Across three continents we have seen questions being raised about the compensation paid to the heads of academic institutions. This has mostly played out in the USA and the UK where the impact of the financial crisis and the downturn in the economy has impacted but the ability of students and governments to pay for education and scholarship but also put a dent in the ability of these institutions had been hit further by a downturn in the return on investment from their endowments.[1] In Australia this has, until recently, been less of an issue as the resources boom and the influx of overseas students kept the system afloat. However, even in the “lucky country” there have been financial concerns raised most recently because of the Labor government’s reduction of university funding via an “efficiency dividend”.[2]
This pressure for efficiency and austerity has prompted many people to question the level of pay received by university presidents and vice chancellors. But is this fair? From one perspective these individuals are heading billion dollar organizations whose complexity rivals many large corporations. From another perspective they are quasi public servants with few opportunities to really change the institutions they run. While each perspective guides us to a different market for talent, it is clear that there is an active market for these positions. It is this aspect of the debate we investigate.
What follows is a short summary of an analysis of the compensation of three groups of university leaders in 2010-2011: (a) the 100 most highly compensated US university and university system presidents, (b) the 100 most highly compensated UK vice chancellors, and (c) all 37 Australian public university vice chancellors.[3]
The analysis itself reveals a number of basic conclusions.
- First, on average, the most highly paid group is Australian vice chancellors. Just looking at the 10 most highly paid we see large disparities. In Australia the median is A$885,238. In the USA the median is US$714,191. In the UK the median is £363,420. Overall, the average compensation of a top 100 US public university president is A$480,409; that of a UK vice chancellor A$456,867; and that of an Australian vice chancellor A$721,607. See Table 1.
- Second, speaking generally, the larger the institution and the better its research quality (as measured by the Shanghai Jiaotong ranking) the higher the pay of the university head.
- Third, the better the research ranking the higher the overall pay of professorial staff (full professors in the US and chairs in the UK and Australia) and the higher the pay of the university head.
Table 1: Basic Information on University Leadership Compensation
However, these simple facts hide a number of subtle factors that we investigated statistically. A variety of different analyses were conducted and many different models examined. For the most part, nearly every model led to the same conclusions.
- UK vice chancellors and US public university presidents are paid comparably based upon the research quality of their faculty and the size of their student body.
- Australian vice chancellors are generally significantly overpaid based upon the research quality and size of their institution. This is particularly the case for what are known as the G08 institutions.[4]
Table 2 presents 4 estimated models. From these we see a number of key points.
- There are no gender effects. Male and Female university heads appear to be treated similarly once other factors are taken into consideration.
- In the UK and the US, the larger the institution the higher the compensation of the university head. In the US, this effect is attenuated for presidents of university systems when compared to universities. The student effect increases at a decreasing rate.
- Compensating for other factors there is no relationship between the pay of full professors and the pay of university vice chancellors and presidents.[5] It should be noted that this effect is identical if we condition the effect on the pay within a country.
- The Shanghai Jiatong rankings matter more in the UK and Australia than in the US. Overall, it matters more that a university is in the rankings than where in the rankings they stand. Most US universities are in the ranking.
- University groupings matter in Australia but not the UK.[6] It is quite clear that there is a difference between the Australian G08 and other institutions.
Table 2: Multiple Regression Estimates (in A$): Dependent Variable = Real Total Compensation
If we take all of this into consideration we can estimate what the salary of a university head would be if they were compensated according to the model estimates applying in another country. This is done in Tables 3 and 4.
Table 3: Comparison of Australian Vice Chancellor Pay Against Model and When Calibrated with US and UK Model Parameters.
Table 4: Comparison of UK Vice Chancellor Pay Against Model and When Calibrated with US and Australian Model Parameters
Let’s look at Australian vice chancellors first. If we use the US model parameters to estimate their compensation this would be like asking what if they were paid according to the model used to determined the most highly paid US public university president’s compensation. We come to the conclusion that they are overpaid anywhere from A$49,204 to A$428,207, with an average overpay of A$177,182 (we estimate the average US based pay at $435,000 to $520,000). It should be noted that this effect is sufficiently large not to be explainable by exchange rates or taxation. The reason the difference is even this small is because it includes many regional universities in Australia whereas the US comparison group includes only the most highly paid presidents. It is also the case that we do not account for other compensation (such as the value of housing or other non-pension benefits). If we use the UK model to estimate the expected Australian vice chancellor’s compensation the range is even lower at $391,000 to $477,000. Basically, it appears that the Non-G08 and non-ATN university vice chancellors are paid in alignment with the model, while those in the G08 and ATN are most likely to be out of alignment.
