Higher Education Strategy Associates

Category Archives: Student Aid

Includes student loans and grants, also scholarships, RESPs etc., and policy relating to these issues in Canada and elsewhere.

May 26

Lessons from the Rise of Tax Credits

I’m feeling low on creativity today, so I’m going to go to that old stand-by: telling war stories. And specifically, I’m going to go back and trace the rise of tax credits in the Canadian higher education system and what that tells us about policy-making in Canada.

Tax benefits for education go back to the late 1950s. There was pressure at the time to create a “national system of scholarships”, but this clearly was going to cause problems in Quebec. But Prime Minister Diefenbaker, on the advice of Ted Rogers and with the assistance of Brian Mulroney, found a way around this which was acceptable to Quebec: namely, by making tuition fees tax deductible. Lesson #1: the federal government in part views tax expenditures as a way to get around troublesome provinces.

These tax deductions for tuition and a monthly “education amount” were turned into tax credits in a general tax reform introduced in 1988 by then-Finance Minister Michael Wilson (which is still arguably the greatest thing any Conservative finance minister has done in my lifetime). The tuition credit did not include ancillary fees and the monthly amount was $60/month. And there it stayed until 1996.

Budget 1996 was not a happy time in Canadian history. As far as most people were concerned, we were in year 6 of a recession (a real one, where unemployment hit double digits and a third of the island of Montreal was on social assistance/EI, not like the past few years). The stomach-churning Quebec referendum night was less than four months in the past. The country was broke, and the logic of Paul Martin’s epoch-defining 1995 Budget meant that fiscal room for anything new was just about zero. Yet the government wanted to show that the federal government could still be relevant, particularly around youth unemployment, which was a concern at the time. So what did they do?

They upped the education tax credit to $80/month.

I know that sounds meagre. Trust me, in the context of February 1996, this was a moderately big deal. But it was the 1959 logic at work again. Need to show the feds can do something about an issue that matters to Canadians but is mostly in provincial control? Use the tax system!

Then in December 1996, the Finance Department’s pre-budget polling (which in those days was always, always, always done by Earnscliffe) numbers came in and they showed – totally unexpectedly – that education was suddenly the number two issue for Canadian voters. Terrie O’Leary, Paul Martin’s formidable chief of staff, immediately went to the office of Don Drummond (now Chief Economist at TD, then the ADM at Finance in charge of the budget). The conversation, the best I can reconstruct it from a couple of different sources, went like this:

O’Leary: I want something on education in the budget.

Drummond: (Acutely aware that the budget date was only about ten weeks away and it’s desperately late to start screwing around with it at this point): Not unless you want a replay of the Scientific Tax Credits fiasco.

O’Leary:  <A string of choice expletives to the general effect of “don’t talk back to me”>.

Well, of course Drummond needn’t have worried because when it doubt: tax credits!  The vehicle was already there, so they just juiced it. The $80/month education amount jumped in stages to $200/month, a smaller credit was added from part-time students, and the definition of tuition tax credits was expanded to include ancillary fees. Bonus: unlike the changes to Canada Student Loans and the Millennium Scholarships which were announced in the following year’s budget, there was no tedious negotiations with provinces. Lesson #2: tax credits are sometimes a tool of choice because they’re easy and quick to implement.

Then of course, the economy improved and Paul Martin started getting generous. In the fall 2000 mini-budget which preceded that year’s election (the Stockwell Day election, in case you’ve erased that period from your memory), he doubled the value of the education amount to $400/month for full time students and $120/month for part-timers.  Why? Well, in the preceding election, the Liberals had promised that any surplus money (and we started running surpluses in 1998), would go 50% to new programs and 50% to “debt reduction and tax cuts” (relative proportions not specified). It finally occurred to the Liberals that under this regime tax credits were gold, because depending on one’s choice of definition, tax credits could be counted as an expenditure or as a tax cut. And yes, they counted these as both, to suit the occasion. Lesson #3: tax credits are attractive because the communications around them are flexible.

That was more or less the high point of education tax credits in Canada. After that, they started to gradually fall out of favour. Quebec (2012) and Ontario (2016) have both abolished their credits, and Budget 2016 saw the feds abandon them in favour of higher grants. I suspect they will disappear from the provincial level over the coming decade.

But the point I want you to take here is not that government was misguided about tax credits back then and is smarter now. Apart from a couple of zealots in the Finance Department who prattle on about tax treatment of human capital, no one in the 1990s genuinely thought that tax credits were a particularly good tool to get money to students. What they had over other more direct means of support was convenience, simplicity, and the ability to be implemented completely independently of what a bunch of tiresome provinces think. In the late 1990s – the High Era of Competitive Federalism – that stuff mattered a lot more than it does today. If those conditions ever return, it would be easy enough to see how tax credits as a funding mechanism could return, too.


