A few people have asked me why I haven't blogged about the census controversy.  The main reasons are that I think the subject has been very well covered elsewhere and that I'm pretty sure with respect to my own readership that I'd be preaching to the converted ... you are converted, aren't you?  

Good data is crucial for effective social policy; without it you're running government based on anecdotes.   To give you a sense of how goofy policy-making can be when you don't have good data, I'd like to relate a story from more than a decade ago: how the student debt-scare of the late 1990s came into being.

You may remember that back in 1994-95, Lloyd Axworthy had an ill-fated attempt to reform social security, which included some proposals on student loans and transfers to provinces for post-secondary education.  The exercise was eventually canned in the face of the much more massive redesign of social policy that was imposed government-wide by Paul Martin's 1995 austerity budget, but not before Axworthy and the Liberals had suffered all manner of abuse from students and faculty for the fascist crimes of suggesting that the federal government might give out less in transfers to government and more in transfers to individuals and also that student loans might be income-contingent.

(Fast-forward fifteen years, and we find that the Axworthy scheme has been adopted almost in full without protest.  By re-packaging the voucher idea as an eight-fold increase in education tax credits and re-labelling income-contingency in student loans as a "Repayment Assistance Plan", the Government of Canada has now introduced virtually everything foreseen by the Axworthy plan, sometimes with the backing of the very student groups who pelted the man with macaroni all those years ago.  But I digress).

Anyway, come early 1996, with power heading to the provinces generally and the Liberals convinced that there was no way that the post-secondary file would ever be a political winner, the government of Canada was about ready to dump the Canada Student Loans Plan (CSLP).  They even went so far as to send an ADM round to each of the provinces to inquire how they might divest themselves of CSLP completely and make student aid an exclusively provincial affair. Needless to say, this alarmed the PSE-lobbying crowd in Ottawa.  But the brighter among them recognized some of their own culpability; public internecine fighting over a range of files (tuition in particular), had made it impossible for government to create a policy that could be seen as a "winner" by all stakeholders. 

It was Robert Giroux, then the President of the Association of Universities and Colleges of Canada, who made the astute call that in order to keep the Government of Canada in the game, the PSE community had to give the government a "win" on this file. That meant finding some policies that everyone could agree on, which seemed unimaginable at the time. 

The answer to this problem was miraculously provided by a pair of publications which came out in 1995 and 1996.  The first, a publication from Washington's Institute for Higher Education Policy, called College Debt and the American Family, was part of an overall re-positioning of the student aid debate in the U.S. from banging on about the decreasing purchasing power of Pell grants to focus on debt levels (i.e. from inputs to outcomes).  Helpfully, it contained information detailing college debt by institution type, and showed that at four-year private institutions, the average was about $20,000 (Canadian) at the time. 

The second publication was the Government of Canada's consultation paper for the Ministerial Task Force on Youth, a body which spent much of 1996 touring the country talking about what, at the time, was perceived as a persistent youth unemployment problem.   An innocuous-looking sidebar in this document's chapter on student assistance read, "According to Statistics Canada, average student debt among bachelor's graduates in 1990 was $10,000.  In 1996, it is $17,000 and it is expected to rise to $25,000 by 1998."

This, to put it mildly, was hooey, and not just because Statistics Canada doesn't do forward projections on student debt.  At the time, it was practically impossible to accumulate more than $10,000 in debt as an undergraduate in Quebec or $24,000 in Ontario.  So how could the national number be so high?  The answer was that the $25,000 figure was essentially a made-up number.  First, it excluded Quebec.  Second, it was an extrapolation of the growth curve of borrowing under the federal loan program, which took no account of how any of the provincial grant or remission programs worked; rather, it assumed that growth in provincial loans was matching the growth in federal loans. But, having published the figure, the government felt unable to disavow it.  In actual fact it is only now, 12 years later, that we are approaching that $25,000 figure, and in real dollars we are still some way off it.

