Category Archives: Higher education externalities

Do higher education subsidies produce public benefits?, 1985 version

I spent the last day of 2012 tidying up my chapter in a book I am co-editing on the Dawkins higher education reforms (the aim is to publish in mid-2013 to coincide with the 25th anniversary of the government white paper). The subject of my chapter is the Liberal response to the Dawkins reforms, which turned into a story about the Liberal role in the long, convoluted path from the state-dominated higher education system created in 1974 to a more privately-funded, market-based system.

Mid-1980s Cabinet papers released today provide a bit more of the background. At this point, Labor was still committed to free education for domestic students. But the Department of Finance thought that this was a bad idea, and wrote a memorandum explaining why.

It has the usual material about free higher education not changing the socio-economic profile of students. But it also contains a version of the key argument in my Graduate Winners report:

While these external benefits [better organised and functing political and social systems, potentially lower crime, sickness, disease, application of research undertaken in conjunction with education] are of course very difficult to measure they are widely believed to exist. To acknowledge their existence however is not to make a case on efficiency grounds for the full public subsidisation of higher education: full susbidies would only be warranted if there were no private benefits at all which is not likely to be the case; most people would expect extra income, status, and work satisfaction as a result of tertiary education…

It goes on to note that there is no public benefit from hobby or recreational study, and there is a risk of over-investment – ie, there could be greater economic and social well-being from investing the same money elsewhere.

I don’t think Finance’s document quite describes the underlying logic of its argument, which 1) are there public benefits from a course? (if no, just leave it to the market); 2) If yes, are the private benefits large enough to attract students? (if yes to this question, just leave it to the market); 3) If the private benefits are not large enough, will public subsidies lead to enough public benefit to justify intervention?

But they are certainly right that a public benefit argument can’t possibly justify full public funding of higher education, as susbequent events showed.

What do OECD comparisons tell us about Australian higher education funding?

In The Australian this morning, Simon Marginson suggests that there is something wrong with the methodology in Graduate Winners.

He starts by accusing Grattan of starting with conclusions and backing them with ‘selective studies’ and cherry-picking data from other sources. My colleagues who did the empirical work for Graduate Winners are very unimpressed with this impugning of their professional integrity. We started with the public benefits claimed in the base funding review, and looked for whatever primary Australian data we could find. There was no cherry picking, no selectivity – and nobody has come forward with anything Australian that we missed.

Particularly on the non-financial benefits, I can confidently say that nobody in Australia has ever analysed this issue as carefully and comprehensively as we did. Jim Savage’s work on this did not get the attention it deserved due to the political controversy over tuition subsidies, but his technical paper has to be the starting point for any future work in this area.

Marginson’s alternative methodology is to look at OECD comparisons, stating that ‘it is significant that the Grattan report carefully avoids both the method and content of the OECD. It would have us believe Australian higher education has nothing to learn from global comparisons.’ Given how often this point has come up, in hindsight perhaps I should have included a section on this subject. But I don’t think the OECD funding data in itself tells us much other than that countries have very different mixes of state and private funding, and that these are reflected in their higher education financing systems. A high fee university system would not mesh well with Scandinavian tax rates. But it does fit with the lower tax rates in Australia, the US or Japan.

Similarly, I find Marginson’s claim that other countries report stronger relationships between education and social engagement uncompelling (we report this fact, he did not need to go to the OECD). American colleges and universities expressly inculcate civic values, Australian universities very rarely do so. The differences between the countries reflect the different histories of their higher education systems, and not public funding levels.

If I had 15 minutes to prepare a debating case on higher education, I probably would turn to the OECD for some handy facts and figures. The key OECD document is called Education at a Glance for a reason. But if as was the case I had months and the help of colleagues to explore the Australian data and think through the conceptual issues, that is surely preferable. In my mind the two big issues in higher education public funding are whether it causes significant additional public benefits (on top of those that would be derived from a market system), and whether there are access implications from fees. We focused on these big issues.