May 15

Provincial Budgets 2017

Springtime brings with it two certainties: 1) massive, irritating weekend traffic jams in Toronto as the city grants permits to close down Yonge street for a parade to virtually any group of yahoos, thus making it impossible to go from the cities east to west ends and 2) provincial budgets.  And with that, it’s time for my annual roundup of provincial budgets (click on the year for previous analyses – 2016 2015 2014 2013.  It’s not as bad as last year but it’s still kind of depressing.

Before we jump in, I need to remind everyone about some caveats on this data.  What is being compared here is announced spending in provincial budgets from year-to-year.  But what gets allocated and what gets spent are two different things. Quebec in particular has a habit of delivering mid-year cuts to institutions; on the flip side, Nova Scotia somehow spent 15% more than budgeted on its universities.  Also, not all money goes to institutions as operating funding:  this year, Newfoundland cut operating budgets slightly but threw in a big whack of cash for capital spending at College of the North Atlantic, so technically government post-secondary spending is up there this year.

One small difference this year from previous years: the figures for Ontario exclude capital expenditures.  Anyone who has a problem with that, tell the provincial government to publish its detailed spending estimates at the same time it delivers the budget like every other damn province.

This year’s budgets are a pretty mixed bunch.  Overall, provincial allocations after inflation fell by $13 million nationally – or just about .06%.  But in individual provinces the spread was between +4% (Nova Scotia) and -7% (Saskatchewan).  Amazing but true: two of the three provinces with the biggest gains were ones in which an election was/is being held this spring.

Figure 1: 1-Year change in Provincial Transfers to Post-Secondary Institutions, 2016-17 to 2017-18, in constant $2017

Province Budget Figure 1 Year Change Provincial Transfers


Now, this probably wouldn’t be such a big deal if it hadn’t come on the heels of a string of weak budgets for post-secondary education.  One year is neither here nor there: it’s the cumulative effect which matters.  Here’s the cumulative change over the past six years:

Figure 2: 6-year Change in Provincial Transfers to Post-Secondary Institutions, 2011-12 to 2017-18, in constant $2017

Figure 2 6 year chage in provincial transfers


Nationally, provinces are collectively providing 1% less to universities in inflation-adjusted dollars in 2017-18 than they were in 2011-12.  Apart from the NDP governments in Manitoba and Alberta, it’s really only Quebec which has bothered to keep its post-secondary funding ahead of inflation.  Out east, it’s mostly been a disaster – New Brunswick universities are down 9% over the last six years (not the end of the world because of concomitant enrolment declines), and a whopping 21% in Newfoundland.

The story is different on the student aid front, because a few provinces have made some big moves this year.  Ontario and New Brunswick have introduced their “free tuition” guarantees, thus resulting in some significant increases in SFA funding, while Quebec is spending its alternative payment bonanza from the Canada Student Loans Program changes (long story short: under the 1964 opt-out agreement which permitted the creation of the Canada Student Loans Program, every time CSLP spends more, it has to send a larger cheque to Quebec).  On the other side, there’s Newfoundland, which has cut it’s student aid budget by a whopping 78%.  This appears to be because the province is now flouting federal student aid rules and making students max out their federal loans before accessing provincial aid, rather than splitting the load 60-40 as other provinces do.

Figure 3: 1-Year change in Provincial Student Financial Aid Expenditures, 2016-17 to 2017-18, in constant $2017

Figure 3 1 Year change in student aid expenditures


And here’s the multi-year picture, which shows a 46% increase in student aid over the past six years, from $1.9 billion to just under $2.8 billion.  But there are huge variations across provinces.  In Ontario, aid is up 83% over six years (and OSAP now constitutes over half of all provincial student aid spending), while Saskatchewan is down by half and Newfoundland by 86%, mostly in the present year.  The one province where there is an asterisk here is Alberta, where there was a change in reporting in 2013-2014; the actual growth is probably substantially closer to zero than to the 73% shown here.

Figure 4: 6-Year change in Provincial Student Financial Aid Expenditures, 2016-17 to 2017-18, in constant $2017

Figure 4 6 Year Change in Provincial Student Aid

So the overall narrative is still more or less the same it’s been for the past few years.  On the whole provincial governments seem a whole lot happier spending money on students than they do on institutions.    Over the long run that’s not healthy, and needs to change.

March 30

What’s Next for Student Aid

A few months ago, someone asked me what I wanted to see in the budget.  I said i) investment in aboriginal PSE, ii) system changes for the benefit of mature students and iii) changes to loan repayment (specifically, a reduction of the maximum loan payment from 20%  of disposable income to 15%).  To my great pleasure, the government came through on two of those wishes.  But there is still a lot of work to do yet.

