typically higher than in the sciences. And it’s easy to overestimate the cost of these subsidies. Exactly how much could the average academic teach, if research were completely out of the picture? The traditional 40/40/20 ratio suggests that combining research and teaching time should allow for a doubling of one’s course offerings, but that leaves no time to keep informed about developments in the discipline (an obligation integral to the active researchers’ job, but not necessarily to that of a teacher) — it would mean forcing teachers into obsolescence. A more reasonable model might posit a 50% increase in individual teaching loads, which in practice would mean an additional class per term. The savings from that would register, of course, but I’m not convinced that the opportunity cost (giving up on research altogether) is worth it.

And yet, despite all these caveats, Drummond’s key recommendation to reward excellence in teaching is still heartening, if only because it avoids the pitfalls of the British system: tying funding predominantly to research excellence both leaves some universities drastically underfunded and empties out the category of excellence; worst, it essentially declares the teaching of undergraduates to be of secondary importance to any university’s mission. That teaching and research have to be equally worthy of financial support may seem like an obvious point. Astonishingly, it isn’t — and I am glad Drummond made it.

The Ugly

The only problem is that it’s one thing to say that excellent teaching should be funded; it’s an entirely different thing to suggest how that would work. How is it possible to achieve teaching excellence without significant revenue growth? Drummond leaves that question almost entirely unanswered. As I argued above, in many fields a drastic increase in individual teaching loads is simply not a viable option. The report explicitly praises the Ontario university system’s current efficiency and notes that the system managed to absorb a “36 per cent increase in enrolment since 2002–03” without significant growth in government funding. However, Drummond also knows the price of that efficiency: “more sessional instructors, larger classes, fewer projects, less contact with professors and so on.” In other words, those of us lucky enough to be permanent employees of our institutions teach more students per class than in the past, we can offer them fewer opportunities for hands-on experience in the field, and we encounter them less and less as individuals. Now, if greater cost effectiveness were Drummond’s only objective, we could simply keep going on this path, while reducing our commitment to research as well. That might mean teaching three rather than two fairly large classes every term. Given the reduction of research activity, we would necessarily have to make do with fewer graduate students, and thus less TA support; consequently, our individual workloads would shoot up, with much higher grading loads than in the past (or alternatively, we could reduce the number of assignments). In every way, the model would produce pedagogically inferior outcomes: more large classes, much less individual attention, less writing instruction, more overworked faculty with less time to devote to students on a one-on-one basis. The province would undoubtedly lose many of its most prolific scholars as well. But we would process more students.

To Drummond’s credit, this is not his vision. In fact, class size is one indicator of educational quality his report mentions more than once — if he wants to increase classes in primary and secondary schools, his idea of pedagogical excellence at the post-secondary level is closely tied to smaller classes.

And this is where the report’s analysis, and its proposals, come off the rails. On the one hand, the document anticipates further enrolment growth; on the other, it does not call for commensurate increases in funding. On the one hand, it thinks professors should teach more students; on the other hand, it does associate pedagogical excellence with smaller classes. And those assumptions can’t be reconciled. You can’t have more students and a better educational experience if the growth in faculty complements does not keep pace with enrolment growths — it’s not even possible to maintain the inadequate status quo under those circumstances. If we want smaller classes (but more students), we will need to increase the number of professors significantly. And cutting research out of the picture will in fact worsen the situation, as it will necessarily lead to a loss of graduate instructors and a further increase in professors’ teaching responsibilities. Let me illustrate what I mean.

Here’s a hypothetical scenario — made up but fairly representative, I think, of the average Ontario humanities department. Professor A, who’s currently not teaching a graduate seminar, is offering the following four courses:

That’s a very simplified counting of hours, with very limited allocations of prep time, but it already exceeds 40% of a regular work week (then again, A will of course not be teaching in the summer). We might also wonder what exactly a regular work week is for a professor. Currently, there is, in theory, no limit to the number of hours we work in any given week — if the research demands it, weekends and free evenings easily become negligible conventions. But with the personal connection most professors feel to their research gone, what happens to perceptions of fair workloads and hours?

Now let’s assume this same professor’s department jettisons its research activities and increases faculty workloads accordingly, setting aside 20% of a faculty member’s time for maintaining a familiarity with the field, and an equal amount of time for service — thus leading to an increase in teaching duties by 50%. Such a department could not, of course, maintain a PhD program and would thus lose TA support.

