Getting Past the Cost:
Making University Education Accessible to All
by Andrew Norton
Click here for PDF version

Contrary to popular belief, deregulation of higher education would improve not reduce access to a university education for those from low-income backgrounds.

In mid-2000 The Australian newspaper renewed the national angst over income inequality in a series of reports run over several days. Amidst much talk of globalisation, economic rationalism and the death of the Ôfair goÕ, only one contributor to the paperÕs coverage mentioned a major underlying cause of income inequalityÑhigher education.

The universities are now educating twenty times the number of students they did in the late 1940s, and have been essential institutions in a long-term shift to highly skill, highly paid jobs. With each generation this century, the proportion of people holding these jobs has risen. This in turn has helped create the largest group of well-off people in AustraliaÕs history, but has fuelled a growing perception of income inequality.

Elizabeth Webster at the University of Melbourne has looked at the types of jobs held by men aged 35 to 39 in four birth categories: 1911-1915, 1931-35, 1946-50 and 1961ø65. For the eldest generation, who reached their mid to late thirties in the late 1940s, just over 10% worked in highly skilled occupations as managers, professionals, para-professionals and technicians. For the next generation it was slightly more than 20%, for the next nearly 30% and for the youngest generation around 35%. Over 80% of highly skilled workers have an educational qualification of a bachelor degree or above (Webster 1999).

As well as providing better jobs, higher education offers significant insurance against unemployment. In 1999 the unemployment rate for those who did not complete secondary school was 10.8%. For those who completed school it was 7.7%. For those with at least a bachelor degree unemployment was 3.1%, while 1997 data shows that unemployment is below 2% for those with degrees in some disciplines.

To a large extent, relative material privilege in Australia is now linked to university-level education, and so the distribution of that education is an important political issue. One fear being exploited by opponents of higher education reform is that deregulation would reduce access to this privilege for those of modest financial means.

Opposition to higher education charges

More than a decade after the Australian Labor Party (ALP) reintroduced charges for university education we still hear calls for Ôfree public educationÕ. In the political mainstream some charges are accepted, but there is a powerful fees phobia. In October 1999, the Coalition ruled out deregulating fees without specifying any reasons. The ALPÕs August 2000 platform states that deregulating fees would reduce opportunities for low- and middle-income families.

The ALP assumes that their university policy supports general goals such as a fair distribution of Ônational economic successÕ. If their policy were only to increase opportunities for low- and middle-income families they would be right. But the chosen policy mechanism of large subsidies undermines their fair distribution goal.

The reason for this is that, as can be inferred from the employment statistics discussed earlier, higher education subsidies must have regressive effects on lifetime income distribution. Those with degrees are more likely to be employed and to hold the best jobs, meaning that their total working life income will typically be higher than the income earned by less-educated members of the workforce. By subsidising higher education and thereby reducing costs for degree holders, the ALP transfers money to affluent people.

The best possible justification for this redistribution to the well-off is that only cheap education induces people from low-income groups to undertake education, and so become socially mobile.

This assessment ought to be challenged. We have reasons to believe, and evidence to show, that low-income people are prepared to invest personally in higher education where the benefits are seen to exceed the costs. The main obstacles to increased higher education participation are not prices but the culture of working and lower middle class Australia, the quality of schooling, and the quota system limiting the number of university students.

Who goes to university?

While the majority of university students come from relatively affluent families, their affluence is far from being the only thing to influence their educational decisions. It has long been believed that middle class parents aspire for their children to attend university and do much to help them by sending them to private schools, and by providing encouragement, support and advice.

Research by the Centre for Population and Urban Research at Monash University confirms that culture is an important influence on university attendance. Using data from the 1996 census, they are able to identify university attendance rates of 18 and 19 year olds living at home by household income group and by parental occupation, usually of their father. This shows that for children of labourers and tradesmen income levels seem to make no substantial or consistent difference to university attendance rates. Household income exceeding $50,000 a year actually has a negative impact on attendance.

This pattern is reversed for children of other kinds of workers, but parental occupation still makes a large difference. For children from homes earning $1,000-$1,499 a week, 21.7% of the children of clerical workers attend university, 30.2% of the children of managers and administrators, and 37% of the children of professionals.