Let’s start with the Post-Secondary Student Support Program, which the Government of Canada gives to individual First Nations to support band members’ education costs.  The Budget provides a $45 million (14%) bump to this program but also said the Government would “undertake a comprehensive and collaborative review with Indigenous partners of all current federal programs that support Indigenous students who wish to pursue post-secondary education”, which I think is code for “we’d prefer a new mechanism which is somewhat more transparent than PSSSP”.

Let’s just say I have my doubts about how easy this collaborative review will be.  Indigenous peoples – young ones especially – have a lot of issues with the federal government at the moment, and it will be difficult to try to manage a focussed review of this one subject without a lot of other agenda items intruding.  I’ve written on this subject before, and there certainly are ways which the funding could be arranged to be managed more efficiently.  That said, some of these ways involve taking management away from band councils and giving it to some other aboriginal organization operating at a larger scale and not all bands are going to find that appealing.

Anyways, the takeaway is: if the feds are expecting a replacement to PSSSP to be in place by fall 2019, they’d better get to work yesterday.

Now, what about the new measures for mature students/adults returning to school?  This was a welcome budget initiative, because the policy discussion has perhaps been focussed too heavily on traditional-aged students for the past few years.  There are however, maybe two cautions I would put on the initiative and how it will roll out.

The first is the budget description of the $287M over three years for programs benefitting these students as a “pilot project’.  I am fairly certain that is PMO-speak, not ESDC-speak.  First of all, I’m moderately certain the law doesn’t allow pilots; second, the idea that provinces are willingly going to spend time and money re-jigging all their program systems to accommodate program changes that are inherently temporary in nature is kind of fanciful.  So I suspect what’s going to happen here is that over the next few months CSLP is going to come up with a bunch of different ways to help this population (change cost allowances for older students, and maybe for dependents too), re-jig how prior-year income is calculated, raise loan limits for this population, raise grant eligibility, etc etc) and then roll them out in roughly ascending order of how irritating they are for provinces to program.  It’s not going to be a big bang, which may limit how well the policy is communicated to its intended targets.

But there’s a bigger issue at play here which the government missed in its haste to get a budget out the door.  One of the biggest problems in funding re-training are the artificial breaks in funding and jurisdiction that occur at the 12-month mark.  If your program is shorter than that, you’re covered by various provincial labour market initiatives and on the whole your compensation is decent.  Longer than that, you’re on fed/prov student aid, which in general is not as generous (and more to the point is repayable).  It would be useful for the two levels of government to work together to provide a more seamless set of benefits.  Perhaps regardless of program length, learners could benefit from 8 months of the more generous treatment and then move on to a slightly less-generous mixed loan/grant system.  This wouldn’t be a quick shift: my guess is that even if you now started talking about how to achieve this, it would still take four or five years for a solid, specific solution to come into view (if you think universities are slow, try federalism).  But still, now’s as good a time as any to start, and perhaps the dollars attached to the mature students programs may be a good conversation starter.

My third wish – the one that didn’t get any traction in this budget – was for improvement in student loan repayment.  I’m not that disappointed in the sense that I’m not greedy (no budget would ever have given me 3-for-3), but I do think there I work to be done here.  Perhaps this gets enacted as part of the follow-up to the Expert Panel on Youth being chaired by Vass Bednar and due for release at some point this spring (although who knows, if the Naylor Report is anything to go by, we could be waiting into 2019).  Or perhaps not: it’s not like CSLP hasn’t already been given a huge whack of work for the next couple of years.

But if that’s the worst problem we have in student aid in Canada, I’d say we are in pretty good shape.


(As a coda here, I’d just like to pay tribute to the Canada Student Loans Program’s Director-General, Mary Pichette, who is leaving the public service shortly.  Mary’s been involved two big rounds of CSLP reform: the one in 2004/5 which first created the grants for low-income students, and second the ones around the 2016 budget (not just the increase in grants but the many smaller but still important changes to need assessment as well. 

 I won’t say –I’m sure she wouldn’t want me to – that those two reforms were down to her.  But they were down to teams that she led.  She did a lot over her two stints in the program to make the policy shop more evidence-based and her legacy is simply that she’s made life easier for literally hundreds of thousands of student across the country.  They can’t thank her, but I can.  Mary, you will be missed.)

March 02

Bravo, New Brunswick

Readers may remember that about this time last year, I was giving the Government of New Brunswick a bit of stick for a botched student aid roll-out. Today I am pleased to give credit where it is due, and congratulate the folks in Fredericton for fixing the problem and developing a much better student aid system.

Let’s go back 12 months to pick up the story.  In February 2016, the Ontario government had come up with a fabulous new system which basically made a promise of grants equal to or greater than average tuition for students from low to mid-family incomes.  At family incomes above that, students received a declining amount of money out to about $110,000 at which point the grant flattens to a little under $2,000 (a remnant of the government’s ludicrous “30% tuition rebate” from 2011) and then falls to zero a little over $160,000.  With a bit of clumsiness this eventually, sort of, got branded as “free tuition for low- and middle-income students, which it isn’t, quite, but close enough for advertising.  Cue what is seen to be a major policy success.

