The non-ether origins of a $42,000 HELP repayment threshold

The Australian‘s High Wired column reports on an exchange between education minister Simon Birmingham and his Green shadow Sarah Hanson-Young on how the government’s proposed $42,000 threshold for HELP repayment was set:

The Greens’ Sarah Hanson-Young was in her element in Senate Estimates yesterday. First up, she wanted to find out where the $42,000 new HECS repayment threshold came from. ‘Plucked out of the ether’ was the answer she was looking for. Which she didn’t get, but the Grattan report on reducing the HECS threshold was mentioned. (And where did that figure come from? Plucked out of the ether, maybe.)

The $42,000 figure was proposed in this Grattan report released in March 2016. There is no science that says exactly what the initial HELP repayment threshold should be. There is always going to be some policy and political judgment involved. But our report did make a non-ether case for a substantial lowering of the initial threshold, which at the time the report was released was based on policy and political judgments made in 2003 (the base level) and 1993 (AWE indexation).

In the intervening years just from 2003, HECS had become HELP and turned into a very different, and much larger, program – with many more students in the core public university undergraduate programs, thanks to the demand driven system, and many extensions of income contingent loans – to higher education students outside the public university system, to diploma students in vocational education, OS-HELP, SA-HELP and other loan systems such as for student income support that use the same basic repayment system.

Total HECS/HELP lending more than tripled between 2003 and 2016, and the risk of non-repayment also substantially increased as we brought in debtors with weaker earnings prospects than was the case with a smaller, more educationally elite, group of eligible borrowers.

It’s sometimes said that the threshold should be high to provide a financial benefit before repayment is required. Arguments like that were made in the late 1980s, and have stuck in popular understanding of HELP. But it is not clear that there is any principle behind this idea. In our report, we argue for seeing HELP in the context of other government income protection programs, rather than a special, very generous deal that graduates should receive for unclear reasons. $42,000 is still a bit on the high side by that standard, but we took into account previous Labor statements that $40,000 would be too low. We want a stable system, so looked for a threshold that Labor would at least keep in office, even if it opposed it on introduction.

So while $42,000 is not pure science, unlike the current threshold it is based on something more than long ago policy decisions made with different circumstances in mind.

Effective marginal tax rates and HELP threshold reform

Media reports in the last few days have pointed out that some women face effective marginal tax rates above 100% under proposed changes to the HELP thresholds. But this is not new – the nature of the HELP repayment system is that for every HELP debtor, male or female, there are income zones that reduce take-home pay.

This is because as income reaches each threshold the debtor must repay a percentage of his or her entire income. Under the current thresholds, a debtor earning $54,868 in 2016-17 will repay nothing. But someone earning $54,869 will repay nearly $2,200. So up to an income of about $57,000 HELP repayments mean the debtor takes home less money than if he or she earned $54,868.

Of course, in most cases the debtor’s long-term financial position is not worse as their HELP debt is reduced. In this sense, HELP repayments are different from the welfare benefit phase-outs that normally contribute to high effective marginal tax rates – for the debtor, that is money lost to consolidated revenue, rather than improving their personal balance sheet. But for people concerned with immediate cash flow more than their long-term financial position, HELP repayments can cause problems.

Research by Richard Highfield and Neil Warren shows that HELP debtors manipulate their taxable income to avoid or delay repayment. They also make the point that this is costly for income tax revenues, even if the HELP debt is eventually repaid. Read more »

Thought leaders and public intellectuals in the ideas industry

I have never liked the term ‘thought leader’. But Daniel Drezner’s new book, The Ideas Industry, persuades me that even if the language is unappealing the concept is useful in describing how the contemporary world of ideas works. In some cases, the category of people known as thought leaders can also make the marketplace of ideas more effective.

