Every once in awhile, someone important says that what Canada/America really needs are income-contingent loans.  I usually reply, “we have income-contingent loans in Canada/America, that’s what the Repayment Assistance Program/Income Based Repayment program does”. To which the rejoinder is “no, no, that’s not income-contingent, what I mean by income-contingent is recovery of the loan is done automatically through the tax system, so you don’t run into all these messy issues around borrowers in repayment having not signed up for things”.

At this juncture, I could point out that the size of the loan payment and its method of recovery aren’t the same thing (I wrote a monograph about this about a decade ago), but I usually just keep my mouth shut because, really, my interlocutors have a point.  RAP in Canada and IBR in the US would both be much better programs if borrowers in trouble automatically received relief, instead of going through the tedious application/income verification process they do now, and the easiest way to achieve this would be to run repayment through the tax system, as they do in Australia, the UK, and New Zealand.

So why don’t we?

The New America Foundation investigated this question in a recent paper, and enumerated a number of challenges in moving to a tax collection system.  One of these reasons is specific to the US (they tax families not individuals, so setting the tax rate on an individual is awkward if he/she is marries), and need not detain us here.  The other reasons can basically be boiled down into two big categories.

First, how do you integrate employers – who do the tax-withholding in Canada – into such an operation?  How do they know how much to withhold?  How do they know when to stop withholding (i.e., when the borrower is finished repaying)? And are we actually going to require students to tell their employers about their outstanding loans?  Part of the issue here relates to people who do not have a single, full-time job that provides all of their income.  How does withholding work when students have two jobs?  Or when wages are not the sole source of income?   Of course there are fixes and workarounds to these questions, but every fix and workaround creates even more complication.  And complication is what ICR is meant to avoid.

(In Canada of course, we’ve got quite specific reasons why income-contingent loans are difficult: namely, most students are not receiving one loan, but rather two – one from the province, and one from the feds – and these don’t always have identical conditions.  You’d need to to align both levels of government across the country for this to work.  That’s not impossible, of course, but it’s tricky.)

But there’s one final reason why governments are reluctant to recoup debts through the tax system, and that’s for fear of damaging something called “tax morale”.  Basically, tax morale is a way of measuring one’s sense of moral obligation to pay taxes, or one’s belief that taxes contribute meaningfully to society.  A 2004 paper in the Journal of Economic Psychology examined the effect on tax morale of Australia’s Higher Education Contribution Scheme, which collects student debt (technically “contributions” rather than debts, but the distinction can be a bit fine). The result, perhaps unsurprisingly, was that students with HECS debt were likelier to have lower tax morale than those who did not.  That might sound trivial, but to governments, it is not.  Our system of taxation depends on voluntary disclosure and reporting.  Messing with that has big consequences; putter around with it at your peril.

None of this should be taken as a reason to not collect student loans through the tax system.  There are a lot of potential benefits to such a policy.  My caution here is simply that implementation will be complicated, may lead to different kinds of errors and difficulties (especially for individuals with multiple jobs), and have drawbacks in terms of tax morale.  For good reason, governments don’t undertake system changes with this level of complexity lightly; there would be a serious risk to service delivery if something went wrong.

Maybe, just maybe, this is the next big project in student aid, now that we seem to be getting the switching-tax-credits-to-grants thing right.  Just don’t assume that this would be a simple process.