REVIEW OF HIGHER EDUCATION FINANCING AND POLICY

AUSTRALIAN RESEARCH COUNCIL INFORMATION PAPER

The Council's mission is to provide advice on research funding and research policy, and to promote the conduct of research and research training of the highest quality for the benefit of the Australian community. The Council has a special responsibility for research in the higher education sector, basic research, and research training in the higher education sector.[1]

Since the ARC provides advice to the Government on matters which will be covered by the Report of the Review of Higher Education Financing and Policy, it is not appropriate for the Council to make a formal submission. To assist the Review Committee, however, this document provides information and analysis relevant to its Terms of Reference, and to major themes identified by the Review Committee as they relate to the research and research training responsibilities of the Council. The Council understands that this information paper will be publicly available.

Key Points

THEME ONE: THE ROLE OF HIGHER EDUCATION RESEARCH AND RESEARCH TRAINING IN AUSTRALIAS SOCIETY AND ECONOMY

• The universities conduct research across the full spectrum, but they have a special responsibility for pure basic research, which underpins the rest of the nation's research effort. University research represents one quarter of all national research expenditure.

• The Council has identified five major benefits of research; its contributions to the quality of our culture, the education of graduates of high quality, the direct application of research results, increased institutional capacity for consulting, contract research and other service activities, and the development of international links.

• Universities are the sole source of research-trained graduates, who are responsible for a significant proportion of the total output of higher education research, and acquire generic skills that equip them both to undertake research, and to be valued in a wide range of occupations beyond research.

• ARC support, allocated on the basis of merit and demonstrated capacity as determined by expert peer review, dominates competitive funding for university research and is the principal avenue for direct support of pure and strategic basic research in Australia's R&D system as well as fostering university-industry collaboration in research.

THEME TWO: FACTORS AFFECTING THE DEMAND FOR, AND PROVISION OF HIGHER EDUCATION RESEARCH AND RESEARCH TRAINING

• Levels of research activity among Australia's 36 publicly funded universities are very unequal.

• The average value of an ARC Large Grant has increased in real terms, and the success rate has been between 20% and 25% in recent years. The principal alternative to the Large Grants Scheme as a source of funding for basic research in universities is the Small Grants Scheme where the success rate is significantly higher, but the average grant has declined in real terms.

• Higher teaching loads have followed from deteriorating staff-student ratios, and many academics are now seeking funding for teaching relief.

• The balance between competitive external funding through ARC programs and internal funding through institutional operating grants appears to be appropriate when considered in the light of international comparisons.

• The so-called 'teaching-research nexus' is difficult to support at the undergraduate level, but good university teachers need to be abreast of research in the field, without having to be actually engaged in it. With respect to research degrees, however, a teaching-research nexus is clearly required, with supervisors needing to be both research trained and active researchers.

• Growth in higher degree by research (HDR) load for the Unified National System (UNS) as a whole has been very rapid in recent years, especially in the post-1987 universities. Moreover, there is a considerable gap between research activity and income among the new universities, and the research higher degree load they are carrying.

THEMES THREE AND FOUR: REGULATION, ADMINISTRATIVE FRAMEWORK AND FINANCING OF HIGHER EDUCATION RESEARCH AND RESEARCH TRAINING

• There are three regulatory mechanisms governing HDR load. The annual profiles process does not currently directly identify an HDR load target for institutions. Some years ago, the profiles process was used to inject substantial additional HDR load. Subsequently, efforts were made to prevent shift of load from undergraduate places to HDR places. These have not been very effective.

• Australian Postgraduate Award (APA) awards that provide a HECS exemption in practice cover about 80% of HDR. About 20% of HDR receive a stipend Award. Increases in HECS exemptions and stipend scholarships have followed growth in total HDR load.

• ARC funding recommendations are delivered by the Department of Employment, Education, Training and Youth Affairs (DEETYA). While the National Health and Medical Research Council (NHMRC) operates similarly, such arrangements are unusual in international terms. The Government has announced that it intends to re-establish the Australian Research Council as an independent body delivering its own programs.

THEME FIVE: FUNDING OF HIGHER EDUCATION RESEARCH

• General University Funds (GUF) represent an average of 64% of all research expenditure by higher education institutions. An estimate of the imputed salary cost in 1994 GUF is $700 million, 58% of the total. The methodology for the collection of these data is, however, questionable.

• The Research Quantum (RQ) within operating grants has not kept up with growth, and has declined following recent cuts in operating grants. While formally the RQ represents the whole operating grant contribution toward research, it is unlikely that it even covers non-salary expenditure on research.

• Grant support from Commonwealth agencies generally only covers the direct cost of research. Project-related infrastructure is provided through the Research Infrastructure (Block Grants) Program. A significant increase in support through this program will be required to achieve a goal of 40c in each dollar of direct grant support.

• Encouraging the strategic development of research consonant with national needs, and of research and research training activity in new and emerging fields, requires funding which complements performance-based schemes.

THEME ONE: THE ROLE OF HIGHER EDUCATION RESEARCH & RESEARCH TRAINING IN AUSTRALIAS SOCIETY & ECONOMY

Higher education research

The strong connection between research and research training is a particular feature of higher education - research training requires concomitant research activity. Nevertheless, research is undertaken within higher education institutions in its own right, over and above its connection to the teaching mission of universities. While not all institutions, and not all academics are equally active in research, a marked feature of higher education research is that most of it is undertaken by staff who also have teaching responsibilities.

Five benefits of research

1. Contributions to the quality of our culture

   Strong, independent universities are a hallmark of a democracy. They are both creatures of the culture and a force which shapes it. Research plays a vital role in determining the nature and extent of that influence.