A smaller proportion, 29.4%, of the children of clerical workers with household incomes above $2,000 a week attend university than do the children of professionals earning less than $500 a week, with 31.4% attending (Birrell et al. 2000: 55). To some extent, clerical families carry their attitudes to education into affluence, and professional families carry theirs into relative poverty.

Cultures can be very persistent, but they are not immune to the power of financial incentives. Economic changes are arguably driving up the higher education participation rates of lower-income families. These incentives are probably working in two directions.

From the top, we have the very visible mass affluence of a professional and managerial class that is increasing rapidly in size, adding nearly 400,000 to its workforce between 1993 and 1999 alone (Cully 1999: 101). The wealth available from membership in this group, and the sense that it is relatively open, must be drawing people to university education.

From the bottom, we have deteriorating labour market prospects for teenagers, which changes the work-education trade-off they make. Overall, the number of full-time jobs for teenagers more than halved in just a decade from 1988 to 1998, lowering the possibility of financial benefit in working rather than studying.

At the same time, it became easier for teenagers to support themselves while studying as the number in part-time jobs increased by more than 60% (Lewis & Mclean 1999: 18). Studying was also made more attractive through student allowances more than doubling between the early 1980s and the mid-1990s (Lewis & Koshy 1999: 49) The possibility of parental support probably also increased, with long-term increases in real household income in all income groups.1

In these economic conditionsÑincreased financial appeal of education, decreased financial appeal
of leaving school, and greater resources available to support young people studyingÑan economically-minded person would expect an increase in university attendance across all class groups. That is exactly what has happened. The evidence for this comes from a national panel study carried out by the Australian Council for Educational Research (Long et al. 1999).

This panel looked at four categories of young people born in 1961, 1965, 1970 and 1975, and divided them up in several ways, including family wealth assessed on the nature of their home and possessions rather than income directly.

For the oldest group, who reached university entrance age in 1980, 16% of those in the lowest 25% of family wealth went on to higher education, compared to 29% of those from the wealthiest 25%.Ê By the time the youngest category reached university entrance age in 1994, 27% of the poorest group went to university, a leap of about 69% in less than 15 years.

Do students from poorer families get less from education?

If students look at their higher education decision in an economic way, they will weigh up the costs and the benefits. The difficulty is that while expenditure can usually be approximated, it is less easy to do so with the benefits.

Graduates may, on average, have better jobs and less unemployment, but averages can conceal considerable variation between individuals. Fearful of the risk that they may end up below average, some individuals may err on the side of caution, and so underinvest in higher education.

The Higher Education Contri-bution Scheme, known as HECS, was introduced in 1989 as a way of charging students but managing the risk of higher education investment. The risk was managed by making repayment of higher education debts contingent on the student reaching a minimum level of income, and then collecting not a set sum but a proportion of that income.

In other words, the idea was that those who never get much of a return on their higher education investment never pay back much either. The HECS system also controls risk through a flat debt charge with no real interest rate, meaning debt only grows by undertaking more education. 2

With this risk controlled, participation in higher education should, from an economic point of view, depend on a combination of short-term affordability (being able to pay living expenses) and long-term returns through higher income. If short-term affordability is cared for through family support, part-time work and welfare, is there any reason students from low-income backgrounds might believe they will earn a lower financial reward from their education?

In Australia, no research exists, to my knowledge, that directly studies either the perceptions of students from low-income backgrounds about their earnings prospects, or of the outcomes they actually achieve. But we can approach the issue in other ways.

One reason for low-income students to doubt their future earning potential is that without parents or friends with professional or managerial jobs, graduating low-income students are likely to have weaker social networks in the job market.

According to the Australian Bureau of Statistics, around 15% of professional and administrative jobs were found through friends, relatives and company contacts, and between a quarter and a third had employers approach them. In both cases, existing social networks were important.

While people without networks are at some job search disadvantage, this is not an insurmountable obstacle. About 40% of jobs were found through means available to any person with a bit of initiative (figures: ABS 1998: 20).

The role of networks in job hunting also suggests that on-campus education is particularly desirable for people from low-income backgrounds. The campus environment provides people from all backgrounds with an excellent opportunity to expand and diversify their social networks.

