Category Archives: Higher education

OECD comparisons are a poor guide to higher education funding policy

Rodney Tiffen’s Inside Story article on university funding reflects conventional wisdom: that there should be more public funding of higher education. But as is common in these pieces, it only really alludes to possible arguments, rather than spelling them out and exploring their implications.

We start with the familiar claim that Australian spending on higher education is relatively low, as a share of GDP, compared to other OECD countries. There are problems with the data used as it does not include the significant subsidies via the HELP loan scheme, but the basic point is right: on a per place basis Australian students pay more than is common in the European countries that make up most of the nations in the analysis.

Tiffen describes Australia’s low position as a low rank, when in reality the information on public spending is just a list. A rank only makes sense if higher numbers are inherently good and low numbers are inherently bad. But even relying purely on OECD numbers the available evidence counts against this conclusion. Consider higher education attainment rates compared to fees. As the chart below shows, low attainment is mostly found in low fee countries.

I have described this before as the paradox of public funding: that high subsidies can lead to low attainment. A low fee policy is usually aimed at demand-side considerations. But in higher education the problem is usually the supply of student places rather than demand for them. When public funding is limited, it is better to give smaller amounts to a larger number of people (eg Australia) than larger amounts to a smaller number of people (eg many European countries).

fees and attainment 2
Note: 2003-04 fees data used to approximate fees likely to be have been paid by the cohort in the attainment data.

This is a paradox, not a political science law. As the Scandinavian countries show, it is possible to have no fees and high attainment. But this typically occurs as part of a much broader political and economic system. High taxing countries have high public spending on social services, low taxing countries have high private spending on social services. Again OECD data can show the pattern as it applies to higher education.

fees and tax

For graduates, the difference between paying high taxes generally and paying extra ‘tax’ via HELP repayments is likely to be subtle. It is possible that the free-on-delivery mode of education shifts education expenses to a later period in life, when high annual tax bills are more manageable. That could be a benefit of the Scandinavian system, but I have not seen any research that demonstrates the point. But there are mostly likely to be significant issues when there are high taxes and high fees, as seems to be the case in England.

OECD comparisons are heuristics rather than arguments. If Australia differs markedly from other countries it is worth asking if this matters. In Australia’s case, I’d say the answer is no. Fees that are high compared to other countries, but modest compared to private benefits, have helped us deliver higher education to the masses on a tax take that is less than many other OECD countries.

IT, engineering and economics students least satisfied with teaching quality

Yesterday I reported data from the 2014 University Experience Survey suggested that students at non-university higher education providers were, on most indicators, more satisfied with their education than students at universities.

There are also significant differences between disciplines on satisfaction with teaching quality, as seen in the chart below. I have taken out disciplines with fewer than 1,000 respondents in the UES, as well as most ‘other’ disciplinary categories as too vague. This took out both the discipline with the highest satisfaction (language and literature, 89%) and with the lowest (mechanical engineering, 72%).

satisfaction by discipline

Most of the relatively low-satisfaction disciplines are popular with males and international students, who report lower overall satisfaction than females and domestic students. But I can’t tell on the available data which way the causation might be running – whether students in engineering, IT and commerce faculties are less satisfied at least partly because they tend to be male and/or from overseas, or because males and international students are less satisfied because they are enrolled in engineering, IT and commerce faculties.

Non-uni higher education provider students more satisfied than uni students

The 2014 University Experience Survey results have been released in the last few days, and this time they included students from non-university higher education providers (NUHEPs). A total of 1,444 students from 15 NUHEPs completed a survey. Given that there are about 130 NUHEPs the results aren’t conclusive, but they are interesting.

As can be seen in the chart below, NUHEP students were generally more satisfied with their educational experience than university students. Each of the categories includes multiple related questions that are combined to produce an overall satisfaction rate. For example, the teaching quality scale contains questions on whether teachers explained things clearly, gave helpful comments on work, whether assessment tasks challenged students to learn, and other similar topics.

