Misreadings and criticisms of Graduate Winners

The AFR published a response to Graduate Winners from Caroline McMillen, VC of the University of Newcastle. It provides an opportunity to respond to misreadings and criticisms.

Article starts, my responses in block quotes:

Access to a high-quality university education is the key to a stronger Australian workforce, economy and society. In turn, these are all important contributors to establishing a stronger place for Australia in the world.

An accessible university education is essential to ensure that Australia in what has been called the Asian century becomes a beacon for innovation and competitiveness.

The proposals contained in the Grattan Institute’s Graduate Winners report would jeopardise that future.

The report, which was made public last Monday, presents in measured language a reductive future for higher education in Australia, where students are motivated only by their graduate earning potential and the state withdraws its funding from what is currently recognised as a world-class university system.

Incorrect: The report shows (pages 56 to 59) that interest in the field of study is the top reason for choosing a course, and that a financially-based motivation model cannot explain why so many students with good ATARs choose humanities and performing arts, which have relatively poor employment and income outcomes.

The proposal is to shift the entire benefits and the risks of undertaking a university degree onto each individual student.

Incorrect: The report recommends a 50% cut in tuition subsidies for most courses; the taxpayer further takes risk through the HELP repayment threshold of $49,000 a year.

Implicit in its calculation of public versus private benefit is that all students undertake higher education at the start of their careers; that all students who study subjects that have historically produced high-earning graduates will also earn high salaries; and that all students have a tolerance for a large lifetime debt.

88% of bachelor-degree commencing students are less than 30 years old, and 95% are less than 40 years old. That we did not model mature-age students was a reasonable criticism and we have since done an analysis of students starting at age 30. For females, the median rate of return falls from 16% to 9%, and for males it falls from 14% to 10%. ‘Breakeven points’ (where the graduate comes out financially ahead of the median year 12 only person, after deducting education costs and opportunity cost of being out of the workforce) move up a little. But they are still not very sensitive to the level of fees, since these are still a small percentage of career earnings.

We show what we think are worst-case risks for not coming out ahead (information that should be more freely available, and which needs much more analysis to help students make choices that are in their own best interests). Further analysis that is not in the report shows that these largely flow from not working full-time; 85% of graduates who do work full-time come out ahead of the median year 12 only person of the same sex working full-time.

Any sensible person will seek to avoid unnecessary debt. The question is whether taking out debt is prudent or not.

This model would present difficult choices for those with, for example, law degrees keen to explore alternative careers with their legal education but under pressure to earn quickly and substantially.

This dynamic would also have major consequences for public organisations and small businesses seeking to employ affordable lawyers.

A bad example. Law gets very little subsidy as it is, and the impact of the Graduate Winners recommendation would be <$3,000 for the whole course. Even for someone working in legal aid it would make very little difference to their overall financial situation.

Of key concern is that the report largely dismisses the impact its reforms would have on mature-age students, indicating that any evidence of a consequence for such students is “contradictory”.

See above. It is more difficult to predict the behaviour of mature age students. Their opportunity costs are usually higher if they already have jobs, and for most courses this will be a bigger issue than student charges. “Mature age” can mean as young as 21, so some of the contradictory results in the literature may be due to this poor definition.

The British university sector has recently been subjected to a variation of the reforms suggested by Andrew Norton, author of the Grattan Institute report. In the UK, the balance of the financial burden of study has been placed onto students following drastic reductions in government subsidies.

It is of note in this context that the Universities and Colleges Admissions Service in the UK reported that applications from mature-age students have declined 15 to 20 per cent under the new fees regime.

Though the report also notes that this half this fall could be due to higher acceptances in 2011, ie the higher fees were announced in 2010 and so those with the chance to start university earlier at lower cost did so.

Another critical point about the UK system – like Australia before 2012, they have a capped system. So small decreases in applications have no effect on enrolments. They just reduce the number of people who miss out. Government funding typically leads to caps as a way of controlling expenditure, creating a paradox: public funding may increase demand, but it also decreases supply.

The Norton model is more extreme than the new UK system, and this raises questions about the viability of such a model for developing a skilled, educated and innovative workforce in regions such as our own.

Mostly incorrect: Graduate Winners is more radical in not proposing a grandfathering scheme (so that new charges only apply to new students), but given sunk costs and still lower overall degree costs drop-outs should be modest. However it less radical in other respects. For clinical subjects the UK subsidy cut is 25% compared to the 50% in Graduate Winners. However for lab subjects their cut is 70%, and for all other subjects it is 100%. Also, these cuts apply from year one. Graduate Winners proposes phasing down over four years.

As highlighted by another recent report by the Australian Workforce and Productivity Agency, higher skilled jobs are expected to grow at 1.6 times the rate of low-skilled jobs and the strongest growth in employment to 2025 will be in professional occupations.