2. Graduates of high quality

   Research is the vehicle for education and training at the advanced undergraduate and postgraduate (research) student level and contributes, in less specific ways, to teaching at all levels in the higher education institutions. The ARC has the important responsibility of providing the highest quality research environment for such advanced education and training.

3. Direct applications of research results

   This is the most obvious and direct benefit of research; the application often stems from several pieces of research, sometimes in apparently unconnected fields, perhaps undertaken over extended periods of time in different locations. It can lead to the commercialisation of a new product or process in manufacturing, mining or agriculture, or to an improvement in environmental management, health care or social welfare.

4. Increased institutional capacity for consulting, contract research and other service activities

   Through the acquisition or construction of sophisticated equipment, through the improvement of the library and other information systems, as well as through the advancement of the skills and knowledge of the staff and students, the capacity of institutions to offer consulting, contract research and other service activities is enhanced. This represents a powerful resource for industry and the community in general.

5. International links

   For a country like Australia, with a small population, a highly developed and respected research community can provide an important channel of information on research developments in other countries, and the country's influence in international forums is enhanced by its standing in research.

Basic research & economic performance

Basic research is experimental and theoretical work undertaken primarily to acquire new knowledge without a specific application in view.[5] The United Kingdom Treasury recently commissioned a study of the relationship between publicly funded basic research and economic performance.[6] The report notes that there are two main views of the economic benefits of basic research:

• to perform economically useful research which firms will not undertake because they cannot exclusively capture the benefits (the 'market failure' or 'public good' argument), and

• to provide research training and maintain access to international networks.

1. discovery of useful information,
2. creation of new instrumentation and methodologies,
3. development of new skills among researchers and graduate students,
4. particularly those involved in solving complex problems,
5. providing access to networks of experts and information, and
6. the creation of 'spin-off' companies.

The Industry Commission noted that the body of knowledge created by basic researchers

can be drawn on as required by those wishing to apply knowledge to produce improvements in existing products, processes or services, or simply to understand the social, cultural and physical environment. It can also be drawn on by basic researchers as a prelude to more applied research. The university is the repository of such knowledge.[7]

Universities are also important in establishing durable conduits to the research done outside Australia. This transfer of knowledge from abroad cannot occur without local people working in the area to absorb and disseminate the results. Many of these people are in universities.[8]

Statistical characteristics of higher education research

Source: Australian Bureau of Statistics, Research and Experimental Development All-Sector Summary, 1994-95

This compares with an OECD average share of total R&D expenditure by higher education of 17.8%, and an EU average of 20.6% It is on the high side of average, close to that of Canada (23.8%), below the Netherlands (28.8%) - the most comparable country in GDP per capita terms - and above the United Kingdom (17.5%) and the United States (15.6%).[9]

By type across the various performing sectors, basic research (pure and strategic) represents slightly more than one quarter of Australia's expenditure on research and development. (See Figure 2.)

Source: Australian Bureau of Statistics, Research and Experimental Development All-Sector Summary, 1994-95

Higher Education is the dominant performing sector in pure basic research, responsible for 78% of all expenditure on that type of research. With 39% of all expenditure on strategic basic research, higher education is also the largest performing sector in strategic basic research. (Commonwealth research performing organisations are responsible for a further 30% of expenditure on strategic basic research).

Thus, within the higher education sector, the largest proportion of expenditure is on basic research. Nevertheless, a third of all expenditure is on applied research. Experimental development is largely conducted by the business sector. (See Figure 3.)

Source: Australian Bureau of Statistics, Research and Experimental Development Higher Education Organisations, 1994*

Higher education research is conducted across all broad fields, with the natural sciences, technology and engineering having the preponderant share. Medical and health sciences, and biological sciences have particularly large shares of total expenditure on research. Among the social sciences, education stands out. The humanities represent a relatively small proportion of all research expenditure. (See Table 1.)

General university funds - predominantly Commonwealth operating grant - are by far the most significant source of research funds. Together with Commonwealth national competitive research grants and other Commonwealth funding, the Federal Government totally dominates research support in the higher education sector. (See Table 2.)

Dissecting higher education research by socio-economic objective, advancement of knowledge dominates (48%). This suggests that, in addition to pure basic research, a significant proportion of strategic basic research has been included in this category. The very low proportion of expenditure on defence related research is noteworthy. (See Figure 4.)

Source: Australian Bureau of Statistics, Research and Experimental Development Higher Education Organisations, 1994

Research education and training

HDR completions

Source: DEETYA Selected Higher Education Student Statistics, 1996

There has been a very substantial expansion in the level of research training undertaken in higher education institutions, so that in the decade 1986-95 HDR completions rose 238% from 1,792 to 4,272 per annum, with 26,500 completions for the decade.

Since the end of the binary divide, all Australian universities have been engaged in research training, awarding bachelors degrees with Honours, Masters by research and PhDs. Indeed, they have the exclusive responsibility for research training at these levels. In addition, the higher education sector shares responsibility for postdoctoral research training with government research-performing organisations, such as the CSIRO, and the Co-operative Research Centres. The universities are thus the source of Australian research-trained graduates who will find employment in industry, government, and in the universities themselves.

Total higher degree research (HDR) load in 1996 was 25,718 equivalent full-time student units (EFTSU). About four fifths of HDR were enrolled in universities established before the end of the binary era, somewhat more than half in eight of the oldest universities. Some 4,272 postgraduate students completed research higher degrees in 1995, 2,491 (58%) being doctorates.[10]

By broad field, research higher degrees reflect both the differing importance attached to research qualifications and the size of the field. (See Figure 6.)