NUHEP satisfy

The area where university students are more satisfied than NUHEP students is ‘learning resources’ which includes questions about the quality of teaching spaces, library facilities, online learning materials, the quality of student spaces and common areas, and related topics. Possibly the big university campuses with their economies of scale are better on these things.

The positive responses on ‘learner engagement’ are noticeably lower for both groups. For example, only 53 per cent of students said they had a sense of belonging to their university. The recent first year experience survey picked up a negative trend in this area.

Although there is room for improvement in some areas, for most questions responses were more positive in 2014 than 2013. That supports the conclusion of the end-of-degree course experience questionnaire (trend data at p.76 of Mapping Australian higher education) that teaching quality in Australian universities is slowly but steadily improving.

Should higher education providers be responsible for HELP doubtful debt risk?

I’m glad that Liberal Democrat Senator David Leyonhjelm is pushing the government to look at HELP doubtful debt. But I’m not convinced that the policy response discussed in a letter from Christopher Pyne to Senator Leyonhjelm is the right one.

Like several similar suggestions in the last year, the latest proposal aims to make universities partly responsible for student debt. Universities are encouraging students to take HELP loans, while transferring all the risk to students and taxpayers. It seems only fair that universities take some of the risk themselves.

As reported by Fairfax Media,

Mr Pyne proposes “a mechanism to make a proportion of each higher education provider’s direct grant funding contingent on its performance against a key set of indicators” – including the debt not expected to be repaid (DNER) by their graduates.

There are several kinds of risk factors related to student debt. As Grattan’s doubtful debt report showed, projected repayment rates differ by course. Someone with a medical degree earns more than someone with an arts degree, for instance. Other risk factors relate to personal characteristics. Men, for example, are more likely to repay than women because they spend more years in the full-time labour force, on average. Higher-ATAR students are more likely to finish their degrees and probably get higher paying jobs. There are macroeconomic risks, where recessions or periods of slow growth mean that repayment rates decline (as is happening now). Then there are risks associated with particular providers, in how well they prepare their students for work and how much they do to help them find jobs.

It only makes sense to penalise higher education providers for repayment issues they can reasonably foresee and do something about. There probably are things they can do to make their graduates more employable, although nobody knows what long-term impact these have. In Mapping 2014-15 we did not find major differences between types of university for employment levels, although we did find salary differences. But I expect it is going to be fairly difficult to identify a unique provider effect at the required level of confidence, amidst all the many factors affecting employment outcomes.

Providers could offer fewer places in non-vocational degrees, or degrees leading to occupations that appear over-supplied in the labour market. But in a fee deregulated market its not clear why providers wouldn’t factor the doubtful debt cost into the fees charged, knowing that this is an expense they share with their competitors.

Targeting personal characteristics is complicated, because some risk factors such as being female are protected under anti-discrimination law. Plus the government is saying that they would take account of equity factors such as gender. Charging lower fees that women have a better chance of repaying before they leave full-time work is a non-discriminatory way of targeting this doubtful debt risk. However, as with course-related risks it is not clear why providers would not charge high fees, knowing that some of the profits need to be set aside to later fund a doubtful debt levy. This is especially the case because they won’t want to charge males lower fees just because females have a higher doubtful debt risk.

Low ATAR or other forms of academic disadvantage aren’t protected by anti-discrimination law, and of course this is the most common form of discrimination in the higher education system. But with low SES students being over-represented among those with low ATARs, a policy that encouraged general discrimination against low ATAR students would run contrary to equity goals and policies. We are far better off tackling the low-ATAR problem directly, rather than encouraging action via doubtful debt penalties.

Repayments are affected by macroeconomic conditions, but we should not penalise providers for circumstances they probably can’t predict and certainly can’t avert.

