Every once in awhile, someone comes up a “new” concept in student financing and people get very excited about it. As in most other policy fields, the “newness” is a matter of perspective and debate: there’s only so many ways you can lend students money and many of the “new” ideas are just old ideas that got discarded for various reasons and resurrect either because circumstances have changed or because proponents aren’t aware of the history (or both).

The latest in a line of such ideas is “Income Share Agreements” or ISAs, which have been growing in popularity in the United States for a few years, particularly with Silicon Valley types who like to think they have re-invented or (God Forbid) disrupted literally everything. For instance, here is a rather breathless Andrew Sorkin piece on ISAs from the NYT a couple of months ago; Linda Nazareth of the McDonald-Laurier institute had a similar booster article in the Globe and Mail in February.

What are ISAs, you ask? Well, to answer this question, it’s worth taking a brief tour of the history of student lending.

When you give somebody money for a purpose – going to school, say – we tend to think of that money as coming in one of two forms: as a gift (i.e. a grant/bursary/scholarship/whatever) or as a loan. Now in regular life, loans have to be collateralized: you have to have some assets which can be used as security on a loan. But student loans work on a different principle: most students have no assets, and the education students are buying with their loans can’t be easily collateralized. So what governments do is simply assume a bunch of risk on students’ behalf and lend them the money, usually at concessionary rates with lots of assistance during repayment (yes, I know, loans with interest rates above prime don’t sound concessionary, but given that they are interest-free while in school and default rates are relatively high compared to commercial loans, they are).

Now, over the years various governments have, with the best of intentions, complicated student lending substantially, to the point where lots of well-meaning casual observers of the scene simply have no idea what governments are doing any more. For one thing, many started offering various forms of income-based assistance; suspending repayment below a certain threshold, scaling payments to income, etc. In some cases – like in New Zealand and the UK, we even have given up the normal infrastructure of loan repayment and turned administration of the whole thing over to the tax system. Combined with scaling payments to income, that starts to make student loans look a lot more like a form of income tax or “graduate tax”, even though they are clearly not because each borrower has a balance and the tax stops when the balance drops to zero.

Of course, the thing is that lending isn’t the only way to temporarily transfer money to someone. Think of this from the perspective of a company, which can either borrow money from a bank or other financial institute and repay that amount plus interest, or it can sell a portion of ownership in the company and allow the buyer to reap rewards through a future stream of dividends and/or capital gains. Though many people credit Milton Friedman’s paper The Role of Government in Education  as the work that originated the idea of student loans, he was actually advocating for the state to assume equity shares in graduates’ future income streams. For a long time, this idea went under the name of a “human capital contract (HCCs)”, though if it were collected along with taxes it could equally have been called a “graduate tax”.

The clear difference between HCCs/graduate taxes and student loans – in theory at least – is that the commitment in the latter is limited while the former is not. In a loan, you stop paying the loan when the principal is repaid; in a graduate tax, you simply pay forever, and some people end up paying a little while others end up paying a lot. A variant of the latter was tried at Yale in the early 1970s.  It didn’t go well.

One of the barriers to public acceptance of HCCs is that many people instinctively recoil from the analogy of their future labour as being something which can be divvied up into shares. Concepts like “indentured labour” seem leap to mind. From an analytic point of view, agreeing to pay someone a share of your income over a period of time is almost indistinguishable from paying them a fixed amount over a period of time, particularly if you just express the latter as a fraction of income. And if your government has already transformed your loan into something repayable as a fraction of income (i.e. an income contingent loan) and has started collecting the loan through the income tax system, then both become difficult to distinguish from graduate taxes.

(The UK has complicated this even further by adopting a loan system which saddles all its students with high balances and relatively low repayment obligations, meaning the loans will, in a majority of cases, never be paid off before they get forgiven 35 years after graduation. Thus, graduates with high salaries will experience it as a loan which goes to a zero balance while graduates with lower salaries will just experience it as a tax on income for 35 years, and by design the student can never know which it will be until a couple of decades into repayment. Bizarre.)

Now that all that is as clear as mud, we can get back to ISAs. Despite the insistence by various Silicon Valley types that they’ve found something new, ISAs are pretty much just Friedman’s HCCs, privately administered, and modified for reasons of public acceptability to look more like loans and less like a graduate tax. Nothing more, nothing less.

The real question is: given all the various forms of public loans which are available, and the increasing degree to which they have been made income-contingent, what is the actual market for ISAs? When does it make sense to use them and when does it not? More on this tomorrow.