Source: DEETYA Selected Higher Education Student Statistics, 1996[11]

HDR employment

The 1995 median starting salary for a male who had completed a doctorate the previous year was $42,000 pa, and for a masters by research $40,000 pa. The median salary for female doctoral and research masters graduates was $38,000 pa. The entry point salary for academic staff at level A holding a doctorate was also $38,000 pa. The equivalent figures for pass bachelors degree graduates were $28,000 for a male and $26,000 for a female.[14] It should be noted, however, that although HDR graduates had higher starting salaries, they will have made major salary sacrifices to undertake their degrees. There are no data on how HDR starting salaries compare with those of bachelors graduates after four years in the workforce, although it would be fair to assume that the gap will have closed at least somewhat. Similarly, there are no data available on the career salary high points of the two groups, although again it might be expected that higher degree graduates would be higher on average.

Across all disciplines, occupations requiring a specialist research qualification represent at least a third of employment destinations of masters by research and doctoral graduates (if one assumes a research higher degree is required of academics). Very likely the percentage is higher; but how much higher is clear from existing data. (See Table 3.)

Source: Graduate Careers Council of Australia, Postgraduate Destination Survey 1995

The employment destinations of doctoral and research masters graduates also vary by field. In most professional fields, practice in the profession is a common employment destination; for example, psychology (22%), social work (100%), architecture (50%), accounting (20%) and law (33.3%), and almost all fields of engineering. In most fields in the natural sciences and humanities, by contrast, university teaching or research work were the principal employment destinations. An interesting exception was geology, where 64.7% of HDR graduates were employed as geologists. HDR graduates in professional fields also attract substantially higher starting salaries than the overall average for HDR graduates.

The Council believes that there is a pressing need for more detailed study of the employment of HDR graduates to determine to what extent their qualifications, qua research qualifications, are required. Such a study would need to track HDR graduates beyond their initial career destination, and ascertain the views of employers as to the importance of research training for the work HDR graduates undertake. The Council proposes to engage consultants to undertake a project along these lines.

Academic researchers

A doctoral degree is held by most academics in the natural and social sciences, engineering and medicine, but is comparatively uncommon among academic accountants and lawyers. The number of academic veterinary scientists and architects is relatively small, whilst the proportion with doctoral qualifications is high among the former, and low among the latter.[15]

Source: DEETYA Selected Higher Education Staff Statistics, 1996

Australian Research Council funding programs

All OECD members, and many other countries besides, have a national research council or councils which, like the ARC, provides a vehicle for governments to direct resources towards higher education research in ways that are consonant with their policy goals. Expert peer review is always employed to provide assurance that support will be given to those most likely to deliver results.

The ARC dominates agency support for university research, providing half of all national competitive grants. (See Figure 8.)

Source: Higher education financial and publications research data collection, 1994 (unpublished)[16]

The ARC Large Grants Scheme is the single largest competitive grants scheme. It is also the primary source of support for project-based basic research in the higher education sector. The Small Grants Scheme, where the universities act as agents for the Council, provides support for research which has a more modest cost. The Special Research Centres Scheme block-funds groups of researchers also working toward the basic end of the spectrum. (See Table 4.)

The ARC is thus the principal avenue for direct support of pure and strategic basic research which, in Australia's R&D system, is mainly conducted in higher education institutions. In consequence of this, the Council has closely examined the role and pattern of support for basic research in Australia. Preliminary advice on The Strategic Role of Basic Research was given in early 1994. The Council then embarked on an extended process of analysis and consultation, beginning with The Likely Profile of Research for Social Economic, Cultural and Environmental Benefits in 2010 (1995), and culminating recently in the Funding Strategies for Basic Research: Part 2 Balance of ARC Funding Across the Disciplines (1996). These studies are concerned to determine whether the pattern of ARC support is appropriate in view of evolving needs, and the ARC's role within the national R&D system.

The Council sought to establish the linkages between basic research and the socio-economic needs of Australia through a major two-phase study, Using Basic Research Part 1: Review of current theory and international practices (1995), and Part 2: Socio-economic connections to academic research in Australia (1996). The latter concluded that

basic research performs an important role in supporting a wide range of Australian social and economic activities. Nearly 70% of the 594 Australian enterprises responding to our industry survey relied considerably on basic research (xiv).

The 1996 Federal Budget saw a substantial increase in the funding provided through the Collaborative Research Grants Scheme, where Commonwealth support is matched by industry. Australian Postgraduate Awards (Industry), where the Commonwealth stipend is supplemented by industry support in cash and kind, are aimed at providing research training directed toward the needs of industry. The focus of the Key Centres of Teaching and Research is the provision of research relevant to user communities, including industry, and the communication of its outcomes to those communities. Key Centres are financially supported to a very considerable extent by their users.

To assist it in developing these programs, the Council commissioned Crossing Innovation Boundaries: the formation and maintenance of research links between industry and universities in Australia (1993). This study found that

Estimated business involvement with academic research is probably five times higher than recorded funding figures suggest, that is, 10% to 15% of total higher education research involves industry linkages (xvii).

The most important role of universities within the national innovation system is in providing broad support for the national knowledge base and developing skilled personnel attuned to innovation needs (xix).

ARC research grants are usually awarded to Chief Investigators who both teach and research. But these grants also support full-time postdoctoral Research Associates and Senior Research Associates. The Special Research Centre staff are also usually full-time researchers. In addition, the Council provides support through the research fellowships program for full-time researchers from postdoctoral to Professorial level. To assist it in judging the appropriate balance between full-time and part-time researchers, project and block funding, the Council commissioned The Effects of Resource Concentration on Research Performance (1993). This study indicated that concentrating resources in groups of block-funded, full-time researchers did not produce increased productivity but did help achieve international visibility.