This broad pattern of employment growth is almost certainly right, though modelling future demand for and supply of graduates is very difficult (the report mentioned has four different scenarios). But we are starting from a position of an over-supply of graduates outside of health and engineering, so it is unlikely that we need to significantly increase our student intake beyond current trends except in these areas.

Around 60 per cent of domestic undergraduate offers made by the University of Newcastle are to non-school-leavers. This is important given that the Hunter region has a lower proportion of the population with a bachelor or higher qualification when compared with the whole of Australia (21 per cent vs 30 per cent nationally).

The median household income in the Hunter is also significantly lower than the national median income ($904 vs $1245 per week).

These data, and similar stories from many regions around Australia, suggest it may be wise to pause before applying a one-size-fits-all model that would likely deter precisely those applicants with the most to gain from access to an affordable university degree.

Needless to say, such a narrowing of access would also limit the capacity of Australia to build a productive and competitive economy.

Norton’s analysis also largely ignores the various safeguards for protecting low socioeconomic status students for which the British education sector lobbied successfully during the development of the new fees system.

It would be possible to subsidy/price discriminate for low SES students. However, we found no evidence of unusual price sensitivity among Australian low SES students and so did not pursue this idea.

These include the government’s Office for Fair Access, the increased salary threshold for fee repayments and the commitment to bursaries for students from the poorest households. Norton’s view seems to be that a national salary threshold of $49,000 before repayments begin is a sufficient safeguard. The 2005 Callender and Jackson study published in the Journal of Social Policy, however, identified poorer students in the UK as being more debt-averse than their wealthier peers.

That study was an opinion survey of school students, and highlights the limits of such surveys in predicting behaviour. See the applications data (red line) for subsequent years. And it raises the basic issue of whether when people hold attitudes are not in their own best interests we should indulge their misjudgments, or try to change them. I would think it should be the latter – especially for people we hope to include in a place where they should learn to put evidence and analysis above intuitions and impressions.

The Graduate Winners model assumes that every student will be willing to take on significantly greater debts than they now accumulate on the promise of higher earnings in future.

Most students would pay between $7,000 and $19,000 over their course. It will knock 1-2% off the median rate of return.

National participation rates in higher education for low SES students increased by 24 per cent between 2007 and 2011, and for indigenous Australians the increase was 26 per cent. It is not clear why shifting the long-term funding responsibilities for higher education from the state to the student will do anything but reverse this progress.

We could price discriminate, but I think low SES students are smarter than these theories assume.

As a university with the highest number of enabling programs in the country and with proportions of low SES and indigenous undergraduates close to double the national average, we at Newcastle consider fair access measures to be non-negotiable in any discussion of fee deregulation.

Beyond the materiality of the proposed changes to higher education, the power of the symbolism in the reforms suggested in the Grattan report run counter to the principles on which Australia’s universities were founded.

To charge students the full cost of a degree sends a clear signal that Australia’s universities are the natural homes for only those who can afford to study, and implies that knowledge is only ever traded rather than shared.

Not at all. I am surprised that these pre-HECS assumptions survive despite the loan scheme’s success.

It is time for a broader discussion based on the principles that a world-class university system and intergenerational mobility are both critical if we are to build our productivity and competitiveness on the world stage.

Caroline McMillen is the vice-chancellor of the University of Newcastle.

5 thoughts on “Misreadings and criticisms of Graduate Winners

  1. Hi Andrew,

    Thanks for your insights. Do you have any response to McMillen’s claims: “Implicit in its calculation of public versus private benefit is…that all students who study subjects that have historically produced high-earning graduates will also earn high salaries”?

    I think this is a legitimate concern and may overestimate the private returns of higher ed. For one, the signalling effect of a bachelors degree for today’s 21 year old will be much less during their career than it was for a graduate 40 years ago when we had an elite higher education system. Therefore, I would doubt that we can reliably predict a steep and steady salary increase for today’s graduates based on cross-sectional data.


  2. Peter – One finding from our research is that graduates don’t need to earn especially high salaries relative to their peers to come out ahead. And if private returns are sinking, initial tuition charges are the least of their problems.

    However, I don’t think declining private returns are very likely unless we get a much large increase in attainment than seems likely. McMillen can’t have it both ways on this – if demand for skills is going up, then all other things equal salary premiums should also be maintained or increased.


  3. Andrew – The argument is not that private returns will continue to decline with higher attainment, it is that they have already declined.

    For example, Graduate Careers Australia report that starting salaries of bachelors graduates has dropped from 100% of FT male average weekly earnings (AWE) in 1977 to 78% in 2011. The decline appears to be steady and 2012 is the lowest on record. Even if higher ed attainment remains stable, it is highly unlikely that today’s graduate will achieve the same returns on higher ed as the 1977 graduate.


    Still, we would probably find that the relativity between year 12-educated and AWE has worsened even more so since 1977. So it doesn’t necessarily point to a weaker return on higher ed relative to yr 12 education. But, I’d say that again it is the signaling effect that is the primary reason why it is much more risky to not complete a bachelors degree.


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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s