If we are going to move to assessing students for their risk of doubtful debt – a major conceptual change to HELP – we should not do it via providers. They are never going to have as much information about lending risk as the government itself. The government can access huge amounts of information to assess likely repayment prospects. Pyne’s letter to Leyonhjelm also proposes linking ATO and Education Department data, something that should have been done long ago, and which will allow a far more nuanced understanding of who is likely to repay. Adding in social security and immigration data would help too.

The government also has much better incentives than the providers. After all, if a HELP debtor doesn’t pay back all the money left outstanding is the government’s loss. The higher education provider is only putting a share of their money at risk, and as noted above there will still be incentives to take on high-risk students.

I don’t think we should rule out a more individualised approach to HELP lending. Arguably, in VET FEE-HELP and parts of the higher education market the government is currently an irresponsible lender, letting people take on debts that aren’t in their interests. But if we are worried about HELP doubtful debt, as we should be, there are things we can do that are easier, faster and will save much more money. Regular readers you have heard it all before, but:

* Lower the first threshold (in the defeated bill);
* Index all the thresholds for repayment by CPI instead of AWE. This would speed up repayment, which is important due to women departing full-time work in their thirties.
* Remove the write-off of debt on death. Our report on this last year concluded that much of the debt written off will be held by women in higher-income households, who will have estates that can easily repay.
* Require repayment from overseas debtors (although this is probably small for doubtful debt).
* Introduce loan fees (like those currently applying for some students) to encourage upfront payment. Money that is never lent cannot turn into doubtful debt.

For some reason, complex ideas that only indirectly tackle problems seem to be winning favour at the moment over simple ideas that directly tackle problems. Making higher education providers responsible for HELP doubtful debt risk is another of these.

A three bills strategy could break the higher education impasse

The government’s higher education reform package faces its second defeat in the Senate this week. Over the last week controversy has focused on the government’s threat to not renew $150 million in funding to the National Collaborative Research Infrastructure Strategy (NCRIS). This is being denounced today as reckless and bullying.

While threatening not to fund NCRIS looks like a tactic to pressure the Senate to pass fee deregulation, it also reflects the narrowing of options to achieve another important goal of the Pyne package, which is to control total spending on higher education. This has increased rapidly in recent times, with the main teaching subsidy program up 40 per cent in just five years. The previous government tried to introduce cuts, the current government is trying, and the next government will try. With big Budget deficits forecast, higher education cannot escape attention.

In my view the issue is not whether higher education will be cut, but how to do it in ways that cause least harm to public policy objectives in teaching and research. The situation we are in now is that programs are going to be cut according to whether or not the measures need parliamentary approval, not whether the affected programs are low priority or could be funded in other ways.

NCRIS is vulnerable because unlike most other university programs it is funded via the annual Budget appropriations bill. All the government has to do is not include its funding in the bill and the money is saved. The Parliament will not get to consider a bill on NCRIS. I think the demand driven system is also vulnerable in the medium term. From 2017 funding agreements between the government and universities could be used to freeze funding (which is all the government is trying to do; the 20 per cent cut proposed to per student funding is what is needed to control spending while student numbers grow). The Parliament has no power over funding agreements, so they are an obvious way to save money.

The minister doesn’t want to pursue either of these options. He has made much of securing conditional support for NCRIS, which Labor had left out of the forward estimates. He likes the demand driven system so much that he wants to extend it. But with internal government requirements to deliver savings, these programs are among the few feasible options if the Parliament won’t co-operate.

The multi-purpose bill higher education bill currently before the Parliament – with Budget measures, the demand driven reforms and fee deregulation – is complicating efforts to get a good outcome. If Senators are opposed to any of it, they will vote against all of it. Fee deregulation faces the strongest opposition, so leaving it in the bill is an obstacle to achieving the bill’s other two objectives.

In a couple of Senate inquiry submissions I have explained the different sources of the Pyne package and suggested proceeding with them separately.