Commonwealth stipend support for research training at the masters and doctoral levels is provided though the Council's Australian Postgraduate Awards schemes. In 1995, there were 4,572 students holding an Australian Postgraduate Award - APAs (with Stipend or an Australian Postgraduate Award (Industry). (See Table 5.)

The distribution of APAs tends to reflect the relative size and research activity of fields within the higher education sector, whereas the distribution by field of APA(I)s tends to reflect industry need for research trained graduates.

ARC research and research training programs have four strategic objectives:

• supporting basic research and maintaining the research base,

• training and supporting researchers of high quality,

• developing links between industry and higher education research, and

• developing national and international research links.

THEME TWO: FACTORS AFFECTING THE DEMAND FOR, AND PROVISION OF HIGHER EDUCATION RESEARCH AND RESEARCH TRAINING

Research performance

Source: Higher education financial & publications research data collection[17]

Five institutions account for nearly half of research expenditure, and the top quartile for almost 70% of R&D funds. Eight institutions are responsible for 67% of research expenditure, and producing 51% of research publications.[18] (See Table 7)

Source: Patterns of Research Activity in Australian Universities (1996)

The picture is similar if one controls for the size of institutions by looking at the Research Quantum (RQ) as a percentage of the total operating grant. On average, in 1994, it amounted to 5.4% of the operating grant, while the range was from 11.4% for the University of Western Australia to 0.4% for the Australian Catholic University. Three universities are above 10% - Western Australia, Adelaide, and Melbourne. Eight universities are below 1%.

On the other hand some 19 institutions - more than half - were among the leading six in at least one field of research.[19] Research performance is, to some extent, dynamic; this is reflected in the Research Quantum (RQ) within institutional operating grants. The RQ is distributed on the basis of a performance-based formula, with research income the dominant factor, followed by publications and research higher degree completions. (See Table 8.)

Source: Performance-based Funding of Universities (1996)

The number of reported teaching and research staff also varies by RQ quartile, but it will be evident that research performance is not directly correlated with the number of researchers. (See Table 9.)

Sources: Performance-based Funding of Universities (1996)
DEETYA Selected Higher Education Staff Statistics, 1996

Thus while the 1^st quartile of the RQ is also the largest in staff terms, they earn much more than proportional research income, produce many more publications and train many more research higher degree students. The relationship between proportion of academic staff and proportion of research income, publications, and research higher degree students is closer to 1:1 for the 2^nd quartile, and reverses for the 3^rd and 4^th quartiles.

Success rates

Three factors affect success rate: the number of applications, the sum of money available, and the average grant size. Beyond restricting the number of new grant applications an applicant may make to two per round, and the total number of grants they may hold at once to three, Council has not sought to constrain the number of initial applications, which grew by 30% from 2,321 in the 1990 round to 3,012 in 1997. To this must be added applications to the Small Grants Scheme, administered by institutions on behalf of the Council. Applications for Small Grants doubled from 2,027 in 1990, to 4,273 in 1996.

Funding available for new (Initial) Large Grants grew from $20.7m in 1990 (about $25.4m in 1997 dollars), to $35.6m in 1997, a 40% increase in real terms. The average grant in 1990 was $37, 342 (about $45, 776 in 1997 dollars), and in 1997 it was $55, 614, a 21.5% increase in real terms. The number of Initial Large Grants grew from 555 in 1990 to 640 in 1997, a 15% increase. Maintenance of, and indeed a modest increase in the value of the average grant has been a conscious policy of the Council over several years.

In sum then, funding for the Large Grants Scheme has increased significantly in real terms, as has average grant size; and the number of grants awarded has also grown. Nevertheless the level of demand has meant that four fifths of those seeking funding have been disappointed. There are a number of considerations, however, which are pertinent to an assessment of the import of this fact.

The number of applicants for Large and Small Grants is a small proportion of the total potential clientele for the two schemes. There are about 7,000 Large and Small Grant applications each year at present, from the 26,000 teaching and research staff of Australia's universities.[20] About 3,700 Large and Small Grants are current each year. Some Awards have more than one Chief Investigator, and Chief Investigators may have as many as three grants. Nevertheless we may safely infer that only a small proportion of those eligible to apply in fact does so.

Of course, in some fields - mainly humanities and some social sciences - external funding is not required, and many academics are funded through other sources - most notably those working in the fields of medicine and agriculture. Opportunities for funding in applied research generally have expanded substantially, with the ARC's Collaborative Research Grants Scheme and the Co-operative Research Centres Program each providing about $30m annually for research in higher education institutions.[21]

But there is also some evidence that the level of demand for Large Grants is partly a function of the perceived likelihood of success. In the early years after the establishment of the ARC, there was marked volatility in demand, with an increase following a relatively high success rate, and a decrease in demand following the ensuing low success rate. This so-called 'hog-cycle' may have stabilised.

Higher teaching loads have followed from deteriorating staff-student ratios, and many academics seek funding for teaching relief through the Large Grants Scheme. It is universally agreed, though, that there has also been a marked increase in the perceived pressure on academics to engage in research and generate research outcomes.

The principal alternative to the Large Grants Scheme as a source of funding for basic research in universities is the Small Grants Scheme where the success rate, consistently about 50%, is more than double that for the Large Grants Scheme. The average Small Grant in the first year of the program, 1990, was $12,987 (about $16,000 in 1997 dollars). By 1996, funding, demand and the number of awards had doubled, but the average grant had declined to $12,705

Consistently, the high cut-off score for funding in each of the Discipline Panels of the Council's Research Grants Committee has meant that many proposals considered worthy of support cannot achieve it. The outcomes of the grants supported through the Large Grants Scheme have been subjected to a program of ex post evaluation by independent expert peer review across a large range of disciplines, and the high overall quality of these outcomes is a marked theme of the resulting assessments.