Another way of assessing the job prospects of low-income students is to see if there is any clear connection between universities that have large proportions of low socioeconomic status (SES) students and those that achieve poor graduate employment and income results. As can be seen in Figures 1 and 2 (see pages 6 and 7), a statistical examination of this using Graduate Careers Council of Australia data on unemployment and federal Department of Education, Training and Youth Affairs (DETYA) data on the home SES area of students shows only a very small relationship, with a 1% increase in low SES proportions increasing underemployment by 0.3% and negative though negligible effects on annual starting salaries.

A further possibility is that low SES students are less likely to complete their degrees, and so perhaps get a lower return on the money they invested before dropping out. DETYA has done a study of students who enrolled in 1992. It found that there were only
small differences by socioeconomic status in completion rates by 1997. The most disadvantaged females had a 70.9% chance of completing and the most advantaged had 73.9%. For males the difference was larger, 61.6% compared to 67.9% (Urban et al 1999: 39).

These figures, however, overstate the direct influence of socioeconomic status while at university. The DETYA researchers found that once they controlled for other factors such as Year 12 results there was little socioeconomic difference in completion rates. This suggests that socioeconomic status has its effect earlier, while the students are still at school.

Government schools are less successful than Independent schools in achieving good Year 12 results. Only 44% of Victorian government school students received a Tertiary Entrance Rank (now renamed ENTER) of above 80 in 1998, compared to 67% of students at Independent schools. The difference is even more marked with TERs 90 and above, with twice the proportion of Independent school students achieving this result compared to government school students (Birrell et al. 2000: 59).

For those going to university on a low TER score, DETYAÕs analysis suggests that they are less likely to complete. After grouping the TER scores into ten deciles, they found that those in the bottom decile had nearly three times the non-completion rates as those in the top decile (Urban et al. 1999: 31). As well as reducing completion prospects, low TER scores limit access to professional degrees with high graduate incomes such as law, medicine, commerce and engineering.

Figure 1. Low SES and Median Starting Salary/Underemployment

Should students from low-income families be deterred?

On the basis of this evidence, students from low-income backgrounds probably will, on average, earn less from their education over their lifetimes than students from affluent backgrounds. The major reason is not their social background in itself, as they can at least partly remedy social network shortcomings while at university, but their relatively weak Year 12 results, which make them less likely to complete and put them into less lucrative courses.Ê

If students do complete, however, there is no reason to believe that their earnings would be uneconomic, provided they do a degree with vocational possibilities. The Melbourne Institute for Applied Economic and Social Research has done research on the dollar returns to education. It found that the average graduate is just under $300,000 better off over their working lives from having gone to university rather than finishing education at the end of Year 12 (Borland et al. 2000: 22).

The Industry Commission has done some by profession calculations of the rate of return to higher education under a full-fee system, and came up with figures of a 6.6% return to a teacher, 9.4% for an engineer and 11.7% for a lawyer (Industry Commission 1997: 98, 101).

The major risk factor then is not that the returns from completion will not be high enough, but that completion will not occur. There are many reasons why people do not complete. Not all are related to factors universities can do anything about, so claims in this area must be moderate. Nevertheless, there are things potentially within the universitiesÕ control.

The first step in lowering attrition rates is to get a better match between student and course, or between student and institution. A recent survey of first year students found that 32% of them did not get into their course of first choice, and of this group nearly a quarter received their fourth or fifth choice. One in five students hoped to change to a different course after their first year (McInnis et al. 2000: 15).

Universities will always reject a proportion of those who apply. However, this figure of a third not getting their choice suggests that a more flexible system, including letting in new providers in a way that did not put them at a competitive disadvantage, would do a better job of avoiding this initial source of student dissatisfaction. Newcomer universities, particularly, would be interested in enrolling those students unable to enrol in their desired course in one of the existing institutions.

A second step would be to improve the quality of university life, both academic and social. While a comfortable two-thirds majority of first-year students agree that the quality of teaching in their degree is good, it tends to be the follow-up matters that cause most discontent.