Looking at the UK vice chancellors we see that that if these individuals were paid based on the US model they would be paid approximately A$38,462 to A$114,219 less. However, if they were paid according to the Australian model they would be paid A$92,042 to $102,290 more. If they were paid according to the formula from the G08 institutions this number jumps to between A$238,674 to A$287,760! These estimates suggest that most UK vice chancellors seem to be paid in line with an overall market model and are dramatically underpaid in comparison to G08 vice chancellors.
Our findings suggest that there could be some inefficiency in the Australian market for university leadership that are not exhibited in the US or the UK. For example, it does not appear to be the case that the major factors that create the reputation of the university have much effect on the vice chancellors’ remuneration in Australia. Size of the institution, research and teaching quality, and alumni relations all seem to largely chart their own course. It should also be noted that even over a long period the likelihood that a university administration would dramatically increase either the size of their institutions or the institutions research standing is quite small. Indeed, in Australia, the RH Martin institute argued quite compellingly that the increased in ERA measured research quality between two Australia Research Council audits was due almost entirely to gaming the system rather than changing the quality of the research being produced.[7]
Hence it is quite unclear what it is that makes the compensation levels so different (particularly given that most university vice chancellors either come from other Australian universities or UK universities where the compensation levels for comparable positions is lower). There is also no reason to believe that Australian universities are any more difficult to run than major US universities or those in the UK or that the individuals chosen to run the Australian universities possess any special skills that demand greater compensation. For example, it should be kept in mind that we only used the top 100 most highly paid UK and US university leaders while we used all 37 public university vice chancellors in Australia. If we included even lower paid US and UK university heads the level of the overpayment to the Australian vice chancellors would be even larger.[8]
In addition, we find that structural factors do stand out. Again, we find better model fit when we simply account for G08 universities as members of the G08 than if we account for them based on their Shanghai Jiaotong Rankings (where the G08 do much better than the others). Hence, what seems to matter is not the research quality as much as simply being a G08 university.
It is hard to believe that there is something distinctive about the individuals at Australian universities that lead to the levels of compensation seen. This is further seen when we look some individual characteristics. While gender does not matter, it does seem to be the case that receive an ‘honour’ does. In table 5 we present the model we used to estimate the level of pay but include whether or not an Australian vice chancellor received an Australian honour (an AO, AC, AM or Centenary Medal) and whether a UK vice chancellor received an honour (a KBE or CBE or an OBE). It does not matter whether the vice chancellor in the UK received an honour but in the Australian case it does. If you were a G08 vice chancellor you effectively gain A$34,063 with an honour. If you were an ATN vice chancellor it is worth A$53,527. If you were in neither of these institutional groupings the bonus is the full A$62,932.[9] It should also be noted that Australian vice chancellors collect these honours at a much higher rate than their peers in the top 100 UK institutions (51% versus 26%).
Table 5: Multiple Regression Estimates (in A$): Dependent Variable = Real Total Compensation
Overall we see an interesting set of results that indicate a fairly consistent model of university leadership compensation. However, what we are presenting is not without limitations. For example, this is a one shot view of compensation that does not take into account movements for performance. We are also limited as to the information that we have as to other factors that may influence compensation both structurally (e.g., taxation) and individually (such as the prior performance of the individuals). In addition, we have excluded the compensation of private university presidents as the feeling was that they are less comparable to vice chancellors in the UK and Australia. However, when considering the information in its totality, and if one is willing to assume that the market for talent is a good judge of the available talent and that the US and UK represent relatively robust and competitive talent markets it is likely that this information is revealing something about the value of leadership talent compensation. In that case, it appears that the market in the UK and US is quite similar in terms of compensation on average while the market in Australia could be considered as an area of concern where there is a significant gap between the supposed requirements of the position and the level of compensation.
This article is co-authored with Grahame R. Dowling
An Appendix is available here.