Now forced to do all the grading and provide all the writing advice and instruction normally covered by her TA in her lecture courses, this professor’s weekly load would exceed 60% of a standard 44-hour work week immediately, even if her course load remained at 2.0 FCE (or 2/2). But since these figures are based on a two-semester model that excludes the summer, and since this instructor will not be expected to conduct research, the summer term now becomes available. Perhaps she could offer an additional large lecture course then, in the more compressed schedule typical for summer classes, with the lower weekly averages balancing out the overload from the first two terms:

However, while this model shows that a professor without research duties could absorb the extra hours resulting from a lack of TA support, and could offer a small number of extra classes if his or her teaching load were spread over an entire year, it also illustrates that this additional load does not allow for any reduction in class sizes compared to current levels. Our hypothetical professor teaches an average of 97 students a term, with only one seminar in the mix. If her department committed to Don Drummond’s desired small-class pedagogy — doubtless the format best suited for instruction in writing, in critical thinking, and in close analysis — that number would plummet quickly:

The smaller classes take up less time, since they take up fewer contact hours and less grading time. But they require more intensive preparation — unlike lectures, which can be delivered in more or less the same format year after year, smaller, seminar-style classes need instructors prepared to take the discussion in unpredictable directions; in a sense, every time such a class is taught, the instructor will think of it as a new course. But even without those considerations, smaller classes are necessarily far less efficient than large ones: the same in-class and preparation hours as for a larger course serve a much smaller number of students. Thus, although our hypothetical instructor now spends over 30 hours of her week in Term 1 and almost 35 hours a week in Term 2 teaching undergraduates, she only reaches an average of 60 students per term — a reduction of 38% compared to the previous model. It is also worth noting that even if the hypothetical department could somehow maintain a PhD program, there would be no sensible role for TAs in classes this small; graduate students would thus have to be given their own courses to teach (with the result that a significant number of undergraduates would receive instruction from less experienced teachers than under the old model).

In sum, while there may be room for finding “efficiencies,” that mission is by definition at odds with the goal of lowering class sizes and focussing on pedagogical excellence. Better teaching is more expensive teaching, and there is only so much teaching one person can be asked to do before quality declines. There are many other issues I’m leaving unaddressed here. My point is simple: the three assumptions of the Drummond report — higher enrolments, stagnant funding, a need for improved teaching — just cannot be reconciled.

The Puzzling

Finally, I’m uncertain exactly how the report imagines the government will determine how much teaching and how much research any one university will be expected — and ultimately, be allowed — to do. If funds are tied to, e.g., specific class-size targets and teaching loads, what would happen if a university failed to meet those criteria? Would funding be cut? Revoked? Or increased (to allow the institution to improve its staffing levels, say)? Conversely, if a university were expected to maintain a certain level of research prominence, what would that mean for its teaching goals? If some institutions will be defined as striving for pedagogical excellence, will others be allowed to offer merely adequate teaching, while focussing more clearly on research? Or would the new system lead to very lopsided funding levels, essentially stripping some universities of the financial power to conduct research, while leaving others able to manage both research and teaching?

Assuming that the latter would be the most likely outcome, the report’s actual recommendation is not the creation of a two-tier system as such, but to concentrate university funding on pedagogy — quite literally the opposite of the post-Browne UK system. Drummond is in fact quite clear about this, suggesting that the province should “refocus resources and rewards towards teaching in post-secondary institutions” and “refocus provincial funding to reward teaching excellence.” Pedagogy will thus be the central mission of all Ontario universities. Research, on the other hand, will become a more privileged undertaking — not exactly a luxury, but somewhat more of a side-show.

I’ve been trying to wrap my head around what this might mean in practice. It would certainly reduce the overall research activity and output in the province. That may or may not be a good thing. It would mean that some of us, especially in the sciences, would have to spend more time in the classroom — though probably not everywhere. Old hierarchies would be reinforced; but some new hierarchies might also develop. Ideally, I suppose, some institutions would achieve prominence for offering the best undergraduate education — a prominence not coupled to a reputation for research excellence. The question, however, is whether such an ideal is attainable in practice; whether, in other words, fairly large public universities can be turned into something more like large liberal arts colleges.

Further, there is the question of how significant savings could be realized by such a refocussing. Would it mean shutting down and dismantling research labs, or turning them into teaching spaces? Decommissioning parts of library collections? In any case, the fiscal rationale for a turn away from research really only applies in fields where research comes at a significant cost — in other words, the natural sciences. However, that research is also easiest to sell to the public, as its practical value is least difficult to demonstrate. The scholarly work of humanists, cheap as it is, is also the toughest sell, no matter how little of a drain on university resources it actually constitutes, and no matter how little would be gained in practical terms if humanists spent less time in their studies and more in the classroom. And finally, there is the obvious question if such a strategy would not lead to a brain drain that would lessen the quality and reputation of Ontario’s university system, for no very good reason at all.

Ultimately, there is a simple answer to all of this. The report accurately identifies the features of an excellent education. It authoritatively projects the development of Ontario’s student population. But it also assumes that professors spend so much time conducting research that a mere refocussing of our attention on pedagogy would lead us to teaching excellence — that, in other words, universities currently allocate their instructional resources inefficiently, not that universities are massively underfunded. There is very little evidence that these assumptions are at all justified.