You can probably see where this is leading.   Students and university presidents might not agree on tuition fees, but they can unite behind the idea that student debt is a bad thing.  Once PSE groups started to point out that, according to the government's own figures, Canadian students were graduating with more debt than students from Harvard and Yale, the Chretien government was shamed into paying attention to the file (there is simply no easier way to get Liberals' attention than to make unflattering social policy comparisons between Canada and the U.S.).  Thanks to some sustained lobbying, the fact that demographics were pushing post-secondary education higher on the public agenda (1996 happened to be the year that that first "baby boom echo" kids turned 18) and a remarkable turnaround in public finances, the Government of Canada did a complete U-turn on student financial assistance.  Far from abandoning it, they embraced it to the tune of several billion dollars: improving student loans, creating the Canada Millennium Scholarship Foundation and the Canada Education Savings grant and juicing education tax credits.

All because of some bad data.

Now, it's true that much good came of this money.  But it's equally true that this money could have been spent a lot more effectively if there had been some basic data available to answer some common-sense policy questions, like, "How much debt is there, really?", "To what extent do finances generally, and debt specifically, affect access to post-secondary education?" and "Do tax credits have any effect, whatsoever, on anything"?

What does all this have to do with the census?  Not much, specifically.  But the census isn't the only portion of Statistics Canada's budget that's been tampered with.  Much of the funding for data collection related to students has either been cut (the five-year Follow-up of Graduates) or allowed to lapse (the Youth in Transition Survey).  And that's on top of the disappearance of the Millennium Scholarship Foundation's research program, the Canadian Policy Research Networks and - to a lesser extent - the Canadian Council on Learning, as well. 

The Conservative government has a done a great deal in terms of post-secondary funding: research and infrastructure funding, the Canada Student Grants, the dedicated PSE portion of the CST transfer, the tax exemption of scholarships, the changes to immigration law to allow foreign students to work here more easily, etc. They cannot be faulted for inattention to the sector and their generosity to the sector should not be in question.  

But it is equally true that they have stood watch over an evisceration of our collective ability to monitor problems and evaluate policy in post-secondary education.  For a government that came to power preaching careful stewardship of public finances, it's a baffling course of action.  Bad policy decisions are only ever one piece of bad data away; Conservative policy on data collection is guaranteed to increase the frequency of these bad decisions.

This week's column is about tuition and its effects on access and retention.  But before doing so, I would like to take a short detour through the field of tax studies.  Bear with me, it'll be worth it.

In 1998, University of Michigan professor Joel Slemrod edited an intriguing book entitled Does Atlas Shrug?  Playing on the title of the interminable Ayn Rand novel, Dr.  Slemrod and some associates tried to figure out whether or not taxing the rich really did make them "shrug" (that is, give up the risk-taking, go-getting behaviour that makes advanced market economies tick). 

To the extent that the studies could determine anything - getting a decent natural experiment for this type of study is quite difficult - they tended to show that risk-seeking go-getters would generally engage in risky, go-getting behaviour regardless of the tax rate.  They might have made less money and the economy might have been less efficient than if taxes were lower, but risk-takers tended not to change their behavior much in response to changes in the tax code. 

This was an intriguing result, because it undercut much of the rhetoric used by the anti-tax movement.  Taxes cannot be raised, they say, because investment will go elsewhere, or because talented individuals will begin to prefer leisure to work and thus deprive the economy of part of their services.  And yet, the Slemrod results show that this isn't necessarily true.

Of course, this doesn't mean that tax rises are pain-free.  After all, when taxes rise, everyone feels an immediate loss of welfare (though this may be partly off-set by the public goods purchased with these tax dollars), and dead-weight efficiency losses remain significant. But what the Slemrod results show is that a loss of welfare does not necessarily entail a change in behaviour. 

What does any of this have to do with higher education?  Well, interestingly, the kinds of arguments that Dr.  Slemrod countered in his book are exactly the same as the ones that underpin the argument against tuition fees. 

Tuition hikes, like tax hikes, clearly make students and families less wealthy, leaving them with less disposable income.  Indeed, as higher education edges closer and closer to universality, tuition for most people does start to look increasingly like a tax - a universal service-charge for education past the age of eighteen.

