There are two constructs that make it extremely difficult to talk sensibly about who should pay for higher education.  The first is “affordability” and the second is “independence”, in the sense of students’ independence from the rest of their family.  It’s worth exploring these concepts in detail to see how they complicate analysis

Let’s start with affordability, a term which unfortunately has become equated with “accessibility”, which it clearly is not.  Simply put, “affordable” is something you can pay for.  It is, in a strict sense, a ratio: the price of a good as a fraction of some measure of purchasing power, however you define that.  Talking about whether the sticker price of tuition is affordable or not is therefore massively misleading in two ways.

The first, obviously, is that the sticker price of tuition is not the net price after grants and other subsidies (y’all know my views on that, I won’t belabour).  The second is that as long as we have variation in incomes, any given good or service can never be equally affordable for all.  Something which is affordable for one is not affordable for someone else and vice-versa.  Therefore, saying that tuition at a given level is “unaffordable” is – at best – a clumsy way of saying “it is unaffordable for more people than it should be”.  In fact, the entire point of student aid is to allow governments and institutions to use price discrimination in favour of poorer students: to make sure it is affordable to them while not handing out subsidies to those who can afford to contribute more.  The idea that tuition must be low or zero in order for education to be “affordable to all” in effect requires one to pretend that effective income-based student aid does not or cannot exist.  That simply isn’t true, and politicians and policy wonks should know better than to claim it is.

But there is a trickier problem nested in here and that is: whose income should be counted when trying to work out that affordability ratio?  A net tuition fee of $4,000 would be affordable for a family with an income of $80,000: it’s only 5% of income.  But that same fee can look immense to a student who might only make $10,000 a year.  This is why people who want to emphasize the affordability of tuition usually talk in terms of family income, and those who wish to talk about how crazy expensive it is tend to talk about it in terms of student income (or better yet, as a multiple of the minimum wage).

Neither lens on affordability is incorrect.  Rather, it’s a matter if understanding when to use each lens.  For some students – those just starting out in post-secondary education, say – you probably want to count parental income, since most students at age 18 are receiving help from their parents or broader family.  Almost no one would think that’s an appropriate measure for, say, a student entering a doctoral program, or a mature student in their 30s returning to college.

Student aid programs around the world take one of two approaches to this.  The Scandinavian ones effectively assume students are independent of their parents as soon as they turn 18 and all students get maximum benefits.  In Australia, students are “independent” when they turn 25, in the US the arbitrary line is at age 24.  In Canada, where the Government’s (imo) bizarre reading of the Supreme Court’s Gosselin decision means the Canada Student Loans Program cannot make hard distinctions between citizens based on age. Rather, we base “independence” on a bunch of weird characteristics (such as marital status, being more than 4 years out secondary or two consecutive years in the labour market, or – in Quebec – being 20 weeks pregnant), but a basic rule of thumb is 22 years old.

The problem with this type of approach is that wherever one sets the line, it’s all-or-nothing.  On one side of the line, we assume parents are giving students their full extent of support, and on the other we assume they are getting nothing at all.  This is probably unrealistic.  The best evidence we have on parental support – at least among those able to provide it – is that they give their kids a lot money in the first year or two and then gradually start to pull back as their children start earning more and/or savings start to dwindle.

This suggests that there may be another method for distributing student support: reducing parental contributions with age.  That is, assess a full parental contribution for the first two years of post-secondary (assuming a student starts at age 18) then maybe drop it to half for the next two and down to 25% for the two after that.  It would increase aid to upper-year students without abandoning the principle of parental contributions.  It would also reduce the phenomenon of the sudden “fifth-year” student borrower – the kid from a wealthy family who suddenly borrows the max in their fifth year because they are suddenly no longer considered dependent (and who, by some measures, account for 20% or more of all undergraduate borrowers).

In short: student support is complicated because emerging adulthood is complicated and there is no settled way of dealing with people who are of the age of majority but clearly not financially independent. Wishing away these complications will result in policy that is expensive, ineffective, or both.