I think we need three bills: Budget measures, demand driven reform, and student funding rates. One of my submissions suggests reforms to HELP that are not in the current bill but which would save money, lessening the need for big cuts to per student subsidies. While it would still be difficult to get a higher education savings bill through the Senate, it would have a better chance than one that includes fee deregulation. It would also give the government a better political position on NCRIS – indeed, it could include a special appropriation for NCRIS so that it is clear that this is just a Budget issue, and not a tactic around fee deregulation.

So far as I know, none of the cross-bench Senators opposes the demand driven reforms. If there were offsetting savings in the Budget legislation, the government could bring a separate demand driven bill forward and expect it to pass the Senate.

Fee deregulation as it stands will almost certainly fail to pass the Senate. As there is no consensus on alternatives, we need more time to work through the issues. The universities want the extra revenue deregulation would bring, but I don’t think this is urgent for 2016. While the government has invested a lot of political capital in fee deregulation as the big idea of the reform package, a sequenced legislative strategy would maximise their chances of achieving something in higher education in this parliamentary term.

Should high university fees be taxed?

If domestic undergraduate fees are deregulated most people, including eminent education economist Bruce Chapman, believe that at least some universities will charge significantly higher fees than now. Chapman has now detailed a proposal to tax excessive fees, to ‘inhibit and limit the extent of price increases’ (number one in this list of Senate inquiry submissions; The Australian‘s version here.)

The basic idea is that the government will establish different bands of fees, which are taxed at different rates – the tax being a reduction in grants that would otherwise be payable to the university. To take an example from Chapman’s paper, fees for humanities up to $6,499 a year (a bit higher than current student contributions) would pay no tax, fees between $6,500 and $11,499 would pay 20% on the margin, fees between $11,500 and $16,499 would pay 60% on the margin, and fees of $16,500 and over would pay 80% on the margin.

The effects of this can be seen in the context of UWA’s plan for a flat $16,000 fee for all undergraduate courses. The tax would be about $1,000 for the $5,000 in the first marginal section, and another $2,700 for the $4,500 up to $16,000. With current subsidies of around $5,500 a year for humanities courses, UWA’s subsidy would be reduced to around $1,800. (For high fees in low subsidy disciplines, the fee tax could mean that the government taxes more than it contributes for that discipline).

Chapman is not endorsing these particular tax rates; they are to illustrate the concept. However, I am not sure that conceptually this is the best way to target the problem of high fees. First, we need to be clear about what the problem is with high fees.

As Chapman says, it is likely that some fees will be well in excess of the costs of teaching. Much of the profit is likely to fund research. There are two public policy problems with this. The first is that students/graduates will incur higher private costs without a commensurate increase in private benefits. The second is that higher fees will generate higher costs for taxpayers, through the interest subsidy on HELP debt and HELP debt that won’t be repaid.

To solve the first problem, the tax policy relies heavily on deterrence. To the extent that universities do charge taxable fees the problem is exacerbated – the money goes to the government, which is even less likely to benefit the student than the university spending money on research. Research spending might at least contribute to the general prestige of the university and the graduate’s qualification.

To solve the second problem, the tax policy is likely to be more effective as it raises revenue that will offset some of HELP’s interest and bad debt costs. However, it means that students who pay upfront are compensating for costs that they won’t generate. Other students who do borrow could over-compensate. Using the tax rates in Chapman’s submission, and a fee of $30,000 for a law student, we estimate a tax of more than $11,000, leading to government savings of $3,000 in excess of the additional HELP costs.

If we are worried about higher private costs without increased private benefits, it might be better to target university spending rather than revenue. In the UK and USA universities report on spending classified according to function (teaching, research etc) that allows us to see the relationship between student-driven funding and spending. If we did that in Australia we could prohibit public universities from moving beyond certain ratios between student funding and spending, and taxing them if they did. That way the student isn’t any worse off than he or she would otherwise have been, since the money wasn’t being spent on them anyway, and it is only the university’s profit being taxed.