And just as advocates of lower taxes say that Atlas will shrug if welfare is diminished through tax rises, so advocates of lower tuition say that Atlas will drop out if welfare is diminished through tuition rises.

But is the proposition that changes in welfare cause behavioural effects any more true for tuition than it is for taxes?  Or is this argument just a smokescreen for what amounts to a middle-class hand-out?

The best available evidence from international experience shows that in fact rises in tuition have very limited behavioural effects.  At a jurisdictional level, there is simply no observed correlation between tuition fees and any of the normal indicators of access. 

Take, for example, the following evidence we've come up with at Higher Education Strategy Associates:

University Participation rates are uncorrelated with tuition levels:  Within Canada, Nova Scotia - the province which for thirty years had the country's highest tuition fees -has consistently has the country's highest participation rate while low-cost Quebec has a participation rate below the national average.  Internationally, both Canada and the US - both relatively high-tuition jurisdiction - have higher university participation rates than most no-tuition OECD jurisdictions.

Changes in participation rates are uncorrelated with changes in tuition levels: Ignoring Ontario, where the "double cohort" distorts the data, the two biggest jumps in enrolment in Canada in the early part of this decade were in Manitoba (29% increase) and British Columbia (33% increase).  The former saw tuition levels drop by 20% in real terms while the latter saw an increase of nearly 75%. 

The stratification of the student body is uncorrelated with tuition levels:  The Educational Equity Index (EEI) measures how "elite" a student body is by comparing students' parents' educational levels to those of the general population.  In Canada, the two least-privileged student bodies were in low-tuition Manitoba and high-tuition Ontario; the two most-privileged student bodies were in low-tuition Newfoundland and medium-tuition Prince Edward Island.  Internationally, Canada ranks among the top five countries for having the least-privileged student bodies, along with free-tuition Ireland and Finland, low-tuition Netherlands and high-tuition Canada and the United States. 

Changes in the stratification of the student body are uncorrelated with changes in tuition:  Even after adjusting for inflation, tuition has more than doubled since 1986.  Yet in a series of Statistics Canada studies, from 1986 onwards shows exactly the same pattern: youth from rich families are about twice as likely to go to university as youth from poor families.  Despite all the tuition policy changes, that relationship has remained very consistent over time.

In sum, there is very little international experience that should make us believe that the reduction of tuition fees would have much effect at all on the participation rates of youth from poorer family backgrounds.  The reason for this is bindingly obvious: kids from poorer families aren't just disadvantaged in financial assets, they also have smaller pools of social and cultural capital on which to draw.  This means they are less likely to be able to get the grades to go to university and they are less likely to want to go on to university in the first place. Finances are thus at best a third-order cause of inequality of access.

None of this means that financial burdens have no effect on students.  Roughly one-in-five Canadian youth who do not go on to post-secondary education cite finances as the prime factor behind their decision and a majority of these are from lower-income backgrounds.  Various American studies have shown that reducing the net cost of higher education (that is, the cost of tuition minus grants) also increases access to higher education for students from lower-income families.  And some data that will be coming out soon from the MESA project (an effort I have been involved with for the last five years) will also show that changes in net tuition - that is, tuition minus grants - may play a role in altering the composition of student demand.

In other words, finances do matter for some students, some of the time.  But that suggests that the right policy response is a program of targeted grant assistance to low-income students.  It does not imply that reduced tuition for all students makes any sense at all.  Indeed, the evidence that such a policy would increase participation from low-income families is slim to nonexistent.  In the jurisdictions where tuition roll-backs have been enacted (Manitoba, Newfoundland, Ireland) there is no evidence whatsoever that the social composition of enrolments have changed.

So, do tuition hikes cause Atlas to Drop Out?  On the whole, and assuming there is a decent system of student aid available, no.  But they do mean that Atlas (and her parents) will have a little less money to spend on life's luxuries, just as they would if their taxes were increased.

