In an earlier post, I argued that student interests drive course choices, but also that more than 80 per cent of first-year students hope for improved employment outcomes.
It follows from this that, within their cluster of interests, prospective students would plausibly choose courses with the best apparent employment and income outcomes.
This theory helps explain trends seen in applications data. While supply-side and timing problems mean we do not always have enough ‘job-ready graduates’, rarely do we lack ‘job-ready applicants’.
The most common and persistent skills shortages in the Australian labour market are in health-related occupations. It has sometimes been hard for both the education and migration systems to keep up with demand. Health care and social assistance employs more people than any other industry, including 500,000 jobs added in the last decade. Government estimates included in the Tehan higher education reform discussion paper predict that this growth trend will continue.
In line with these labour market needs, health courses have significantly increased their share of university applications since 2010, as seen in the chart below. Charts at the end of the post show that this is a nearly 30 year trend.
Most health students will get student contribution reductions under the Tehan plan, but no financial incentives via altered student contributions were needed to produce the changes in enrolment share since 2010.
Nursing did have a special discount price relative to other fields between 2005 and 2009, but a Deloitte analysis in 2011 was not convinced that this explained changes in application share in that period. Subsequent growth in nursing applications supports the conclusion that medium to long-term job prospects, not course fees, drove student preferences towards nursing.
I do not, however, think that much of this was at the expense of other university courses. Nursing attracts mature-age and lower-ATAR students who would not otherwise have gone to university. Even without counting single-preference direct applications, nursing has the highest-rate (68 per cent) of second and lower preferences being in the same field. Nursing is the only university course that interests many nursing applicants.
For smaller fields of education, important changes are hard to see in an applications share analysis. Engineering lost applications share between 2010 and 2019, but the chart below shows that during the mining boom engineering applications grew 20 per cent from their 2010 base, but then fell back when the boom turned to bust. At the margins, people with engineering in their cluster of interests were first steered towards it, and then against it, as employment prospects fluctuated.
Teacher education also lost applications share in the last decade, as it did in the early 1990s as well (first and last charts in this post). As the chart below shows, undergraduate teacher education applications are in absolute as well as relative decline. In both cases, falling applications were triggered by deteriorating employment outcomes (a big data file, for those who want the detail).
Prompting the latest teaching decline, full–time employment rates went from a recent peak of 82.9 per cent of those looking for FT work in 2008 to 70 per cent in 2014. Negative media stories highlighting poor outcomes would have influenced the choices of prospective teaching students.
Recently teaching employment outcomes have not been too bad. So why are applications still going down? NSW and Victoria mandating minimum academic entry requirements is one factor. The LANTITE literacy and numeracy test, which is blocking some teaching graduates from professional admission, is also likely to be deterring potential applicants. These are obstacles to the desired employment outcome.
The only major 2010-2019 applications share trend with no possible direct relationship to objective employment trends is the rise of science. As I have long argued, in most detail in Mapping Australian higher education 2016, never-great science employment outcomes turned bad as the over-supply of graduates produced by the applications and enrolments boom entered the labour force.
I believe, however, that there is an indirect relationship with employment for some students, who hope for entry to postgraduate health-related courses, or transfers to undergraduate health-related courses for which they lack the necessary ATAR for post-school admission.
Science is the main case study for arguing that student contributions can potentially influence applications. From 2009, the Rudd government slashed science student contributions to increase demand. The chart below shows an immediate sharp spike in applications. But in 2013, with the Budget in a bad way, science student contributions were put back up again. As the chart shows, increased student contributions had no obvious impact on demand. Despite a couple of pauses along the way, science has enjoyed a decade-long boom.
For science, the difference between 2009 and 2013 is publicity. The student contribution price cut in 2009 was combined with general promotion of STEM, which created an impression that jobs were plentiful. In an applications pool that was up 5 per cent overall between 2008 and 2009, science went up by 17 per cent and IT and engineering each went up 10 per cent. In 2013, the student contribution increase hardly received any publicity.
A low-profile price change was a weak signalling rival for continued, if often inaccurate, promotion of STEM job opportunities. In March this year, to use one recent example, the science minister used misleading data to encourage women into STEM courses. And of course in the higher education reform package science student contributions will again be discounted, this time by $2,000 a year, because science is a ‘national priority’.
Career-focused, school-leaver university applicants would almost never be rationally influenced by student contributions. Even if students pay the full teaching cost, as they almost will in law, business and humanities if the Tehan package is legislated, it will amount to a small percentage of career earnings for most of them. If money worries them, the relative job prospects and likely wages of any potential course in their cluster of interests will affect their long-term finances much more than any student contribution changes, up or down.
With a loan scheme like HELP, student contribution discounts are, in themselves, a redundant mechanism for steering students towards ‘job-ready’ courses. If good jobs really will be available, a much larger financial incentive is already in place.
If price signals work at all for career-focused students, the effect is primarily from the associated publicity. Shifts in university applications are usually not mysterious; prospective students hear about changes in the labour market and absorb other information relevant to their course choices. At the margins, this publicity influences decisions about whether to apply for university, or which courses applicants list as their first preference.
Because courses choices are more open to influence from publicity than prices, there are much cheaper ways of steering demand than attention-seeking changes in student contributions. Reductions in STEM student contributions will save students about $250 million a year. A marketing campaign costing 1 per cent of that amount could influence people who are unsure about their direction, without delivering windfall gains to students who would choose STEM anyway.
Provided information about employment prospects is as reliable as it can be, within the inherent limits of our ability to predict the future labour market, there is nothing wrong with trying to guide student choices. I hope the National Careers Institute proves to be helpful. But using student contributions to influence student decisions is not cost effective. This is one reason why, although I share the government’s goal of wanting to increase student places, I don’t think the reform package should be implemented in its current form.
Supplementary applications charts