One cause of high levels of unsatisfied demand is the balance between competitive external funding for basic research through ARC programs, and internal funding through institutional operating grants. When the ARC was established in July, 1988, its programs were partly funded through a 'clawback' of funds from operating grants.[22] There was substantial growth in demand between the last round of the Australian Research Grants Scheme,1,399 applications, and the first round of the Australian Research Council Grants Scheme, 2,168 applications. The growth continued the following year, when to 2,321 Large Grant applications were added 2,027 Small Grant applications.

This result was an outcome of a deliberate policy of 'concentration and selectivity' inaugurated in the 1988 White Paper on Higher Education, and subsequently endorsed by the Industry Commission.[23] The different success rates between the Small and Large Grant schemes suggests that institutions are not as rigorous in their selectivity. The most recent data suggest that Australian government funding for research through the operating grant, as a percentage of total support for non-military R&D (32.2%), is almost identical with that of the UK (32.0%) and not far from the EU average (35.0%).[24] The balance between competitive external funding and internal funding through institutional operating grants appears to be appropriate.

Qualification patterns among academics

The formation of the UNS has had profound effect on the funding and management of higher education research. Firstly, staff from the former college sector, including from disciplines such as nursing which had little previous research tradition, set about seeking funds for research from within their institutions and external sources. Secondly, staff in the professional faculties of the pre-1987 universities, such as accounting, law and architecture, in which there had previously been little research, responded to the burgeoning research culture and the need to attract external funds.[25]

This process towards the universal PhD in academia has been neither planned nor questioned. Rather it is an outcome of the abolition of the binary line, and amalgamation of colleges into older universities, or transformation of entire institutions, together with the dominant 'hands on' research culture, which now characterises academia.[28]

The Teaching-Research Nexus

But such an interpretation is difficult to sustain.[31] In Australian higher education institutions at the undergraduate level Ramsden and Moses found "no relation or a negative relation between teaching and research at the level of the individual and at the level of the department, across all subject areas."[32] Ramsden reports that nearly 20% of pre-1987 Australian university staff reported they had not published a single journal article in five years, with 10% of academics responsible for more than a third of all publications.[33]

A much weaker form of the teaching-research nexus distinguishes scholarship from research, and insists that good university teachers need to be abreast of research in the field, without having to be actually engaged in it. As the Industry Commission observed,

undergraduate teaching and graduate coursework teaching involve academic staff in research which has the purpose of maintaining familiarity with advances and changes in the body of knowledge in particular fields. This sort of research is often described not as research in its purest sense but rather as scholarship.[34]

Academic teachers... may be researchers, scholars or professional practitioners - of course, these are not mutually exclusive categories. But a teacher who is none of these things is unlikely to be able to provide a higher education.[36]

With respect to research training at the honours, research masters, doctoral and postdoctoral levels, the [teaching-research] nexus is clear and straight-forward. Students, who themselves will be involved in research, need to be supervised by staff who are research trained - normally at least to the level at which they are supervising - and who continue to be active researchers in, or close to, the topics being studied by the students they are supervising.[37]

Research higher degree load

The growth has been even more marked in the post-1987 institutions, where it has averaged 31.6% per annum between 1989 and 1995. Of course, the 1989 base was very modest for these institutions, totalling less than 1,000 EFTSU. By 1995, however, it had grown fivefold, from just over 8% of total HDR load to just under 20%.

Institutional shares of the RQ provide a measure of research activity (see Table 6 above). The fourth quartile are all post-1987 institutions. In the third quartile, 6 of 9 universities are post-1987 institutions. Only two post-1987 institutions were in the second quartile. The top quartile are all pre-1987 institutions. These data indicate there is a considerable - albeit narrowing - gap between research activity and income among the new universities, and the research higher degree load they are carrying.

Source: DEETYA Selected Higher Education Student Statistics, 1996

The dominant element in the Composite Index determining RQ shares is research income (currently 82.5%). For the UNS in 1995 an average of $22,647in research income was available per EFTSU of higher degree research load. Institutions ranged from $44,136 per EFTSU of higher degree research load to $6,276 per EFTSU of higher degree research load. (See Table 10.) It is noteworthy that using this measure, all research income, spreads the post-1987 institutions between the four quartiles more than using National Competitive Grants (NCGs) alone. Thus, for example, the Northern Territory University ranks higher, and La Trobe University lower, when HDR is divided into total research income than is the case when it is divided into NCGs.

Nevertheless, in 1995, the most recent year for which both sets of data are available, while the seventeen wholly post-1987 institutions had 19.5% of all HDR, they reported only 9.8% of research income from all sources used in the Composite Index. This 2:1 ratio also holds at the level of broad field of research, too - except in the social sciences, where the ratio is 1:1 (See Table 11.)

Table 10 Research income per EFTSU[38]

Sources: DEETYA Selected Higher Education Statistics.
Higher Education Financial and Publications Research Data Collection

Table 11 Post-1987 institutions share of HDR & research income

Sources: DEETYA 1995 Higher Education Students Statistical Collection Table 2, Composite Index Collection.

The data in Table 10 permit comparisons between institutions which are broadly similar in their range of research activity. The larger and older pre-1987 institutions, the other pre-1987 universities, the former Institutes of Technology and the other post-1987 universities provide one useful taxonomy.[40] Where institutions have divergent disciplinary profiles, the data in Table 11 enable reasonable comparisons to be made by broad field.

Some of the older universities within the UNS have significant endowment, which they can apply to support research. However, institutions are broadly funded at the same level for the salary cost of teaching and research staff through operating grants. As discussed below, this is the major element in GUF expenditure on research.