And therein lies the key to understanding the real reluctance of Canadian politicians to raise tuition fees.  The middle class - for whom university education is increasingly as normal as primary school - sees fees as a tax which reduces their spending power and reacts to them accordingly.  Concern about the ability of the poor to access PSE has nothing to do with it and is actually something of a smokescreen.  Reducing tuition is simply the big-government equivalent of middle-class tax cuts - no more, no less.

If you're looking for some summer reading on higher education (and, really, who isn't?), I'd like to recommend to everyone a recently-published book entitled The Great Brain Race: How Global Universities are Reshaping the World, by the Kauffman Foundation's Ben Wildavsky. It's a good, breezy introduction to the kinds of issues that senior administrators are pondering with respect to internationalization and what has come to be known as the "world-class university movement."

The book tackles a number of inter-related themes about the "global university."  The vast expansion of educational opportunity in the developing world, and the rapid modernization of China and India have broadened both the supply of and the demand for highly-qualified personnel and educational services. Universities, which for the past two centuries have regarded their role as being handmaidens to state power, have, in response, suddenly started to revert to their medieval role as ecumenically transnational centres of learning.  The shift has been jarring and sudden, and it's still not clear where the transition is leading.

Wildavsky brings a refreshingly jargon-free approach to his examination of this shift.  He is not a theorist; if you're looking for some more overarching understandings of globalization and knowledge, go read Manuel Castells instead. This is much more a magazine-style of writing (as befits his background as a former editor at U.S. News & World Report); six 30-page chapters, each focused to a large degree around a couple of people or institutions doing something that he believes is a harbinger of a larger coming trend. 

This approach of focusing on "the next big thing" means that Wildavsky spends most of his time looking at outlier institutions rather than the ones most professors and students inhabit.  For instance, New York University's attempt to turn itself into a global entity with a number of co-equal campuses around the world - which takes up most of the first chapter - doesn't herald the arrival of chain universities with globalized brands.  But if NYU succeeds - which is far from a sure thing, given the risks and money involved - it would change the face of international higher education and there will certainly be imitators.  It would also permanently alter the way the world sees NYU; from a slightly parvenu second-tier Ivy League school to a visionary institution with a first-mover advantage in a whole new category of world super-institution. 

Fundamentally, this book is about how students and professors have become more mobile, how institutions have come to see themselves as being in competition for these students, and how they are altering their behaviour in order to improve their chances of winning these competitions.  But, despite Wildavsky's repeated use of the market language of "competition," it's worth remembering that universities do not really compete with one another the way companies do.  When companies compete, they are doing battle for market share and, ultimately, profits.  This is not even vaguely true of universities; in countries like Canada, few genuinely wish to increase their enrolment and none are legally entitled to show a profit or distribute earnings to shareholders (of which they have none, in any case). 

Rather, the competition is simply for prestige.  This is a very important concept within universities; prestige being in fact the main currency of academia, both at the level of the individual academic and the institution as a whole.  But it's an odd currency, in that it is difficult to count and measure at the level of the institution.  Global rankings, in this sense (the subject of another chapter), serve a useful purpose within academia by bringing a certain transparency to the measurement of prestige, thus allowing people to see who is "winning" (the charge that rankings "created" competition for prestige between institutions is, in North America at least, nonsense - it's always been there, but wasn't acknowledged openly). 

But what does an institution do with prestige once it has it?  It hoards it, basically, like Smaug sitting on his treasure pile.  What many people fear is that prestige is being concentrated in fewer and fewer hands; this, in part, is what has led governments around the world to start creating multi-million-dollar "excellence" programs designed to concentrate talent at a few domestic universities so that they can "compete" on this world stage.  As we saw here in Canada during the G5 (or "Big Five") debacle of last summer, there's a genuine debate to be had about whether or not this is a good thing (for what it's worth, I think there are merits both to the concentration and dispersion of talent, but there's very little evidence to suggest one is much better than the other), but Wildavsky, unfortunately, declines to enter into this larger debate.

