More pennies for your thoughts? University fees and the Grattan report

“Graduates…have attractive jobs, above-average pay and status. They take interesting courses and enjoy student life. I think they are getting a bargain compared with their lifetime earnings potential.”

So says Andrew Norton, higher education programme director at the Grattan Institute, in his recent report ‘Graduate Winners: Assessing the public and private benefits of higher education’. Norton’s principle thesis is that graduates, rather than society as a whole, are the beneficiaries of higher education – and should therefore pay more, potentially much more, of the true costs involved in providing that education.

Discussion around the funding of higher education is important. Indeed, I would agree with Norton that a university education provides graduates (and even those students who do not graduate) with significant opportunities and benefits, so it is fair to question the degree to which an individual should contribute to the costs of that educational experience. But to follow the accounting approach laid out in the Grattan report is to cede an argument that should be philosophical and sociological to the realm of economics. It’s like the old logical fallacy “So have you stopped beating your wife yet?” By adopting the neat rhetorical device of comparative earnings, Norton and others like him are attempting to make the debate about quantifying the benefits graduates accrue, rather than what a University education is really for.

For all it’s careful analysis and sound-bite friendly calculations though (quotable statistics are a much easier sell than philosophical nuances, after all), when you look behind the numbers, Norton’s study is founded on what may be a less than secure foundation – the assumption that the higher lifetime earnings (and other benefits in relation to the general population) of graduates arise explicitly because of the degree they have obtained. But hang on a second, let’s unpack that assumption – because what Norton seems to be doing here is committing the schoolboy error of a equating correlation to causation.

How so? Well, are the higher earnings of graduates truly related to their University education, or are both simply related to another controlling variable? Since the 1950s, by way of analogy, both atmospheric CO2 and obesity levels in Australian society have increased sharply. To imply a causative relationship would be to suggest that in some way atmospheric CO2 causes obesity – as if we’re all just making our own carbohydrates directly through photosynthesis, perhaps – whereas in reality both are symptomatic of the common controlling variable of increased wealth.

In essence, the problem for Norton here – as he himself recognises in the inner workings of the report – is that the comparison between graduates and non-graduates is not unbiased. Yes, there are wonderful success stories of prodigiously driven, inspirational individuals who forgo the graduate path to achievement – stand up and take a bow Richard Branson and Steve Jobs – but in general, merit-based entry criteria mean that it is a brighter, more ambitious, and more self motivated cadre that undertake University education. If we also assume that career and business success correlates strongly to the same aspects of intelligence and drive, then it should not be a surprising observation that graduates are in the top bracket of career earners. How, then, do we calculate how much of the extra earning of graduates should be attributed to their University education, and how much to their innate ability?

Now here’s where I’m eternally grateful to Norton because, like a magician revealing how the illusion is performed, he tells us how it’s done. Admittedly in an appendix where an over-worked education reporter with a filing deadline to meet isn’t exactly likely to stumble across it. He assumes a number. Empirical basis? Scientific calculation? Theoretical underpinning? Nope. An arbitrary 10%. We’ll just assume University graduates enjoy a 10% earnings advantage due to their innate ability, and the rest of their career success is the result of their University experience.

Now, yes, this assumption is critical to make the calculations tractable so Norton can come up with his headline-grabbing $600,000 lifetime earnings differential…but the flip side is that the modeling is then simulating an artificial system rather than the behaviour it purports to capture. It’s like the old joke about a physicist charged with figuring out how to increase the milk production of a herd of cows, whose solution begins “First, assume a spherical cow…”

Norton does allow in the appendix for alternative calculations with an ability premium of up to 40%…but why stop there? Why not 50%? Or 100%? Or maybe they start with a premium of 150% but all that ‘student life’ they enjoy actually stunts their achievement. Okay I may be guilty of reductio ad absurdum here – but the point is that the ability premium has no innate basis – it exists only for that convenient mathematical purpose. There is no way we can resolve whether 10, or 20, or 200% is the appropriate figure to use. Like a Chinese high speed train project then – the Grattan analysis is built to look impressive and modern, but laid on potentially insecure foundations.

Why is that important? Surely the comprehensive modeling and calculations Norton undertakes are sufficiently robust that the conclusions of the report can’t be undermined by this one trivial simplification, can they? I’m not so confident on that score.

Let me illustrate by way of a detour through my own professional back yard. Physicist William Thomson (later Lord Kelvin) – a towering scientific intellect acknowledged as one of the greatest exponents of thermodynamics of the 19th Century – ‘proved’ in 1862 that the Earth was somewhere between 20 and 400 million years old. Earth scientists of the day thought that figure far too young to account for the global history evinced by the geological record, but none could fault his calculations. Indeed, Thomson’s calculations were accurate – brilliant even – and still stand as a valid construction. But they produce the wrong result. Out by a factor of at least 10, with the Earth’s actual age – should you be interested – now measured to around 4.54±0.05 billion years. How can this be? Because Thomson based his workings on the simplifying assumption that the Earth lacks an internal heat source, when in actuality radioactive decay has been continuously pumping out heat throughout geological history.

Although numbers may look good when quoted in a nice banner headline, any calculation, no matter how exquisitely conceived and accurately undertaken, is only useful if the underlying assumptions are true.

The relationship between higher education and society is a social compact that should be open for ongoing discussion, and as a part of that, we should be debating how Universities should be funded. In that context, the Grattan report should be seen as an important contribution, offering a coherent and well argued monetarist view of how the University system should operate. But it is nothing more than that. And if you think the economic modeling Norton presents should be privileged over other philosophical and sociological viewpoints in the discussion of the University system…I have a friend in Nigeria who would like to talk to you about a fortuitous windfall he’d like your help with.

Advertisements

2 thoughts on “More pennies for your thoughts? University fees and the Grattan report

  1. Dr Geoff Batt Post author

    As an addendum, following in Norton’s footsteps, Australian Federal Tertiary Education Minister Chris Evans earlier this month ascribed a 45% earnings advantage to the holder of a University degree in Australia. Although Evans was quoting from the recent OECD report ‘Education at a Glance 2012’, rather than the Grattan Institute document, the statistics quoted are remarkably similar. Unlike Norton’s comprehensive document however, the OECD report lacks information on the statistical approaches and assumptions used – and is therefore much more intellectually damaging. There are few things harder to refute than an unfounded statistical estimate, and once these figures enter the lexicon of debate (particularly in the hands of a partisan commentator), they often become virtually impossible to kill off.

    Reply
  2. Pingback: Of Cats and Cashflow: Human sentimentality and the flaw in economic rationalism | rocky subjects

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s