It was such a success that New Brunswick decided to copy it later last spring.  Like Ontario, they built on the change to Canada Student Grants and eliminated some of their own tax credits (including the egregiously wasteful graduate tax rebate) to fund a “Tuition Access Bursary”, which guaranteed a grant equal to tuition (up to a maximum of $10,000, which was more generous than Ontario) for students from families making under $60,000. Which is great, right?  Well, yes, except the problem is, there was no phase-out for the grant.  At $59,999 in family income, there you were raking in $6500 or so in grants and at $60,001 you got $1200 in grants (the federal middle-income grant) and that’s not great social policy.  Making it worse was the fact that families in that $60K to $70K would also be losing a lot of money in tax credits that both the federal and provincial governments were ending in order to pay for this new benefit; my back-of-the envelope calculation was that in this range, parents were going to be about $1,200 worse off as a result of the change.

In any case, because I and others pointed out this flaw, the government after a brief period of defensive blustering decided it was best to go back to the drawing board and revisit the formula.  They did so and last week came up with a new “Tuition Relief for the Middle Class”, which basically involved taking a sliding declining scale of grants for families earning between $60-100,000 onto the existing Tuition Access Bursary (which has been renamed the “Free Tuition Program”).  Arguably, the New Brunswick program is now somewhat better than the Ontario program because 1) it’s not just “grants up to “average” tuition”, a caveat which I suspect is going to leave a lot of people slightly cheesed off when the program starts and 2) It still manages not to subsidize people up to that absurdly high $160K + threshold that Ontario insists on maintaining.  Ontario gets points for making its aid portable, though – New Brunswick’s program is only available to students who study in-province, which I think is a shame.

The announcement – which you know, hey guys, it’s a good news story! – was marred somewhat by some media sniping about how the number of beneficiaries is about 30% short of what was estimated last year.  To me this is neither here nor there: government cost estimates on year 1 of a new program are often a matter of throwing numbers at a dartboard.  The good news is that there is still money to either raise the entry threshold for the Free Tuition Program or (better still) expand the debt relief program or top up the amount of money available to high-need mature students and parents through the New Brunswick Bursary Program.

Now, all we need for this to be perfect is for New Brunswick to come up with a smart, credible monitoring program to examine the effects of these changes on participation over the next few years.

(New Brunswick folk: that’s on the way, right guys?  Right?  Well, you know where to find me if you need a hand…)

Anyways, as I say, credit where it is due.  Well done, New Brunswick.

February 02

Manitoba’s Golden Opportunity

It’s tough to be in government these days: prolonged slow growth means it’s difficult to keep increasing spending at a rate at which citizens have become accustomed.  Instead, with rising costs and little appetite to raise taxes or fees, governing often seems to be one long exercise in nickel-and-diming.  Higher education – in most of Canada at least – has felt some of this, but in truth has been insulated more than most other parts of the public service.

But the key role of government should not simply be to find ways to cut: it should be about increasing the effectiveness of public expenditures.  And in particular, making sure public expenditures are designed in such a way as to promote and not hinder growth.  That’s why, if there was one place in Canada I wish I could be an Advanced Education Minister right now, it’s Manitoba.  Because, as I explain in a new paper HESA is releasing today, Manitoba has a boatload of poorly-performing expenditures in higher education tax credits that could be re-purposed into areas which could really help the province.

Here’s the scoop: Manitoba has two tax credits – the Education Amount Tax Credit and the Tuition Fee Income Tax Rebate – which are neither particularly effective nor have many defenders within the higher education sector.  The former tax credit is a hold-over from the Diefenbaker era which all provinces (except Quebec) got stuck with in their portfolios when the provinces moved from a tax-on-tax to a tax-on-income system back in 2000.  In the past 12 months, the federal government, the province of Ontario and the Government of New Brunswick have all eliminated this tax credit because it was neither progressive nor efficient, and funneled that money back to student assistance.  The latter tax credit is effectively a tuition rebate for students who stay in the province, which is batty and wasteful for number of reasons I’ve previously outlined here. In any case, it is demonstrably too small to achieve its intended goal of convincing students who would otherwise not live in the province to live in the province.  The result is this money is a windfall gain to graduates, paying them to do something they were going to do anyways.  The elimination of these two tax measures could yield approximately $67 million per year in savings which could be spent more productively elsewhere within the higher education sector.

