Sunday, 25 March 2018

Even Stevens? No, Says Amy Stevens

I have a degree in economics.

I'm frontloading this not to show off but to make it plain you really ought to consider what I am about to say. I'm not your usual armchair commentator.

There is a lot of bad economics out there that is promoted as not only good economics but orthodox economics. This actually goes as far as the very discipline itself: what economics is, is very misunderstood.

Economics is a very flexible discipline. It's not quite as adaptable as statistics but the economic frame of mind is highly portable. There are a lot of subjects to which economics goes, "Well, obviously we should think about them this way". It's actually quite rare that you have to coerce a topic into a format understandable via economic concepts.

Education, as it happens, isn't just readily understandable through economics, it's one of economics' favourite subjects. Yes, that's right. Education is just as interesting to Economics as trade. In fact, it might be more interesting. The point is, there's a debate to be had. If this surprises you, you don't know anything about economics.

There are several ways economics likes to understand tertiary education or aspects thereof. It's often seen as a form of signalling, a way of indicating that you (the student) are a good type and worthy of purchase (employment). It's probably more commonly understood as human capital investment; tertiary education improves quality. More obscure angles include bubbles, bundling and bellwetherism. In popular discourse what you often find is things like Amy Stevens' "thinkpiece" entitled, "The millennial problem with free tertiary education."

Let's do a largely paragraph by paragraph take on that... been a while since I've done so.
While typically a National voter, like many I decided 2017 was time for a change. Given I'm at the higher end of the tax bracket spectrum, some might ask what I had to gain from the employment of 'Taxinda'.
If you're a typically National voter you must be voting National's way more than 50% of the time, right? Which would imply that Stevens is old enough to have voted since 2011. Alternatively, this is more likely a misleading way of interpreting the text. That's probably more likely. I went the other way though because because being more precise in the use of language is important. Let's take a look at NZ's tax brackets:
Up to $14,000...................................... 10.5%
Over $14,000 and up to $48,000......... 17.5%
Over $48,000 and up to $70,000......... 30%
Remaining income over $70,000........ 33%
So, we assume that Stevens is pushing $70,000 but isn't quite there? And is paying... not very much on that somewhat less than $22,000? Is this an entry level salary? It doesn't seem like it? What stage of life is Amy Stevens at anyway? Remember this as we skip some paragraphs to get to...
However, one policy that concerns me is free tertiary education, and the extra $50 a week students are getting in their pockets for living costs. 
It hurts that those just out of university - with student loan debts of  $50,000 to $70,000 to our names - are left to foot the bill. 
Okay, I don't believe Stevens for a moment, here. Neither should you.

The typical degree in NZ is three years long. I did a conjoint so it took four. In fact, I did a conjoint purely funded by Studylink and with more courses than required. I managed $42,000-ish. That's quite some way off Stevens' lower range. It might even include what I am doing now... which is a fifth year of study. Let's see some aggregate data.
Students were leaving university with an average of between $16,600 and $17,220 of debt, with bachelor students tending to have the largest volume of borrowing, the report found.
That's like half of what I've got. But the same article also includes:
The average loan balance at June 30 was close to $21,000, the report said, while the average time it takes a graduate to repay their loan is now 8.4 years.
Presumably by average they're reporting means, which they shouldn't be. Mean values are highly sensitive to outliers and skewed data... both of which are to be expected with money related variables. If we go to the actual report, the median is more like $15,000. And if you look at page 32 you'll see that on 30 June 2016 only 6.3% of students had loans of more than $50,000. And that number barely changed for 30 June 2017 (page 38).

Amy Stevens... you are talking out of your arse.

Now, it obviously follows that people are getting hysterical the other way. Student Loan Doom is a feature of the USA a lot more than it is here.

I have $40,000 in debt, yes, but it's interest free. And I am in a small minority... 85% of people have loans of less than 40k and 50% less than 15k. We're not loading up people with debts so great as to cause access problems. And given that these are interest free... well, ask a financial adviser if there's a difference between $40,000 with and without interest.

The access problem that we actually have in NZ is regarding living costs. There are eight or something like that universities in NZ. All of them are in fairly major cities and two are in Auckland... across the road from each other. To go to university you need:

  • to move
  • to live within a reasonable distance of a university (in zone, if you will)
  • to live along a major transport artery and tolerate a long commute (me)
  • have a very, very long commute

Three of these options are burdensome. The first and fourth obviously induce pressures which can compromise the programme of study. That's not good from our human capital perspective. You might even argue that so does the third. To be honest, the train is a drag some weeks and some days.

Fees free education when divorced from compulsory education is the definition of middle class welfare. It makes living life according to middle class sensibilities free for the wealth (middle and upper) classes. For everyone else it doesn't materially impact the burdens that they actually suffer. Notice, for instance, which schools are close to university? High Decile and private schools.
My friends and I all felt relatively well off at university. In fact, we now appear less well off as entry-level workers.
With many of my friends working 'cashies' outside their 9-5 (we wish) jobs, I'm not sure there are enough hours in the day for the echo boomers to subsidise students for their study - or rather, their lifestyle.
A lot of people have a big problem with anecdotes. I don't. I know the actual issue with anecdotes is generalisability.