This is a symptom of what, unfortunately, is this book's greatest deficiency.   While it's a great gee-whiz tour of the leading edges of neat stuff going on in academia on four continents (neither Africa nor Oceania get much of a look-in), it lacks a "so what" element.  So what if a few institutions have gone global?  So what if there are some institutions that are world-class and others not?  Does it matter either for institutional or public policy that this stuff is going on?  After all, most of the challenges this book outlines are very far removed from the lives of most professors and students.   Most universities are not really in a global competition for talent, either at the student or faculty level.  Even among those that are, the concrete nature of the challenges of such "competition" actually affect very few people beyond the vice-President, some deans and a few student affairs types.  At a day-to-day level, few, if any, students and professors will perceive themselves to be participants in any kind of global competition.  So why should any of this stuff matter to them?  Wildavsky doesn't really take the time to say.

The real answer, I think, is that, as long as prestige matters, then university presidents' central mission is always going to be to obtain as much prestige as possible for their institutions.  What Wildavsky is really documenting is that the rules for obtaining prestige have changed - essentially, if you're not internationalized, you're dead.  And it is this change in the rules of prestige that are driving such profound shifts in institutional behaviour.

The final element of the book that I think is really worth highlighting here is its chapter on for-profit higher education and how a few (primarily American) companies are making millions of dollars through the provision of education in Latin America and Asia.  It's easy, of course, to pooh-pooh for-profit education; usually, if the choice is between for-profit and free public education, the nod should go to the latter.  But while Wildavsky doesn't pull punches on issues of quality control, he also gives the sector its due - in much of the world, the choice isn't between private and public, it's between private and nothing.  In the way this chapter gets down to the hard brass tacks of the economics of delivering education in distant places, it's perhaps the book's most intriguing passage. 

This book won't be for everyone; the lack of a strong integrative synthesis will grate for some, as will a lack of reflection on the downside of the war for prestige for those institutions left behind.  But as a peek into fast-evolving trends in global or transnational education that are increasingly consuming the minds of university presidents, it's extremely good, and its educated-but-breezy Atlantic style of writing, so rare in higher education books, make it a pleasure to read. 

(Editor's note:  you may also be interested in a transcript of a speech Wildavsky gave at the Carnegie Council, as well as a review of Wildavsky's book by The Times Higher Education.)

Dear Stephen, Michael and Jack,

Hi guys. Look, the next federal budget is going to have to be a doozy for cuts, given our fiscal situation, and everyone understands that Stephen would probably prefer to go to the voters and try one last time to get a majority before lowering the boom. That means it's more than likely we'll have an election in the fall - before October, if the Conservatives wish to avoid answering questions about the Auditor General's report on the effectiveness of the stimulus spending.

That makes this summer platform-writing time for all three of you. As a public service, I'm here to offer you all some free advice when it comes to drafting the bits of your platform related to post-secondary education (PSE) and innovation.

Let's start with the issue of access to education. 

First, scratch anything that vaguely resembles a policy on tuition fees. No credible research shows tuition to have a substantial impact on access, and, in any case, it is entirely a matter of provincial jurisdictions. 

Even policies like paying provinces to reduce fees (NDP policy from 1997 to 2006) or rebating fees to students (Liberal policy in 2006) have weird distributional consequences because they would involve vastly different subsidies to different provinces, based on existing tuition. 

Money in student aid is a slightly more promising idea because at least it's targeted.  But you might want to keep your powder dry on this one. Student aid costs are already set to rise significantly over the next few years as interest rates rise ... and the jury is still out on whether the recently-introduced Canada Student Grants are going to cost what was originally predicted or if the recession and lax eligibility criteria will blow the lid off costs. 

A closely-related idea to increased student aid is that of giving "completion grants" for apprentices, which if you really think about it, is a massive waste of money. The main reasons apprentices don't finish their training is because they can make too much money without completing (in boom times) or because they can't find get enough work hours to complete (in bust times). Either way, giving completers a couple of thousand dollars isn't going to change outcomes; in good times the incentives of work are simply too big for a couple of thousand dollars to matter, and in bad times, the incentive doesn't mean anything if they can't find work.