$67 million is a lot in Manitoba higher education.  Taking that money away from unproductive tax credits could fund a whole lot of new, useful investments.  These include:

  • Adding $14 million/year to provincial student assistance fund.  Spent correctly, this would be  enough to fund an Ontario-like “free tuition” guarantee to low- and middle-class Manitobans even if tuition fees were allowed to rise by a third (which, given how low tuition is in Manitoba, is probably a not a bad idea).
  • Investing $12 million/year to increasing supports to Indigenous students and expanding community delivery of programming in or near First Nations communities
  • Supporting the expansion of work-integrated learning at Manitoba universities and colleges with the creation of a dedicated $15 million/year fund.
  • Redressing a long-standing imbalance in post-secondary spending by increasing the number of seats in non-Metro Manitoba with a $15 million/year investment.
  • Creating an $11 million/year employer-driven “quick response training fund” to make it easier for employers with expanding businesses to access bespoke training.

In sum, for the price of two badly-designed tax credits, Manitoba could make real investments in access, both in terms of financial aid and providing spaces in under-served areas, increase support to Indigenous students and communities, improve the quality of education and provide more funds for employer-led training that could help relieve skills bottlenecks for investors.  How could you pass this up?  Who wouldn’t do this?

Over to you, Manitoba.

January 19

American Higher Education Under Trump

Tomorrow, Donald Trump will be sworn in as the 45th President of the United States (actually, the 44th person to be President: Grover Cleveland’s two non-consecutive terms screw up the count).  What does this mean for higher education?

First off, let’s recollect that where higher education is concerned, the US, like Canada, is a federation where the main decisions about funding public education are made at the state level. Decreased state investment in institutions and consequent rises in tuition have given the federal government a larger though indirect role in the system because the salience of student aid has risen.  And of course, the government spends an awful lot of money on scientific research, primarily but not exclusively through the National Institutes for Health (NIH) and the National Science Foundation (NSF).  And let’s also recollect that while the President names the Secretary of Education, a lot of control over specific budget items rests with Congress, which, despite being controlled by Republicans, will have ideas of their own.

Recall that Trump barely spoke about higher education during the campaign, other than endorsing an even-more-expensive version of income-based repayment than the existing one which was recently discovered to be costing nearly over $50 billion more than expected (short version: he wants to raise the repayment maximum from 10% of income to 12.5% but shorten the time before forgiveness to just 15 years).  Also, his education secretary Betsy DeVos, is a K-12 specialist (I’m using the term loosely) with very few known views on higher education.  I think it’s a given that their instincts will anti-regulatory and pro-market (which means things are looking up for private for-profits), but it’s hard to see them initiating a lot of new policy.  Which means the policy reins, such as they are, will likely be held by the Republican Congress and not the White House.

So what to expect?  Well, I think we can rule out any continuation of the Obama White House’s free college agenda, or anything vaguely like it.  That idea won’t disappear, but it’s something that’s going to happen in the states rather than in DC (witness Andrew Cuomo’s decision earlier this month to launch his own Ontario-like free tuition-plan).  Beyond that, you’re likely to see some cutting back on institutional reporting requirements, particularly with respect to Title IX, the federal law on sex-discrimination in education, and possibly a push towards more competency-based education.

Where it gets interesting, though, is on student-aid.  It’s not just that we’re likely to see cuts in things like loans to graduate students and (pace Trump’s own views) loan forgiveness.  We may see a return to more private capital in student loans (which would mostly be a bad things); we may also see institutions be required to pay for some of the costs of their own students’ loan defaults (an idea colloquially referred to as requiring institutions to have “skin in the game”.  Some think that the new Congress may push what are known as “Income Share Agreements”, which are kind of like graduate taxes only the entity giving the student money and then collecting a percentage of income afterwards is some kind of private investment firm rather than government.  One of the most crazy/plausible ideas I’ve heard is from University Ventures’ Ryan Craig who mused recently on twitter about setting rules whereby institutions might have to provide a certain fraction of total aid via ISAs in order to be eligible to receive federal aid.

On the research side: who knows?  Clearly, climate science is going to have a hard time.  But health sciences often do well under Republicans; the National Institutes of Health went from $18 billion/year to $30 billion/year under Bush Jr, for instance.  And Trump might decide to do something big and crazy like announcing a lunar base or a Mars mission (the former is a favourite of Newt Gingrich, the latter an obsession of Elon Musk, who suddenly seems quite close with the incoming White House), either of which would have substantial positive ramifications for university science budgets.  So we’ll see.

But put all this into some perspective: as far as Congressional priorities are concerned, changes to student aid are going to come several light years behind repealing Obamacare and dismantling various environmental protections.  The former in particular has some pretty serious budget impacts as repealing Obamacare is going to cost a ton of money.  That’s going to cause a scramble for offsetting budget cuts – one could imagine some pretty big across-the-board cuts in which higher education-related programs will simply be collateral damage.

It’s bound to be interesting, anyway.  Though I for one am glad I get to watch it all from a safe distance.

December 09

Does Student Debt Matter If You’re Not Going to Pay It Back?