Drezner argues that the marketplace of ideas is much larger and more open now than it was in the post-war decades. Technology is one obvious reason; anyone with an internet connection can now publish and social media can be used to bypass the old publisher and broadcaster gatekeepers to audiences. Drezner covers this, but I think the most interesting parts of the book are about how even though it is more possible now than in the past to promote ideas on a small budget, this is also an era of for-profit ideas.

The language of thought leadership is used most by consulting firms, which publish reports as part of their branding – these are the trends and problems your industry faces, come to us for solutions. Drezner says McKinsey spends $400 million a year on these activities. Locally, PwC, Deloitte and others advertise their thought leadership in various fields.

While government spending on consultants seems volatile, there is little doubt that they play a much bigger role in advising governments than they did in the past. So their ‘thought leadership’ is likely to transmit directly to government this way. (Drezner has a chapter on how economics is more influential than other social sciences; that consulting firms are big employers of people with economics degrees is another route for economists to influence government).

Drezner contrasts ‘thought leaders’ with the older term ‘public intellectual’. He has a table of what he sees as the distinctions between them: Read more »

The boom in HELP debtors

The latest ATO taxation statistics come out today, giving us some new information on HELP.

Although growth in higher education student numbers moderated in 2015, VET FEE-HELP was still out of control in the period covered until mid-2016, contributing to a substantial increase in total debtor numbers. They grew by nearly half a million (a 24% increase) between 30 June 2014 and 30 June 2016. With big policy changes in vocational education taking effect in 2016 and 2017, along with continued moderate growth in higher education numbers, the rate of growth should slow substantially in the year to 30 June 2017.

On current policy settings, however, the number of debtors repaying is not likely to accelerate rapidly. Only 22 per cent made a repayment in 2013-14 and the number is likely to to be lower still in 2014-15 (there is a 2014-15 number in the chart, but these have a history of significant upward revision due to late tax returns, so it is too early to say exactly what proportion made a repayment).

The 2014-15 repaying share is likely to be lower because of the number of people who are still students, the high initial repayment threshold of nearly $55,000 a year is delaying repayment for recent graduates, and a large proportion of VET FEE-HELP borrowers are unlikely to earn enough to repay.

There is speculation that the initial threshold for repayment will be reduced in the Budget. These numbers explain why the idea needs considering.

How bad is student mental health?

The media this morning is reporting some dramatic figures on student mental health. According to Headspace, a youth mental health organisation, two-thirds of students responding to an NUS survey reported high or very high psychological distress over the last twelve months, and a third experienced thoughts of self-harm or suicide.

I’ve recently been doing some reading on this subject, as part of a project on student attrition. I’m convinced there is a problem, but I am not so sure that it is this bad.

The latest ABS National Health Survey from 2014-15 shows that the main student demographic, those aged 18-24, has worse mental health than any other age group. This is only moderately so for men, with 11 per cent reporting high or very high psychological distress on the Kessler scale, compared to 10 per cent for all adult men. But it is dramatically so for women, with 20 per cent reporting high or very high psychological distress, compared to 13.5 per for all adult women.

According to other ABS sources, about half of 18-24 year olds are students. An analysis of an earlier NHS (along with HILDA) found small overall differences between students and non-students, especially after adjusting for demographics, with students being on average younger and more likely to be female than the general population. The NHS survey found 10 per cent of its student sample were experiencing high psychological stress, while it was 21 per cent in HILDA.

One important reason for different results is that the ABS surveys and HILDA ask about mental health issues over the last month, while the Headspace/NUS survey asks about the last twelve months. Mental health issues tend to be episodic, so we should expect more people to experience one over a year than a month. There seem to be widely varying results even from surveys with similar methodologies, so mental health is clearly something that is difficult to measure at a population level. But Headspace/NUS getting triple the rate of the ABS or HILDA seems high. They have not published their detailed results, so perhaps we will have a better idea of why when they do.

How students study influences course completion rates

My concern about low-ATAR students is primarily about their high risk of not completing a degree. But in this we tend to focus much more on underlying academic ability than on the circumstances in which students study. Completions analysis shows that off campus and part-time students have non-completion risks that are very similar to those faced by below 50 ATAR students, as the slide below shows.