If it is access that's bothering you and you want to do something useful, then concentrate on individuals with the greatest challenges in reaching PSE: Aboriginal students and lower-income inner-city youth. And concentrate on the barriers that are most serious: inadequate academic preparation and weak academic motivation (the American GEAR UP program is not a bad model for how to do this in a federal system). These students will eventually need money, too, of course, but existing student aid programs are likely strong enough and generous enough to deal with that, so avoid the temptation to enrich student aid on that account.  Focus on the achievement and aspirations and you'll make a real difference.

If you absolutely must pour money into student aid, do yourself a favour and make sure it's cost-neutral; whatever you stick into student aid, you should take away from education and textbook tax credits. These cost a tremendous amount of money and achieve absolutely nothing.  Everyone knows they only exist because they are a cheap and convenient way for the federal government to shovel money to the middle class without having to deal with the provinces.  Cutting these would offend no one in the sector; if you cut them and redistributed the funds elsewhere in education, you would win plaudits.

Turning to research - well, what can I say? You more or less all agree on the essentials, anyway.  We have a granting council system that works pretty well (some of you seem to enjoy micromanaging the system more than others, but that's as may be), which could use a little more funding but isn't wildly out of whack. Everyone seems to like our human-capital increasing programs - the Research Chairs and the new Excellence Chairs, so don't mess with those.

The Canada Foundation for Innovation is a well-loved method of handing out funds for research infrastructure - though with operating budgets strained, you may find that universities aren't quite as eager these days to expand their operating costs with new buildings.  

The problem here, though, is one of using policy to buy votes versus using it to achieve something useful. We all know you love a ribbon cutting, and everyone in the sector does their best to satisfy you on that. But what the sector desperately needs are things that are unglamorous: more money for research overhead, reducing the paperwork burden on research grants - that kind of thing. If you can show the sector that you are serious about these kinds of things, they'll assume you're serious; if you insist on the ribbon-cutting stuff, they'll know you're just posing. 

But now let me turn to the most important question of all - one so important and basic that the sector itself is reluctant to face it head on. You know that we're facing a demographic time-bomb: The Canadian labour force will stop growing in 2016, while the number of over-65s is set to triple by 2030. This means taxpayers are going to have an increasing burden for pensions and health care. The only way to relieve that burden is to increase the productivity of Canadian workers. 

A lot of the productivity increase is going to have to do with changes in regulatory and competition policies, but it also has to do with education. We need to better understand how curricula can be altered such that graduates can generate the extra productivity we will need to generate the money required to maintain our standard of living. Long term, this isn't just the key issue: it's the only issue. And right now, frighteningly, no one has any idea how to approach the issue.

What I've just described must, I know, be terribly unappetizing for you politically - an issue where there are no simple answers and which, in any case, lies almost entirely within provincial jurisdiction. Nevertheless, any medium-term prosperity plan that doesn't get to grips with this issue is going to be doomed to failure. And the provinces are going to be struggling for answers, too - they may welcome a partner in Ottawa if the approach is handled sensitively.

The best solution here is probably a revival of an agency not entirely dissimilar to the Canadian Council on Learning (CCL).  Not CCL exactly, because that well has been poisoned. What is needed is something with a much narrower mandate (e.g. provides and tests a range of possible educational innovations that can raise productivity and growth rates) than CCL, which includes very significant participation of the provinces and educational institutions.   It should not presume to issue definitive "orders" about what works and what doesn't, but rather try to provide a variety of solutions that might work in different systems.  Finally, it should engage large numbers of stakeholders (especially schools, colleges and universities) in deep debates about the myriad ways in which curricula might be reformed.

In sum, what all three of you need to decide is whether your PSE policies are going to be about buying votes or about actually improving Canada.  The latter is much tougher to do, not least of all because of our country's federal nature.  But if any of you can actually get to grips with these tough problems, you'll get serious respect from the community.

Jack, Michael, Stephen: by your promises shall we know ye.  Don't disappoint us.