You can accumulate one hell of a lot of debt these days in the UK.  Just in an undergraduate degree, fees are ‎£9,000 per year plus you can get another ‎£10,702 in maintenance loans per year of you’re studying in London.  Over a three-year degree that’s ‎£59,106 or a tad over $100,000 (yes, really). So, at face value one can understand the spate of stories coming out of the UK these days talking about how their massive debt loads are going to paralyze them for life, stop them being able to buy housing etc.

Except, wait – these are income contingent loans, not mortgage-style loans.  The maximum payment you have to make in any given year is 9% of marginal income over 21,000.  And the debt incurred doesn’t necessarily need to be paid back.  Loans are forgiven after 25 years, regardless of how much you have repaid.  Estimates vary, in part because it depends on what discount rates one chooses and in part because the government criminally keeps messing with the terms of the loans, but at the moment it is expected that between 25 and 40% of student loan balances will never be repaid and a higher proportion of students (perhaps 50%) will receive at least some forgiveness on their loans.  For those who do not repay their loans, the UK loan system is more like a tax than a loan – a 9% surtax on income over 21,000 which lasts for 25 years after graduation (more on that here).

Despite massive nominal debts, students simply aren’t facing massive repayment burden.   A graduate making 30,000 is only repaying 810 per year, or about 3.1% of after tax income, which is a heck of a lot less than the amount that the average Canadian graduate with student loan debt is paying (our grads pay close to 8% of after-tax income on average).  And they’re paying that regardless of how big their debt is, which is not true in Canada either: at any given level of income over $25,000 per year, Canadian student loans borrowers’ rise along with the amount of debt they have up to a maximum of 20% of family income.

(If you’re wondering how that works – how UK loans can be so big and yet borrowers repay so little – it’s precisely because the government expects quite large losses on the program.  Student loan burdens are easy to reduce if you’re prepared to go to extreme lengths to subsidize them).

The point of income-contingent loan systems like those in the UK, with their guarantees, their maximum payments and their generous forgiveness systems is precisely  to do everything possible to make life easier for borrowers, to ensure that their student loan debts are not going to affect their ability to borrow for other things later on.

But perception is everything.  If graduates feel that their large debts constrain their ability to do make certain life choices like buying a house even though (technically) they don’t, then can we say the policy is actually working? There’s an interesting side point here. When deciding on applications for mortgages or other types of consumer debt, it’s unclear whether banks in places like Australia and UK actually treat income-contingent student loan debt differently than Canadian and US banks treat mortgage-style debt.  They should, but apparently nobody knows for sure because no one’s ever checked – not that banks would necessarily fess up if they didn’t.

Now, I’m not saying that these stories coming out of the UK are in fact true; people in opposition to government policies will tend to come up with whatever argument sounds good at a particular moment. But even if such views aren’t widespread, the point raised is a good one.  Student loan policy wonks have always assumed that if you provide guarantees and limit liability/risk on student loans, then students will be ok with debt.  But if the facts of the policy don’t change people’s attitudes about risk, then the policies will fail, no matter how well they deal with the actual problems at hand.

But what’s the alternative?  It’s a bit of a scary thought.

November 30

Comparing International Student Loan Repayment Plans

People talk a lot about student debt and the burden it places on recent graduates.  Not surprisingly, different countries come to different policy conclusions about how this burden should be dealt with.  Today’s column examines how various countries choose to deal with this issue.

What I am going to do today is compare expected loan repayments under five different student loan regimes: Canada, the US, the UK, Australian and New Zealand.  This obviously does not fully examine the issue of loan “burdens” – to do that properly would require information on average debt and average post-graduate income which I suppose I could find but can’t be bothered to do just at the moment.  But it’s still a revealing exercise.

First, a brief description of the loan repayment schemes.  Canada and the Unites States have very similar systems, in that they are technically “mortgage-style” loan systems (you pay them down as you would a home mortgage, in equal installments), but which have “income-sensitive” features to help lower-income borrowers.  In Canada, that means the Repayment Assistance Program (RAP), which requires no repayment if income is below $25,000 and restricts payments to a maximum of 20% of income over that threshold.  In the US it is called PAYE (Pay-As-You-Earn) or REPAYE (don’t ask), which requires no payment if “adjusted gross income” (meaning income minus certain allowable deductions, roughly equivalent to line 260 on a Canadian tax form,) is less than 150% of the poverty line, which in practice means US$17,820 for a single individual.  Repayments are restricted to 10% of income above that level.  In both countries what that means is that repayment is directly tied to income until the point where payments rise above what they would be on a mortgage-style arrangement, at which point borrowers switch into the mortgage system.

The other three systems are income-contingent, meaning repayment is geared exclusively to income regardless of the size of the outstanding debt.  In the UK, the repayment threshold is £21,000 and borrowers repay 9% of their income above this level.  In New Zealand, the repayment threshold is NZ$19,084, and borrowers repay 12% of their income above this level.  Australia is more complicated: borrowers pay nothing until income passes $54,869, but then one pays an escalating percentage, starting at 4%, of one’s entire income (not just the bit above the threshold) as income rises above this.  For those interested, the contributions table is here.