The number of commencing off-campus and part-time students greatly exceeds the number of commencing low ATAR students, as the next slide shows. We are doing some more work at the moment to try to understand the interactions between these factors. In 2015, for example, 24,000 commencing bachelor students were both part-time and off-campus – are these cumulative risk factors, or both just proxies for students with other commitments? Distraction from study could be the actual main risk factor.


Source: Department of Education and Training (domestic students only)

Low ATAR is an issue. But if we are worried about increasing attrition, we have to focus on how students are engaging with their studies and not just how well they did at school.

—–
Update: The day after I wrote this post, the Department released new completion statistics. They are slightly worse than the numbers shown in the first chart.

Low ATAR enrolments are increasing, but still a small share of all commencing higher education students

This week the main university offer rounds start, and with them the annual ATAR controversy. If the last few years are a guide, demand from low ATAR school leavers has stabilised, as the chart below shows. Offers, however, are increasing. Although most low ATAR applications do not result in offers, the number which did more than quadrupled between 2010 and 2016, from less than 2,000 to more than 8,000.


Source: Department of Education and Training

Last year, the number of low ATAR students accepting their offer increased, but still less than half of offers were accepted, which was about a quarter of the original applications.

When I looked at this last year, I said that a significant proportion of low ATAR applicants who accepted their offer were not enrolled after the first census date. While this is a factor in eventual enrolments, we now think this is less of a factor as some students who are reported as having an ATAR during the applications process don’t have one recorded in the enrolment data. A declining share of school leavers are being admitted based on their secondary education. This has dropped from 87 per cent in 2012 to 80 per cent in 2015.

Of the students who do have an ATAR in the enrolment data, those with low ATARs mostly don’t come direct from school. In 2015, there were 2,830 below 50 ATAR students enrolled who had completed school in 2014 (this includes students in pathway diploma courses, not just bachelor degree courses). But in the same year there were 8,000 students with below 50 ATARs in the enrolment data. Of those not being admitted based on their secondary education, about two-thirds were admitted based on previous higher education or vocational education.

Low ATAR enrolments are increasing, and I am among those who thinks that this is an issue. However, it is an issue that needs to be kept in perspective. Only 3 per cent of commencing bachelor degree students have ATARs below 50, and only 7 per cent have ATARs below 60.

Loan fees and the expert panel

As I expected, there has been comment (here, here, or here) about my release of a report on HELP student loan fees at the same time as I am on a government higher education policy advisory panel.

Due to the panel consuming my time for the last couple of months, the loan fee report has appeared later than originally intended. But other than that the report’s release follows a plan developed a year ago to complete reports on two weaknesses in HELP’s finances, the thresholds for repayment and interest costs. They are companion reports for our report on doubtful debt and recovery of HELP debt from deceased estates in 2014.

Loan fees are not new, and nor is my support for them something I suddenly arrived at after being appointed to the panel. I said in response to the proposed Pyne reforms that rather than abolishing loan fees we should extend them. What this week’s report does is work through in more depth whether interest subsidies are necessary to income contingent loan schemes (no); the relative merits of real interest, hybrid real interest and CPI indexation, and loan fees (loan fees better); and arrive at a method for setting a loan fee rate (likely interest costs over the life of HELP loans). It’s the detail that is appearing now, not the broad recommendation to use loan fees.

When the government asked me to be on the advisory panel I said I could only do it if I could also meet my existing Grattan commitments. They agreed to this. The panel is providing private advice to the minister and the department. It is a different situation from a review with a final report I need to agree with my panel colleagues and which the government will publicly accept or reject. So while the timing is not ideal, the loan fee report does not pre-empt other people’s decisions.

Should teacher education places be capped?

NSW education minister Adrian Piccoli has long been a critic of universities over-supplying the teacher education market. In this morning’s Australian, he is calling for caps on student places. If accepted, this would be the second course after medicine to be capped.