For HELP costs, we should tackle HELP’s problems directly rather than focusing on the students paying high fees. Loan fees payable only by those who borrow would assist in dealing with HELP’s costs without hitting the students who pay upfront. Plus there are several other ways of controlling HELP’s costs, as I have pointed out many times before.

Policy considerations aside, this is a complex policy when the government needs a clear, simple and positive case for fee deregulation.

Should government benefits be increased when university fees go up?

Fairfax has a story this morning on the hidden cost of deregulating university fees. Higher education is included in the bundle of goods and services that make up the consumer price index, which in turn is used to index a wide range of government welfare benefits. So if fees increase the CPI will go up, driving up the cost of the social security system. This was an issue in England when their university fees went up.

I am quoted in the story as saying that the government could exclude university fees from the index. I was challenged on Twitter about this.

The CPI is based on a basket of goods and services consumed by households, with the primary input being the ABS Household Expenditure Survey. A well-known criticism of this is that consumption patterns vary significantly between household types. For this reason, the ABS also calculates a range of other indexes for different household types, especially different categories of government beneficiaries. An aged pensioner index has been used, but only when it is more than CPI.

Given that the purpose of indexation is to maintain the real purchasing power of benefits, it is not clear why people should be compensated for an increase in prices in a commodity few of them other than Youth Allowance recipients are likely to consume. This is a general point about the choice of indexation methods, not one just restricted to this particular issue. But there could be special legislation to at least avoid the once-off major spike in prices after the system changes increasing government expenditure on welfare benefits.

—–
Update: Some of my economist colleagues are more sceptical of the inflationary effects. They argue that the Reserve Bank has an inflation target and they take policy action to keep CPI within it, even though it is common for there to be price spikes in particular services or commodities. So while the cost of higher education would go up, this could be offset by price stability or reductions elsewhere.

Pyne package linking domestic and international charges is about subsidies more than fees

When announcing its higher education reform package, the government said that international student fees would be a cap on domestic fees. This idea has been criticised regularly since, including today by Gavin Moodie who notes that such a rule could easily be gamed.

But the draft guidelines released with the reform bill version 2 last December show that legislation is not mainly about capping total fee levels, but trying to ensure students benefit from tuition subsidies.

The problem here is not that domestic students are likely to be charged more than international students. At Grattan we have collected fee information for both domestic and international students for hundreds of courses where there is no regulation requiring domestic students to be charged less than internationals – for postgraduates in public universities, and in all full-fee courses in non-university higher education providers. There is not a single case where domestic students are charged more than internationals, and only a handful where they are charged the same. Presumably a mix of underlying cost differences, market forces, and mission considerations mean that domestic students are not charged more.

There is no practical need to cap domestic fees with international fees, and that isn’t what the government is trying to do. Rather, it is trying to ensure that its tuition subsidies benefit students instead of providing super-profits to universities. So what the guidelines say is that tuition fees for non-Commonwealth supported students (which includes internationals) must be at least the student contribution plus the Commonwealth contribution, in some disciplines a much bigger number than just the student contribution.* It is phrased as a floor price for international/other full-fee students rather than a maximum fee for domestic students.

Take an engineering degree at a Group of Eight university, where we calculate that the average annual international student fee was $33,000 in 2014. If the rule just said that international student fees were the cap for domestic fees, a university could in theory charge a domestic student $32,900 and then add the $12,000 tuition subsidy, giving them revenue per student of $44,900, way more than they get for an international student. Even if we assume a more moderate domestic fee of say $26,000, with the tuition subsidy added that still leaves the university with revenue of $38,000 per student, $5,000 more than for an international student.

However, under the rule as drafted the university could not get domestic fee revenue of $38,000 ($12,000 subsidy plus $26,000 fee) per student without lifting international student fees to $38,100, which might price them out of that market. A university might be prepared to take that loss in courses where there are few international students. But in courses where there are significant numbers of internationals the rule will ensure that domestic students benefit from the tuition subsidy bringing down the fees they pay, rather than delivering windfall gain revenue to the university.