For this comparison, I assume that the Canadian and American subjects each have outstanding loans of $25,000 (in local currency) which are eligible for income-based repayment. As noted above, size of debt is irrelevant for the other three examples.  I have converted everything into Canadian dollars at purchasing power parity using the July 2016 Big Mac Index (C$1=NZ$1=A$.958=$US.84=£.496).

With all that out of the way, Figure 1 shows how much student loan borrowers are expected to repay per month under each of the five systems.

 Figure 1: Required Monthly Repayment on Student Loans, by Income Level, in C$ at PPP, Selected Countries


The essential natures of each country’s program can be seen in this graph.  Canada and the US start out as upwardly-sloping lines but then plateau, which reflects their common nature as a blend of income-based and mortgage-style lending.  Payments in the US system rise more gently because of the different repayment maximum (10% vs 20%) and plateau at a lower level because of lower interest rates.  The New Zealand and UK patterns are simple upwardly-sloping curves.  Australia’s curve is notable not just because it is at zero for a long period but also because it jumps quickly at the point of the threshold.  In the social science literature, this is what they call a “step-function”, and it’s not a great idea because it means at the point of the threshold, individuals actually become significantly worse off (in this case, by $193 per month) by earning one extra dollar.

At low levels of borrower income, Canada, the United States and New Zealand all look quite similar in that they require borrowers to begin repayment at much lower levels of income ($20-25,000) than in either the UK ($43,000) or Australia ($57,000).  At income levels between $20,000 and $34,000, New Zealand demands the highest levels of repayment.  Between $34,000 and $50,000 (the part of the income curve where most recent graduates can be found), Canada has the highest repayment requirements; above $50,000 it’s New Zealand again. Between $25,000 and $75,000 it is definitely advantageous to be in Australia or the UK as these have the lowest payments.  However, by $80,000 repayments in the US system are the lowest and if we were to extend the chart out to $85,000 then we would see repayments in Australia and the UK exceed the Canadian level.

A final point: the Canadian system imposes the highest costs on students in precisely the income-range where most recent graduates fall.  We could do more for them here; specifically, if we reduced the maximum repayment rate to 15% from 20%, we would kink the curve in such a way that monthly repayments would never be higher than they are in New Zealand.  Something to think about for the next budget, perhaps.


October 19

The Yale Tuition Postponement Option

If you pay attention to student assistance, you know about income-contingent loans.  And if you’ve heard about income-contingent loans, you probably know that the first national scheme debuted in Australia back in the late 1980s.  You might even know that the first theoretical exploration of income-contingent loans was made by Milton Friedman back in the 1950s (actually, he was talking more about human-capital contracts, but close enough.  And you might occasionally wonder: why did it take 30 years to go from idea to implementation?  Well, the answer is that it didn’t: there was an intermediate stage in which a couple of universities tried to run their own income-contingent loan programs.

The year is 1971. Private 4-year universities were probably at their lowest-ever ebb relative to the big public flagships: massive amounts of public money had been pouring into public universities while privates had yet to really perfect their practice of extracting mega-millions from loaded donors.  But Inflation is starting to rise in America as a result of a decade worth of a guns AND butter fiscal policy.  And so schools like Yale began to think about raising tuition to meet higher costs and regain their place at the top of the academic dog-heap.

Enter economist James Tobin – a man who within a decade would win a Nobel Prize and is today mostly known for his advocacy of a beloved-of-the-left tax on financial transactions (the eponymous “Tobin Tax”).  Room and board at Yale College at the time was $3,900 (yes, I know, I know).  The university wanted to raise fees by about $1500 over the next five years, and so President Kingman Brewster (the model for Walden University’s President King in the comic strip Doonesbury) asked Tobin to come up with a scheme that would allow the institution raise said money without putting too much stress on students.

The result was something called the Yale Tuition Postponement Option.  Students could choose to defer part of their tuition (the part that came on top of the pre-1971 $3,900) until after graduation.  Repayment was a function of both loan balance and income: borrowers were required to repay 0.4% of their income for every $1,000 of tuition postponed (a minimum payment of $29/month was set).  Repayments could take as long as 35 years although it was expected to take less time than that.

There was a catch, though.  Loan programs lose money through defaults.  These either have to be made up through subsidy (which is what happens in most government student loan programs) or mutual insurance among borrowers.  Yale had no intention of subsidizing these loans, and so went the latter route.  These were therefore in effect group loans – you kept paying until your entire borrowing cohort had repaid.  You could escape this only by paying 150% of your initial loan and accrued interest.