THEMES THREE AND FOUR: REGULATION, ADMINISTRATIVE FRAMEWORK AND FINANCING OF HIGHER EDUCATION RESEARCH AND RESEARCH TRAINING

Profiles

At the end of 1990, the National Institute of Labour Studies Study of the Labour Market for Academics predicted a large impending increase in demand for academics. The authors made the assumption that "the possession of a higher degree is a necessary pre-requisite qualification, at least for appointment at or promotion to Lecturer and above."[42]

In response, 17% of additional funded intakes were allocated to HDR places in 1993,[43] and half the 1994 growth places were allocated at the higher degree level.[44]

Nevertheless, reacting to the large increase in postgraduate load in 1992 (24% for HDR), the Commonwealth noted

Some of that postgraduate growth... has come from decisions by institutions themselves to turn places funded at undergraduate rates into postgraduate places, despite the greater costs associated with the latter... . During profile discussions institutions were alerted to the danger of moving too quickly in this direction and of allowing staff aspirations to exceed the institution's capacity to support high quality research training.[45]

Australian Postgraduate Awards

The current allocation formula, introduced in 1995, was intended to ensure that awards are allocated to institutions according to the extent and quality of their research training environment. Awards are distributed on the basis of HDR load, weighted at 40%, the research income and publications elements of the Composite Index, weighted at 40%, and HDR completions, weighted at 20%. The effect of this formula, however, has been to give older and larger institutions more HECS exemption APAs than they can use, and newer, smaller institutions a HECS exemption APA quota less than their load. As a result, only about 80% of HDR load is HECS exempt.

If the quota were determined 100% by HDR load, the shortfall between system HDR load and HECS exemptions would be shared equally among all institutions. Such an approach, however, would mean no consideration of the quality of the research training environment.

The number of HECS exemption APAs is subject to regular Cabinet review, the most recent of which saw the number increase by 1,600 in each of 1995, 96 and 97.

For some years, a proportion of APAs received a higher stipend when they were undertaking studies in areas their institutions determined were of particular priority. Although this is no longer the case, those receiving APA(Industry) awards still receive the priority level stipend.

ARC administrative arrangements

The ARC provides research and research training policy advice through the National Board, which considers the views of the Higher Education Council, as well as those of the other sectoral Councils under its umbrella. In contrast, the ARC provides funding advice on the programs referred to it directly to the Minister.

One pre-1988 arrangement which continued was that the Minister's department delivered the programs once the Council has given funding advice. That is, the Department disburses funds to institutions, it is to the Department that institutions acquit their expenditure, and to which researchers report the results of their work. The ARC and the Department have assumed joint responsibility for evaluation of these outcomes. The Department also provides research policy advice in addition to that provided by the Council.

There is a similar division of responsibilities between the National Health and Medical Research Council and its portfolio department. In other comparable countries research councils generally deliver their programs directly.

In July, 1996, Minister Vanstone released an Australian Research Council and Higher Education Council Discussion Paper which noted:

The Government's policy statement, Higher Education: Quality, Diversity and Choice announced a number of reforms to the way that the Government will receive advice on higher education issues. The major elements of these reforms are the abolition of the National Board of Employment, Education and Training (NBEET) and the establishment of the Higher Education Council and the Australian Research Council as independent bodies. Quality, Diversity and Choice also indicated that the independent, expert advice provided by the ARC and the HEC would continue to have a vital role in Government decisions about higher education, and that the HEC and the ARC would be given a wider range of functions.

Quality, diversity and choice stated that the Government will give the ARC responsibility for allocating competitive research grants, not just making recommendations to the Minister on their allocation. Giving the ARC responsibility for making final decisions on the allocation of research grants and related funds would involve a major modification to the division of responsibilities between the ARC and the Minister. In making this change, the Government affirms the importance of academic peer judgement in decisions about the allocation of research funding. It implies that the ARC will be established as a full statutory authority with program management responsibilities.[48]

THEME FIVE: FUNDING OF HIGHER EDUCATION RESEARCH

The core funding of Australian universities for research purposes is now almost entirely performance based. This is a major turnaround from the situation a decade ago. Then, core funding of research was implicit and effectively confined to the group of 19 institutions comprising the pre-1987 universities.

The change to a performance-based approach applied system wide came with the introduction of the Relative Funding Model in 1990, and the delineation of separate teaching and research components of the operating grant. The research related component of the model was defined as 'the quantum of the operating grant used to support research activities other than those inextricably to higher degree research training.'

General University Funds (GUF)

The Council would be pleased if the Australian Bureau of Statistics data were to collect and report type of research expenditure by source of funds. Even so, it is concerned at the methodology employed by the Bureau of Statistics. In principle, data are collected by institutions at project level. Nevertheless, a recent seminar organised by the Bureau to identify and foster best practice revealed that some major research-intensive institutions use much more rough-and-ready approaches (for example, prima facie 30% of time spent on research). Even where project level data is collected by institutions, it necessarily involves asking academics to make estimates of their time spent on research for a year in retrospect, usually without the detailed records which would enable them to do so reliably.

The Research Quantum (RQ) and Composite Index (CI)

There are three terms in the CI: research income, publications and HDR completions. Initially, there was a 90% weighting on the research income component which was, in turn, dominated by a National Competitive Grants element. The publications element has increased somewhat, as has the HDR completions, so that the current formula is:

Within the research income element, the weightings are:

It had been intended to further increase the weighting of the publications element, especially if a qualitative component could be added to the quantitative one. Little progress has been made in that direction, however, and, in the meanwhile an audit by KPMG of a sample of the 1994 publications collection found a very high rate of errors. In consequence, institutions have been asked to resubmit their 1995 data in the light of clearer guidelines, and these will be the subject of system-wide audit.