Some 47% of first years disagreed with the statement that the academic staff take an interest in their progress, while 62% were not fully satisfied with the availability of teaching staff, and only 25% fully agreed that teaching staff give them helpful feedback (McInnis et al. 2000: 48).

ÔEmotional healthÕ is the single most commonly cited reason for considering deferring. While many of these emotional problems probably have their sources and solutions off campus, there is a concerning minority of first-year students who are not engaged with the university community. Nearly a quarter say they have not made close friends, and just over a quarter say that they generally keep to themselves. Similar proportions say they feel uncomfortable participating in class discussions and that they are not interested in extra-curricular activities (McInnis et al. 2000: 36)

The first-year survey also asked universities about what programmes had been effective in helping their new students. These included assessed writing early in semester to identify students with problems, peer support programmes, and student-to-student mentoring. These in addition to fixing staff availability problems would make a difference.

Clearly efforts are being made to be more responsive under the current system, but that system is not conducive to them in two major ways. First, it provides no reliable stream of revenue sufficient to staff the universities adequately. Student-staff ratios are up around 19 to 1, compared to 7 or 8 to 1 in top American universities and 12 or 13 to 1 in top British universities. There are too few staff to provide every student with sufficient academic support, let alone pastoral care.

Second, even if there were extra resources under the current system, the incentives for focussing them on student concerns are too weak. For universities with more students than their government quota (which is most of them) the students at the margins are more likely to represent a cost than a source of revenue, making some drop-outs desirable. Turning marginal students into sources of potential revenue rather than costs could do much to concentrate minds on their needs, which in turn would make university a better option for low-income students.

Are low-income students debt averse?

A common argument against fees with loans is that low-income students are debt-averse (Andrews 1999: 13). Yet the proportions of people from low-income groups enrolling in higher education increased after HECS loans were introduced, so clearly not all low-income people are debt averse. The concern can only be with a sub-group.

Les Andrews has analysed this problem by looking at the attitudes of low-income groups to other forms of debt such as mortgages and personal loans. He found that rural people were more reluctant to apply for loans, but concluded that Ôit appears that the SES background of groups had no strong or consistent effect on their level of debt aversion as measured by their willingness to apply for new mortgages or personal loans and the amounts involvedÕ (Andrews 1999: 17).

Logically, people should be less debt averse than normal when it comes to HECS-type loans for education. Unlike other types of loans, educational loans are in the long-term income producing, generating their own capacity to pay for themselves. Unlike other loans, there are no set repayment schedules, which means that the risk of the loan causing financial difficulties is lower.

The experience of the HECS scheme shows that rising numbers of low-income students accept this logic. To the extent that residual debt aversion remains, the constrained nature of the Australian educational market must in part be to blame.

Unlike post-compulsory school education, in higher education the government effectively limits the number of students it will assist through a quota system. Historically, it has set the quota well below actual demand, which means that few universities need worry about filling their places. Since universities are near-guaranteed as many students as they can cost effectively take, there has never been a great need for intensive efforts to explain the benefits of higher education to people from groups with traditionally low higher education involvement.

Figure 2. Low SES vs Underemployment (%)

Abolish quotas

In my view, the quota system of allocating student places undermines the incentive to market higher education to low-income groups, exacerbates course-student mismatch, and promotes student attrition.

Bad as these effects are, they do not make clear its worst effect, which is simply to exclude people from university altogether. As long as the quota system remains in place, increasing the proportion of low-income students will be a zero-sum game with wealthier families, since the number of places is finite. Unfortunately for low-income students this is a game that they have little hope of comprehensively winning, given the well recognised capacity of private schools to increase their studentsÕ Year 12 marks.

In the long-term only cultural change and school reform can significantly increase the proportion of low-income students attending university. In the short to medium term the most promising way of lifting the number of low-income students at university is to abolish the quota system, thus giving them places and the universities an incentive to attract and retain them.

Why subsidise?

From a traditional equity point of view higher education subsidies seem like a bad idea. If there are income-contingent loans, and short-term affordability is dealt with, the appropriate measure of need is lifetime income. The public policy case for giving more to those who are already on average $300,000 ahead is, to put it mildly, not strong.