You can imagine how this went.  A lot of students borrowed, but there was a fair bit of adverse selection (people who worried about their incomes opted-in, people who thought they would earn a lot opted-out).   And as time went on, a lot of graduates groused about subsidizing their less-successful classmates.  The program was phased out in 1977-78 because federal student aid was becoming more generous and because the university was starting to twig to both the problem of adverse-selection program and the problem of keeping in contact with graduates and getting them to voluntarily disclose their incomes.  Eventually, amidst rising alumni discontent, the program was wound up in 2001 and outstanding debts assumed by the University (which by this time could easily afford to do so).

The failure of the Yale Plan was certainly one reason why people were scared off income-contingency for another decade or so, until a reformist Australian government picked up the idea again in the late 1980s.  But from a policy perspective it was not a total loss.  One Yale student who enrolled in the program – fellow by the name of Clinton – thought it was a great idea.    He made it a center-piece of his 1992 election campaign, and an income-contingent tuition option was in place by 1994.  That specific policy never took off, but most of the income-based repayment plans (which are now used by 40% of all borrowers) owe their start to this program.

So, a failure for Yale perhaps.  But a long-term win for American students.

October 11

Hillary’s Higher Education Plans

Barring some sort of catastrophe, it now seems pretty clear that Hillary Clinton will be the 45th President of the United States.  There is a reasonable chance (51.6% in Monday’s FiveThirtyEight forecast) that the Democrats could regain the Senate and an outside chance that they could also regain the House.   Those odds probably change a bit in the Democrats’ favour once some post-grope polls come out later this week, but the basic outline of a post-November 7 world – Hillary in charge, with a split Congress – is now pretty clear.  What does it mean for higher education?

Well, you wouldn’t know it from any of the debates – we’ve now gone 270 minutes without a single second being spent on education – but higher education is a major plank in Hillary’s platform.  But her policies on higher education have evolved somewhat over the course of the campaign, mostly because her primary opponent Bernie Sanders’ success with millennials convinced her she needed a big, expensive, youth-oriented policy, and higher education (apparently) is it.

Hillary’s plan, release just prior to the July convention and known as “The New College Compact” consists of two pillars.  The first involves creating a system of “free tuition” at public universities for students from families with under $125,000 by 2021 (it would start at $85,000 in 2017 and rise by $10K each year thereafter) .  On the fact of it, this is a bit like what the Ontario Liberals and the Chilean socialists have developed, only more generous (i.e., using a higher cut-off point).  But the costing on this plan is – to put it mildly – hazy.  Her costing documents speak of spending $450 billion over ten years, but the tuition take from 4-year public alone is north of $55 billion, and that’s not including either the cost of 2-year colleges or the extra costs that would accrue if free tuition induced hundreds of thousands of students from private colleges to switch into the public system (the New America Foundation has correctly warned that not including funding for system growth could well result in a reduction of access for lower-income and minority students as middle-class students switching from privates could push out less-prepared lower-income kids from a fixed number of spaces).

The problem here is that the US (like Canada) is a federal system, with education a responsibility of the states.  The federal government can promising anything it likes about tuition, but at the end of the day it is states who have the final say.  The best the feds can do is work out a system of carrots and sticks to entice the states into a program.  The wording of the plan seems to imply that states who want to get reduce tuition will sign up for grants from Washington in return for meeting certain conditions – one of them being pouring more money of their own into their systems.  But the progress of Obamacare, which required considerably less from states but has only brough two-third of states on board so far, should give everyone pause.  On top of that, of course, the President alone can’t appropriate funds unilaterally.  Congress would need to be on-side as well, and the Democrats are still a long way from being able to make that happen.  Which is why most higher education analysts in the US seem to assume that the plan is more talk than action: a rhetorical statement which can attract voters rather than a plan likely to be implemented.

The second part of the Clinton plan involves a three-month moratorium on student loan repayment allowing all borrowers – including those in repayment – to re-finance their loans at a lower rate.  There is a fair amount of scepticism about how effective this measure might be.  As Robert Kelchen of Seton Hall University (possibly the shrewdest US student loans pundit out there), wrote in The Conversation a couple of months ago, the most-indebted graduates tend not to be the ones with the high default rates because default is most commonly associated with dropouts and hence lower levels of debt, and also because over 40% of borrowers in the US are now in income-based plans and so changing the level of interest will have minimal effects on repayments.  In other words, it will be a big income transfer to younger Americans, but not necessarily one that will do much to increase access or reduce defaults.

So after the election, what we can probably expect is a situation quite similar to what we had prior to 2014: a President and a Senate with a desire to make college more affordable (though not necessarily in particularly efficient ways), with a House implacably opposed and states offering indifferent support.  But a catastrophic Republican result in the House – which remains a possibility following this weekend’s stampede of defections – might result in some very rapid and drastic policy changes from the new administration.

Stay tuned for November 8th. 

Page 1 of 1312345...10...Last »