It would also be a major precedent, as it would set a low benchmark for justifying capping. According to employment surveys education graduates do slightly better than average in finding full-time work. Among those finding FT work, education graduates do significantly better than graduates in other fields in getting jobs that use their qualifications. That said, full-time employment for education graduates is down 11 percentage points on 2008, the recent peak of graduate employment rates.

The strength of the demand driven system is not that student and universities will always make the right call about where the job market is going, but that it can adapt as new information becomes available. Over the last few years, the message that the teacher market is saturated has been well publicised. Commencing student numbers were already past their peak by 2015, as the chart below shows. The trend would have been further down except for a major move by Swinburne Online, which went from no students in 2014 to 8.5 per cent of the national commencing market for initial teacher education in 2015.

commecning ed students
Source: uCube

There are bigger falls in education more generally – down 10% percent in commencing students between 2014 and 2015 and 13 per cent in full-time equivalents (presumably part-time enrolments at Swinburne Online are affecting that). Under the legislation, capping occurs in full-time equivalent places, not on a head count of students.

The initial 2016 applications and offers data suggests a fall of 2.4 per cent in applications and 4.7 per cent in offers for education, so a further drop in student numbers seems likely.

Cutting student numbers under the pre-demand driven system was a slow, politically painful process. With demand driven funding, it is happening quickly with few people even noticing.

Should we use the OECD’s analysis of the private financial benefits of tertiary education?

I’m quoted this morning in The Australian‘s report on graduate earnings across the OECD, which is in the latest issue of Education at a Glance.

The reported numbers seemed low compared to work Grattan and others have done for higher education, and I have had a bit more time since to work out why.

An issue I noted in the Oz is that the analysis included people with diplomas. In 2012, diploma holders were 28 per cent of everyone with a diploma or higher qualification. Their lower average earnings will bring down the overall average.

Another issue is that the OECD’s data source may be understating graduate income. They used a source I had never heard of for analysing educational returns, the ABS Disablity, Ageing and Carers survey. It was a general population survey so the issue is not that it is a sample of graduates with a disability. However, looking at the way the unit record data is made available to researchers it seems income is only available in ranges, the top one of which is $1,730 a week or more. We hit this problem in the 2011 census as well, with their top range of $2,000 a week or more. 11 per per cent of diploma holders, 21 per cent of bachelor degree holders, and 33 per cent of postgraduate degree holders reported incomes of $2,000 a week or more. As some of these would have incomes well over $2,000 a week, the average is artificially held down by the income category cap.

The OECD numbers are net present value, which means that income expected to be received in the future is counted as of less value than income received now. There is plausible time value of money theory for discounting the future – for example, an 18 year old prospective student would probably rather receive $1,000 now than $1,100 when they finish their 3 year degree, even though there is a favourable implied interest rate on offer.

But in our Graduate Winners report that was not the way we presented the data, which we left undiscounted in the key sections. This was partly because the undiscounted number is easier to understand, and partly because despite the plausibility of time value of money theory in various contexts I was not sure it was so persuasive in this one. In my view, one reason people pursue higher education is so that they will have a good job and a high income in 30 years time. How much theoretical sense does it make to heavily discount the value of achieving a major objective?

Some interesting data on male hourly earnings by years of experience from the latest HILDA report highlights this issue. For the first five or so years, male graduates don’t earn much more per hour than men with vocational education. But after that time a wide earnings gap develops – in the later years that are most discounted by the OECD methodology.

male hourly earnings

The discounting also affects another issue, which is that they assume students don’t work while studying, and the consequent assumed forgone earnings appear with a low discount and are deducted from gross earnings. But in Australia most students work while studying, so the forgone earnings cost is exaggerated, while future income benefits are under-valued.

There is no perfect method of doing educational returns analysis, and every data source in Australia has limitations. But overall I think the OECD numbers are less useful than existing Australian research on the financial benefits of education.