Friday 16 July 2010

Yeah, but will it get you a job? [updated (2)]

The Minister for Tertiary Education here in Kiwiland is mulling over the idea of having university funding be partially contingent on whether graduates are able to get jobs after university.

If we now reckon the point of universities is the increased earning power of graduates, there's little case for state funding of university students beyond potentially having government backed loans (at market interest rates or a bit higher) because of the potential for market failure where human capital cannot serve as collateral. Of course, universities serve multiple goals. But the more emphasis is placed on this one, the weaker is the overall case for public funding.

Imagine that you're an admissions officer at one of the Kiwi universities. Dikat comes through that funding outcomes will depend on proportion of graduates who wind up getting jobs. You might find yourself with incentive to turn down the folks you reckon are less likely to wind up getting jobs. At least that's what I'd do.

On the plus side, it would be very good to start getting some data on where graduates wind up. A lot of students come in to Commerce because they reckon it provides better employment chances than Arts. I think they underestimate the value of the meatier parts of the Arts and overestimate the value of the fluffier parts of Commerce. But I'd love to see what the actual numbers look like.

Update: Matt over at TVHE argues that tertiary ed funding ought to be based on external benefits, that it's plausible that external benefits vary across disciplines, and that those benefits may or may not be correlated with work placement.

I don't necessarily disagree. But it's unlikely that it's the education that provides the benefit; rather, it would be whatever's done with the degree afterwards, right? Shouldn't we then just provide the subsidy for the provision of the service providing the positive external effect, and let folks sort out for themselves whether or not to take the training? So suppose that you want to argue that primary school teachers provide all kinds of value over and above what they're paid. Taking that claim as given, isn't it more efficient to then just provide a wage subsidy for primary teachers rather than to provide subsidized training? The former will also draw in folks from abroad who might be able to provide the service; the latter will just increase the domestic supply. Since I'm agnostic about whether it's more efficient to have locally or foreign trained providers, the wage subsidy is a nice neutral way of letting the market sort that out.

Moreover, for most professions providing such external benefits, there will be multiple paths towards success in that profession. A great social worker, for example, could have a background in formal social work training or could have any number of other degrees, or even none. Why privilege one route rather than others when it's unclear that any particular route is best for all comers?

It's hard to build the case for subsidising students. We can build a case for subsidising research as a basic public good, or for dissemination of knowledge on similar grounds. But it's hard to argue that there are great efficiencies from these kinds of big transfers to middle class students.

Update 2: Note that Education Directions sees this as a likely non-issue; rather, it was a distraction thrown out at the Minister at the end of the speech in Q&A to divert attention from other stuff that's actually on the agenda. That's entirely possible; Dave there is certainly watching these issues more closely than I am.
Joyce has said more since then, but you have to admire a Minister who went into a speech with the main public issue being current caps on enrolments and steered the debate completely the other way. The first 24 hours focused on universities and ITPs screwing students and the government through extra fees, while the second 24 hours has focused on providers attacking employment outcomes as a measure of their performance. Not only has the focus shifted from the government and onto providers making rather awkward defences, but it has shifted to a potential policy that is years away. I thought Joyce would be good, but this is a great political lesson. Next we’ll be talking about whether Martian ecology should be funded in 2050.


  1. You might find yourself with incentive to turn down the folks you reckon are less likely to wind up getting jobs

    ...maybe. It would be a brave move considering the difference that 3+ years at uni can make to a person, but I suppose if you had more applicants than places it might have to do.

  2. Hi Eric,

    I completely agree with the points you've raised regarding my post. My only point was supposed to be "if we subsidise education, it should be on the basis of external benefits". I was not so much trying to say that these benefits exist ;)

  3. @StephenR: If it's a hard funding rule and serious money is at stake, the Uni would just need to run a few regressions on incoming student characteristics, eventual job outcomes, and start turning down folks with bad incoming characteristics. Sure, 3 years uni can make a big difference. But some characteristics are still going to more strongly correlate with ex post success than will others, and a uni with a lot of money on the line would be ill advised to ignore that.

    @Matt: Even if those benefits exist, if they manifest from the employment outcome rather than just from the education, then the subsidy is properly applied to the activity that generates the external benefit.

  4. This is a great idea for perpetuating inequality although I assume that's not the intended outcome.

  5. I would fully expect that the results of such a rational University response to incentives would be some decrease in social mobility, where folks who'd be outliers then have a hard time getting into University. But again, I'm not the one advocating the change in rules to emphasize employability. I'm just stating what the best response to those rules would be.

  6. "If it's a hard funding rule and serious money is at stake, the Uni would just need to run a few regressions on incoming student characteristics, eventual job outcomes, and start turning down folks with bad incoming characteristics."

    Sounds like a fun job for Monday! Though I've been advocating a hardline, rational approach to enrolments for a while now and people seem to balk at the idea. I don't imagine the folks that write the Tertiary Education Strategy would be happy, but then they're the same people who want to introduce performance-based incentives.

  7. @hefe: I've never seen data on job outcomes, so we're missing the left hand side of the regression equation. Dave over at Ed Directions though notes that this whole issue may well be a red herring (see update above).