The proposed rule on fees for full-fee students has weaknesses as a guard against excess fee charging. But I think it is at least interesting in thinking about what subsidies are for in a fee deregulated system. It takes international student fees as a rough guide to the true market worth of a course, and then tries to ensure that the tuition subsidy brings down the price to domestic students.

* The legislation uses the term tuition fee rather than student contribution now, but I will keep the old language to separate the concepts more clearly.

The asbolute number of lower-ATAR students is still small

In The Conversation today, I have some suggestions about how to handle the increasing willingness of universities to make offers to lower-ATAR applicants. One point I should have made (other than just noting that many lower-ATAR students reject their offers) is that the absolute number of lower-ATAR offer acceptances is not that high, despite a high growth rate since 2010 – about 3,500 in 2014, out of 86,500 acceptances by school leavers admitted with an ATAR (or about 4 per cent). The trend in lower-ATAR absolute numbers can be seen in the slide below.

Comparing acceptance data and enrolment data during the demand driven review last summer there were significant discrepancies between lower-ATAR acceptances and enrolments, indicating drop outs before the HELP census date. If the past is a guide, nearly a quarter of those who made it to the census date won’t return in second year, and just over half will complete. So the absolute number of lower-ATAR students in the system is lower than these acceptance numbers might suggest.

lower ATAR offers
Source: Department of Education applications reports

The other interesting thing about this chart is the sharp increase in applications. As cut-off ATARs began to fall with the enrolment boom it is likely that school leavers who had previously dismissed higher education as unrealistic began to think it was possible, and put in an application. Of course universities also alerted them to this possibility. An example from Victoria University is below.

lower ATAR

Is the prospect of higher fees deterring university applications?

The number of applications to university for courses commencing in 2015 has attracted more interest than usual, due to the controversy over higher education fees. Some data has already been released by individual tertiary admission centres, but it is now available in consolidated form. The figures are preliminary, reflecting applications made as of October 2014. Based on recent history, there will be tens of thousands more applications lodged after this date. I am still seeing plenty of university advertising aimed at that goal.

There is a particular complication this year in Western Australia. A change to the school starting age in 2003 has flowed through the school system, leading to a Year 12 cohort that was only about 60 per its normal size. This makes the WA figures hard to interpret, and the report presents trend data with and without WA.

Without WA, school leaver applications are up 2.2 per cent. Possibly this could be interpreted as saying that the fear of fees has had little or no impact on demand. That’s probably right, although the apparent upward trend may be due to people who would have taken a gap year starting in 2015, so that they get at least one year on the fixed student contribution rates. We also don’t know exactly how many students completed Year 12, so we cannot calculate an application rate.

Non Year-12 applications are down 6.5 per cent. However, this may not mean anything at all. For non-Year 12 applicants, there is a longer-term structural shift away from using tertiary admission centres and towards applying directly to universities. Since 2010, the number of TAC non-Year 12 applications has declined every year, while the number of direct applications has increased.

The report also raises the possibility that the demand driven system might have reduced a backlog of unmet demand for higher education. It is plausible that as people who had unsuccessfully applied to higher education in the past get admitted the pool of higher education hopefuls will diminish. And as more people get into university straight from school, there is a smaller potential market for mature-age higher education.

While these theories may be right, it is still possible that there will be no decline in overall non-Year 12 applications when we get the direct applications data later in the year.

If the demand driven system survives it will be our best yet test of theories in this area. Under the old system, the supply of places was always well below demand. Unless there was a huge decline in demand any price sensitivity would not show in enrolment numbers. We therefore had to use applications data to assess underlying demand. But applications are an imperfect proxy for a serious intention to pursue higher education. Large numbers of people reject the offers they receive, raising questions about whether some apparent demand for higher education is really just keeping options open, or contingent on an offer for a very specific course. Actual enrolments in a system without supply constraints will be a better guide to the true level of demand for higher education.