Most of the data that we talk about is just a collection of lots and lots of anecdotes. I have $40,000 of debt... that's an anecdote. If you add my anecdotal experience together with lots of similar ones in a systematic fashion? Now we have a dataset.

It's kind of okay to use anecdotes, then. So long as you're not talking about evidence gathering and let us know some basic details, the anecdotes can be used to shape a discussion.

Stevens isn't talking about evidence gathering. She's mentioning that her friends feel cash stretched. Well, okay, if we knew any details about you or them we'd be able to do something with this. But there's a big difference between someone who finished uni in 2010 and someone who finished in 2017. We need to know this sort of information. Especially when you've already made very, very misleading statement.

As it is, Stevens' friends have nothing to say.
What’s more, the extra students going to university for the 'free ride' will only devalue our education system and flood the New Zealand labour market. It will become more and more difficult to distinguish between a highly skilled worker and a free rider.
That's not what a free rider is.

In economics the free rider problem occurs when you have someone who is able to claim the benefits of a product without the producer's being able to exclude them. If you don't pay any taxes and your country gets invaded, you free ride off your defence forces. If you catch a train which has no fare control measures without paying, you're kind of free riding... even if that's what you're literally doing. Free riders aren't people who consume products that cost $0.

The jargon that Steven is actually after here is "good type" for "highly skilled worker" and "bad type" for "free rider". And the concept she's after is signalling. The introductory problem is known as the market for lemons (i.e. dodgy second hand cars).

The basic argument is that going to university and paying is something that only good types will do. In the extreme version, the fact of attending university does absolutely nothing other than affirm good type-ness. As long as we make a few assumptions about behaviour, the bad types will always prefer to do something other than go to university. As a consequence, anyone who has a university degree has to be a good type.

If you're paying attention you'll notice this isn't quite the same as saying "highly skilled worker". In fact, it usually means someone that will be a decent (not dodgy) employee.

If we start to believe that university education actually improves skills, then we believe that we can make good types into better types. And hence we believe that we can make bad types into better types too... medium or even good types, right?

Because university is only a signal, when you make it free you stop the good types from being able to indicate their superior quality. As a consequence, bad types are able to enter the market and vie for jobs they would previously have been totally shut out of. Worst case scenario, the good types don't even find it worth going to university any more and the whole house of cards (sorry, market) collapses completely. (Woah, you mean self interest can cause market failure? B-but National said... sorry, mate, National positioned itself as the economically aware party but that was just fake news; they're a bad type.*)

In the real world we know this isn't true.

Going to university doesn't indicate good typeness. Sticking with it for three years and passing courses, does. Making university free doesn't change this. And in the right contexts, we might find that good types and bad types are currently both motivated rationally and selfishly to go anyway. Which is to say, it's unclear if this will change anything.

Furthermore, you don't have to do just the three years. There are plenty of options which will leave you at university for a longer period of time. If university is strictly about signalling and costs, you're able to substitute time for money. This can be seen as a way of interpreting marks as well. Employers aren't stupid. They know that putting in the time is associated with good types only and better marks.

I am also not convinced that universities are nothing more than "signal mills". I believe that I know more and have more mental tools now than when I started this blog. I know for a fact that there are ways of looking at the world that I would never have encountered if it weren't for courses I've taken. And, sure, the content is available elsewhere (online etc.) but that doesn't mean that my awareness of these things didn't stem from uni.

In the ideal world, you start off not knowing anything and by the end of the process you haven't noticed that you know more. If you're doing a PhD, you probably didn't struggle much ever. And it's the people with PhDs who shape our understanding of human capital versus signalling theory more than anyone else...

tl;dr -- a university education has an absolute value, and that's never devalued by others having one too
The free education incentive should be focused solely on industries where workers are needed, for example construction and trade. Perhaps then we wouldn't have to resort to immigration measures and could continue growing New Zealand as a small and open economy.

Amy Stevens, you're an idiot. And if anyone doesn't want me to call her that, she shouldn't say dumb things. And by extension, if I don't want people to call me an idiot, I shouldn't say dumb things. I don't think I have. Somehow, I think Amy Stevens thought that too, though.**

* Yes, our analysis does suggest that if a bad type is able to win the most votes in an election, it's very unlikely there are any good types.

**Note... there's another concept related to signalling (i.e. another key idea in the analysis of adverse selection) which I don't recall the name of. I have suppressed such cautionary disclaimers both out of confidence and also as a means of making this link make sense.

1 comment:

  1. Somewhere in that report where I got most of the figures on there's actually a section which is limited to degrees or something... the point is the numbers are more relevant than "all student loans". However, they still do not line up with Stevens it's just that her error is less egregious.

    I'd tell you where to look with more details but I accidentally discovered this a while ago and have only just remembered I never did a correction.