In any case, if we had used OECD data it would have tended to support our conclusion that the level of public funding is not the key variable in higher education systems. As I showed in an earlier post, there is no evidence that lower fees result in higher attainment. Indeed, the data suggests the reverse. One of my colleagues updated our analysis today with the latest Education at a Glance data, and (unsurprisingly) it shows again that high fees and high attainment tend to go together.

Even if OECD comparisons were a better methodology, they don’t always get Marginson where he wants to go.

Should the government redistribute student fees between universities?

In an AFR op-ed today (not behind a paywall – things are improving), Macquarie Uni VC Steve Schwartz suggests some egalitarianism for universities.

If fees are deregulated, the more prestigious universities would charge higher fees than others. Schwartz suggests that if they did, their government subsidy should be reduced, and redistributed to other universities.

The reason is regulatory – the new Tertiary Education Quality and Standards Agency is imposing standards on all universities, but it is hard for the poorer universities to match the standards of the wealthier universities.

I doubt TEQSA will require all universities to be the same. A university licence to operate depends on meeting minimum standards, not being identical to all other universities. That said, there is a tendency in the standards released to date to codify common practices, some of which are of doubtful necessity. If this continues, the universities in the best financial position to try new things will tend to set the standards over the long term. Read more »

Why does the base funding review panel think lawyers should pay less for their education, and teachers and nurses more?

The most contentious aspect of the base funding review report, released today, is likely to be its proposal to change the basis of public subsidy for higher education.

At the moment, the public subsidy is not explicitly based on public benefits. Effectively, it’s just what’s left after student contributions are deducted from total per student funding by discipline. Total funding is loosely derived from a study of higher education expenditure 20 years ago, while student contributions are loosely based on differential HECS introduced in 1997. Differential HECS was in turn based roughly on average private earnings of graduates in particular disciplines. So law and medical students paid the most because lawyers and doctors earn a lot. Education and nursing students pay lower amounts, because teachers and nurses have modest salaries.

According to the base funding review, public subsidy should be based on the government paying for public benefits. They say the public benefits are equivalent to between 40% and 60% of total annual expenditure per student. These public benefits are defined as miscellaneous non-pecuniary benefits to society, plus the ‘direct fiscal dividend’ from the additional taxes graduates pay due to their increased earnings.

Leaving aside whether these numbers are robust (I doubt it, but assume they are for the sake of argument), what is the justification for using public benefit as the basis for public subsidy? The base funding review offers two possibilities.

One possibility is that without subsidy ‘private benefits might not be enough to motivate a student to pay full fees’. So the logic would be that through subsidies the private benefits are increased to a point where it is financially attractive for students to enrol in higher education, and then go on to the produce the claimed public benefits. Read more »

Does higher education reduce crime?

One argument made for higher education – at least when arguing for more funding – is that it helps reduce crime. A visiting OECD official recently made the reduced crime claim for higher education, citing Walter McMahon.

Graduates are likely to have quite low conviction rates – I have not been able to find precise statistics, but in 2009 only 14% of 25-34 year old prisoners had completed year 12, compared to 63% of the general population in 2006.

But it seems more plausible to me that graduates are people who were always at relatively low risk of offending, regardless of whether or not they pursued higher education. This low risk would be a function of better socialisation, and the ability to earn a reasonable income without breaking the law. Education is likely to have whatever preventive effects it is going to have well before higher education.

The broad historical trends would also seem to count against any straightforward link between higher education and crime. Australia’s crime statistics don’t lend themselves to easy long-term time series, but crime and education both escalated significantly from the 1970s. The figure below shows crime and higher education attainment increasing from the mid-1990s to around the turn of the century, before crime started trending down again (as also occurred in other countries).


Sources: Education and Work, Australian Institute of Criminology

My theory would be that a third factor at least partly explains both trends, though I think crime is a more multi-factor phenomenon than higher education. The collapsing labour market opportunities for men with little education over the last 30-40 years made both crime and higher education more profitable relative to the alternative of welfare/insecure jobs. So crime and higher education both increased.

But there is no direct relationship between crime and higher education, and increasing the latter will not decrease the former.