The reason for a general subsidy, at least in the short to medium term, is caution over what might happen if fees rapidly increased by a large amount. A reasonable guess is that some people would abandon their educational plans, to their own detriment and, in the long-term, to the countryÕs detriment.

This would be particularly so if the subsidy was abolished before increased competition between the universities enhanced their capacity to inform prospective students of the benefits of higher education, and before any of the quality improvements likely to flow from deregulation occurred.

This risk was controlled in the reform package put by the federal Education Minister, Dr Kemp, to Cabinet in October 1999. Under the reforms, subsidy levels would have been unchanged. Average annual fees were estimated to increase by $1,000, based on what happens in the already deregulated market for overseas students. So the package protected against large price shocks, while still proposing quick structural change by allowing the price signal to work through top-up fees and abolishing quotas. The only significant policy shortcoming was variable interest rates on student debt. For the risk-related reasons discussed above, flat charges are better.

Conclusion

John Dawkins, the Education Minister responsible for introducing HECS, provided an important turning point in higher education policy. He was the first to launch a large-scale experiment with the idea that places and not prices were the key to bringing low-income groups into tertiary education, simultaneously significantly increasing them both.

His experiment was a considerable success, but he left in the higher education funding legislation a substantial impediment to realising his vision. That impediment was the quota system, which limits competition and makes marginal students sources of cost rather than income.

The irony of the current policy impasse is that the Left, ostensible defenders of the poor, remain implacably opposed to removing this impediment. The reason for this is that abolishing quotas would introduce student choice, which they link to ÔvouchersÕ and fees, which they see as inequitable.

In my view, this logical flow does not work. While choice and fees would be complementary reforms, you can have one without the other. And fees are not inherently inequitable; to the contrary, they enhance lifetime income equality.

Unfortunately the circumstances in which Dr KempÕs submission became public meant that it did not get a fair hearing, even in the government. It is a pity, as this is a case in which free markets and equity walk hand in hand.

References

ABS 1998, Successful and Unsuccessful Job Search Experience, Cat. No. 6245.0, Australian Bureau of Statistics, Canberra.

ABS 2000, Australian Social Trends 2000, Cat. No. 4102.0, Australian Bureau of Statistics, Canberra.

Andrews, Les. 1999, Does HECS Deter? Factors affecting university participation by low SES groups, DETYA, Canberra.

Birrell, Bob and Angelo Calderon, Ian Dobson, T. Fred Smith 2000, ÔEquity in Access to Higher Education RevisitedÕ, People and Place 8(1): 50-61.

Borland, Jeff, Peter Dawkins, David Johnson, and Ross Williams 2000, Returns to Investment in Higher Education, Melbourne Institute for Applied Economic and Social Research, Melbourne.

Cully, Mark. 1999, ÔA More or Less Skilled Workforce? Changes in the Occupational Composition of Employment, 1993 to 1999Õ, Australian Bulletin of Labour 25(2): 98-104.

Industry Commission 1997, Submission to the Review of Higher Education Financing and Policy, Industry Commission, Canberra.

Lewis, Philip and Paul Koshy 1999, ÔYouth employment, unemployment and school participationÕ, Australian Journal of Education 43(1): 42-57.

Lewis, Philip and Ben Mclean 1999, ÔThe Youth Labour Market in AustraliaÕ, CLMR Discussion Paper Series 99/1, Centre for Labour Market Research, Murdoch University, Perth.

Long, Michael, Peter Carpenter, and Martin Hayden 1999, Participation in Education and Training 1980-1994, Australian Council for Educational Research, Melbourne.

McInnis, Craig, Richard James, and Robyn Hartley 2000, Trends in the First Year Experience, DETYA, Canberra.

Urban, Mark and Emily Jones, Gregg Smith, Christine Evans, Maureen Maclachlan and Tom Karmel 1999, Completions: Undergraduate outcomes for 1992 commencing students, DETYA, Canberra.

Webster, Elizabeth. 1999, ÔOccupational Profiles of Men Since 1947Õ, Melbourne Institute Working Paper No. 23/99, Melbourne Institute for Applied Economic and Social Research, Melbourne.

About the Author
Andrew Norton is a Research Fellow with The Centre for Independent Studies and Director of its Liberalising Learning programme.

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)