The RQ has been converted from a proportion of the operating grant to a dollar sum and since that occurred, subsequent growth in the number of places funded by the Commonwealth was not tied to growth in the RQ, leading to a decline in the RQ as a proportion of operating grant. By 1996, the RQ was $218.6 million, 4.9% of total operating grant.[53] In 1997, the RQ is $220 million, having been augmented by a component associated with increased growth. On the other hand, reductions in operating grant across the system announced in the 1996 Budget have been accompanied by proportional cuts in the RQ, which, as a result, will decline to $213.8 million by 1999.[54]

The Relative Funding Model was developed within the practical constraint of not making very large shifts in funding between the two elements of the formerly binary system. Nevertheless, formally the RQ represents the whole operating grant contribution toward research. The logic of its derivation requires that it includes the academic salary imputed to research. The ABS data above indicate that this cannot be the case. Anecdotal evidence supports this conclusion, with some institutions spending substantially more than their RQ on research costs excluding academic salaries.

In the Council's view, proper assessment of the adequacy of total funding for research, and of the balance between competitive and discretionary funding - including the size of the RQ - requires detailed and reliable data on actual institutional expenditure on research. The Council proposes to support a study whose brief will be to obtain such information.

It is important to note that the re-distribution of the RQ through the CI reflects only changes in the relative performance of institutions, not changes in the absolute research activity of the UNS. Thus institutions whose research activity actually increases can, and do suffer a diminution in their RQ. The only way to avoid this effect is to increase the Quantum itself to take account of increased productivity.

There are a number of problems associated with the current CI as a basis upon which to distribute the RQ. While the research income and HDR completions elements appear to be reliable and cost-effective to collect - although there continues to be debate about the weightings within the income element - problems with the reliability of the publications collection have already been mentioned. It remains to be seen whether they can be resolved. The cost to institutions of collecting this information is considerable, too, and in response the Minister has recently decided to reduce from twenty three to four the number of publication categories. Nevertheless, the publications collection may not be warranted unless the weighting on that element in the CI formula were increased substantially. A further difficulty is the essentially quantitative nature of the publications collection, with no quality factor other than peer review. Thus far, however, attempts to develop qualitative measures such as citation impact have proven unsuccessful. [55]

The major alternative model appears to be UK Research Assessment Exercise (RAE), which focuses on the quality rather than the quantity of publications. The outcomes of the third RAE were released recently.[56] Individual departments are assessed by peer review panels, which are provided with lists of staff members, each of whom may list up to five publications.

It has been argued that the RAE has contributed to an improvement in the international visibility of UK research (in contrast with that of Australia, which has declined in recent years). The ARC and the Higher Education Council have jointly commissioned a study to evaluate the practicality and cost effectiveness of adapting the RAE to the Australian context and, in particular, the usefulness of a peer review based research quality assessment process to supplement or replace the research publications collection. The report of this study will be made available to the Review Committee.

Research infrastructure

The initial commitment to these schemes was five years, from 1989. During this time, the belief that the level of infrastructure support for research was inadequate became wide-spread within the UNS. In consequence, the Government sought advice from the National Board of Employment, Education and Training as to both the level of infrastructure support and the mechanisms used to distribute it. The Board's advice was informed by independent consultants.[57]

Following that advice, the Government consolidated Mechanisms A and B from 1995 in the Research Infrastructure (Block Grants) Program (administered by DEETYA). Mechanism C became the Research Infrastructure (Equipment and Facilities) Program, still mainly supporting inter-institutional use of major equipment and facilities, but also now accepting applications from single institutions (referred to the ARC). A small increase in support for research infrastructure accompanied these changes, but far less than the $125 million per annum which the National Board had recommended.

Grant support from Commonwealth sources generally only covers the direct cost of research - equipment and consumables, the salary together with on-costs of full-time researchers and technical staff specifically associated with the project.[58] As noted above, the salary of teaching and research staff is not included; nor are major shared equipment and facilities (including libraries). Similarly excluded from direct grants are provision for building, fitting out and general maintenance of laboratories and other research facilities ('deep' infrastructure).

The project-specific infrastructure requirements of research are provided through the Research Infrastructure (Block Grants) Program. A proportion of the RQ is considered to provide for 'deep' research infrastructure.

An inter-departmental committee on higher education financing recommended prior to the 1995 Federal Budget that the appropriate level of project-specific infrastructure support is 40c in each dollar of direct grant support. Although that Budget provided significant increases in infrastructure funding for research (bringing the level to about 27c in the dollar), the increases were for a fixed period. The present Government will need to renew and increase this commitment if a level of 40c in-the-dollar infrastructure support is to be achieved.

Strategic development of research

For unto every one that hath shall be given, and he shall have abundance: but from him that hath not shall be taken away even that which he hath.

Matthew 25:29 KJV

The Council has asked the Government for funding to establish a new program which would assist in the strategic development of research and research training capacity in accordance with national needs. Applications from institutions would be required to be related to their own missions and strategic planning. Commonwealth funding would take into account the desirable overall shape of a diverse higher education research sector, and the state of development of the research area nationally. It would be provided for three years, covering the full costs in the first year, with institutions required to provide 25% of the funds in the second year, 50% in the third, and to guarantee salaries and on-costs for a further three years. Bids would be sought in the range $250,000 pa - to a maximum of $1million pa. It is proposed that funding be provided for three rounds of $5million each in initial year funding.

[Notes]

[2] ibid.

[3] Industry Commission, Research and Development, Report No. 44, 1995, v.1, 128.

[4] ibid., v.3, QA.2

[5] Australian Bureau of Statistics definition.

[6] Science Policy Research Unit, University of Sussex, The Relationship Between Publicly Funded Basic Research and Economic Performance, HM Treasury, 1996

[7] ibid., 373, emphasis added

[8] ibid., 375

[9] OECD, Main Science and Technology Indicators 1996

[10] DEETYA Selected Higher Education Student Statistics, 1996

[11] Note that each category is exclusive (ie., Science does not include Veterinary Science, &c.)

[12] Graduate Careers Council of Australia, Postgraduate Destination Survey 1995: A national survey of 1994 postgraduates' activities as at 30 April, 1995, 1996

[13] Graduate Careers Council of Australia, Graduate Destination Survey 1995: A national survey of the activities of 1994 university graduates,1996

[14] ibid., outlying values make the median a better measure than the mean or mode which are generally higher.

[15] Don Anderson, Sources of Australian Academics' Qualifications, AGPS, 1993

[16] For details, see Patterns of Research Activity in Australian Universities, National Board of Employment, Education and Training Commissioned Report No. 47, AGPS, 1996, pp.111-113

[17] Research income from all sources except operating grant and endowment. Excludes Australian National University whose Institute of Advanced Studies is block funded. Note, too, that research income data do not reflect endowment resources, nor do they reflect activity funded from operating grants.

[18] Patterns of Research Activity, loc. cit., xvii, xix

[19] ibid., 99-110

[20] DEETYA Selected Higher Education Student Statistics, 1996, 29. The 8,601 research only staff are generally ineligible to apply - indeed many are supported on grants.

[21] There are, of course, two other 'legs' in each CRC - industry and government research performing organisation partners - who receive the remainder of the CRC funding.

[22] This 'clawback' was, of course, re-distributed within, and not removed from the UNS. In addition, substantial new funding was provided through research infrastructure programs, discussed below.

[23] Industry Commission, Research and Development, Report No. 44, 1995, v.1, 383.

[24] OECD, loc. cit.

[25] loc. cit., 14

[26] Don Anderson, Sources of Australian Academics' Qualifications, AGPS, 1993, 3

[27] Don Anderson, Bob Arthur and Terry Stokes, The Qualifications of Australian Academics: Sources and Levels 1978-1996, in press.

[28] ibid.

[29] Paul Ramsden, 'Describing and explaining research productivity', Higher Education, vol. 28, 1994, 207-226

[30] Ruth Neuman, 'Perceptions of the teaching-research nexus: A framework for analysis, Higher Education, vol 23, 1992, 159-171

[31] Marc Romainville, 'Teaching and research at university: A difficult pairing', Higher Education Management, vol.8, no.2, 1996, 135-144

[32] Paul Ramsden and Ingrid Moses, 'Associations between research and teaching in Australian higher education', Higher Education, vol.23 no.3, 1992, 273-295

[33] loc. cit., 216-218

[34] loc. cit., 373

[35] Anderson, and Anderson et al., loc. cit.

[36] Research and Research Training in a Quality Higher Education System, National Board of Employment, Education and Training Occasional Paper, AGPS, 1992, 2

[37] ibid, 1

[38] Excludes Australian National University because HDR data do not distinguish the Faculties from the Institute of Advanced Studies, the latter being ineligible for most ARC and NHMRC funding.

[39] Unfortunately, the broad field categories under which institutions report their research income aggregate medical & health sciences. For HDR, however, desegregated data are available. Thus, while the post-1987 institutions had only 2.6% of the medicine HDR, and none in dentistry, in 1995 they carried 60% of the medical technology, 53.5% of the nursing, and 48.8% of the environmental health HDR load.

[40] It should be noted, however, that some of the other pre-1987 universities have faculties of medicine, while the ANU is in some respects a unique institution.

[41] DEETYA, Higher Education Funding Report for the 1997-99 Triennium, 1996, 29

[42] viii

[43] DEETYA, Higher Education Funding Report for the 1991-93 Triennium, 1991, 22

[44] DEETYA, Higher Education Funding Report for the 1992-94 Triennium, 1991, 15

[45] DEETYA, Higher Education Funding Report for the 1993-95 Triennium, 1992, 21

[46] ibid.

[47] DEETYA, Higher Education Funding Report for the 1994-96 Triennium, 1993, 31-32

[48] See http://www.deet.gov.au/divisions/hed/research/arcdis.htm

[49] Don Anderson, Richard Johnston and Bruce Milligan, Performance-based Funding of Universities, National Board of Employment, Education and Training Commissioned Report No. 51, 1996, 47, quoting DEET, Assessment of the Relative Funding Position of Australia's Higher Education Institutions, 1990, 13.

[50] Australian Bureau of Statistics, Research and Experimental Development Higher Education Organisations, 1994

[51] DEETYA Selected Higher Education Staff Statistics, 1996, 29

[52] DEETYA, Higher Education Funding Report for the 1997-99 Triennium, 1996, 63-64

[53] Anderson, Johnston and Milligan, loc. cit.,48

[54] DEETYA, Higher Education Funding Report for the 1997-99 Triennium, 1996, 64

[55] See Penelope Murphy, Determining Measures of the Quality and Impact of Journals, National Board of Employment, Education and Training Commissioned Report No. 49, 1996

[56] See: http://back.niss.ac.uk/education/hefc/rae96/c1_96.html and The Times Higher, Research Assessment supplement, December 20, 1996

[57] Higher Education Research Infrastructure, 1993

[58] The Co-operative Research Centres Program is a major exception. In that program, total research costs are calculated at three times salary cost.

*The Council has some concerns about the methodology used by Australian Bureau of Statistics to collect data from higher education institutions, especially concerning the imputed academic salary and associated imputed expenditures in General University Funds. These concerns are discussed in some detail below.

**Based on grants.

[Return to Top] [Return to Index]