[1] See, e.g., “Is this where your tuition fees are going? Vice-Chancellor’s salaries continue to rise, ” The Independent, 9 April 2013 (http://www.independent.co.uk/student/news/is-this-where-your-tuition-fees-are-going-vicechancellors-salaries-continue-to-rise-8565600.html); “University vice-chancellors take average £9,700 pay rise, “ The Guardian, 16 January 2012 (http://www.guardian.co.uk/education/2012/jan/16/vice-chancellors-9700-pay-rise); “Macquarie leads the way with vice-chancellor’s pay,” Sydney Morning Herald, 5 June 2012 (http://www.smh.com.au/national/education/macquarie-leads-the-way-with-vicechancellors-pay-20120604-1zs7t.html); “How much is your college president costing you?” US News and World Report, 15 February 2011 (http://www.usnews.com/education/best-colleges/articles/2011/02/15/how-much-is-your-college-president-costing-you-college-president-cost-per-student); “Median pay of public university presidents rose to $436,000 last year,” New York Times , 18 January 2010 (http://thechoice.blogs.nytimes.com/2010/01/18/college/)
[2] “Universities sector to be hit in Gonski reforms,” Sydney Morning Herald, 13 April 2013 (http://www.smh.com.au/national/university-sector-to-be-hit-in-gonski-reforms-20130413-2hry2.html).
[3] The data for this analysis comes from four sources: (a) in Australia, The Australian newspaper and the Universities Australia website; (b) in the UK, The Times Higher Education Supplement; and (c) in the USA, The Chronicle of Higher Education. The conversions were done using the midpoint exchange rate in 2010. The UK data includes salary and pension. The US data is for total compensation (except deferred compensation). The Australian data is salary.
[4] The G08 include: The University of Sydney, The University of Melbourne, Monash University, The University of New South Wales, the University of Queensland, The University of Western Australia, Australian National University, and the University of Adelaide.
[5] Note that this is not the pay of an average academic staff member but only Professors. The argument being that presidents and vice chancellors are typically professorial level staff. It should also be noted that there is considerably greater variation is professorial level pay in the US than in the UK or Australia and that the variance in the UK is also larger than the variance in Australia.
[6] The other groupings are: the ATN (The Australian Technology Network of Universities) – University of Technology Sydney, Queensland University of Technology, University of South Australia, RMIT, and Curtin University of Technology – and the Russell Group – Birmingham, Bristol, Cambridge, Cardiff, Durham, Edinburgh, Exeter, Glasgow, Imperial College London, King’s College London, Leeds, Liverpool, London School of Economics, Manchester, Newcastle, Nottingham, Oxford, Queen Mary (London), Queen’s (Belfast), Sheffield, Southampton, University College London, Warwick and York.
[7] See the reports by Frank Larkins at: http://www.lhmartininstitute.edu.au/insights-blog/2013/03/118-era-2012-part-1-discipline-research-profile-changes-from-2010-to-2012 and http://www.lhmartininstitute.edu.au/insights-blog/2013/03/118-era-2012-part-2-discipline-research-profile-changes-from-2010-to-2012.
[8] This can be seen by the fact that the A$ compensation of the 100 UK vice chancellors used in this analysis was A$456,867 while the average for ALL UK vice chancellors (including these 100) is $410,808. What this implies is that the remaining 62 vice chancellors made, on average, $A336,520 (meaning that the Australian premium is even more substantial). If we use the compensation of all 185 US public university presidents we get an average compensation of A$416,802 compared to the $480,409 earned by the top 100. This implies that the bottom 85 presidents earned on average A$303,805 (again dramatically increasing the premium paid to Australian vice chancellors).
[9] These estimates are taken by looking at two coefficients. The first is the coefficient in Table 5 for the receipt of an honour (A$62,932). This is then adjusted by the change in the coefficient relating to G08 and ATN institutions between model 3 in Table 3 and the estimates in Table 5. The former declines by A$28,870 (= A$195,758 – A$166,848) while the latter declines A$9,405 (= A$136,383 – A$126,978).
[10] Note that US university faculty salaries are based on a 9-month basis while those in the UK and Australia are based on 12 months.
It’s worse than you think. On average the US CEOs have considerably more responsibilities than their Australian counterparts.
In the US they need to do a lot of pressing of the flesh to appease a large group of stake holders (sometimes with rather pointed stakes) such as state legislators (for large state institutions), sports fanatics (can’t be a real university without a football team), and potential donors (who often “donate” with very large strings attached and may be of need of continual ego massaging). These commitments are all in addition to the mundane issues of academic freedom, research, scholarship, and teaching. Also remember that many US institutions serve as the home away from home for the majority of the students. Although the university is no longer subject to strict interpretation of “in loco parentis” the location in a typical college town (Armidale might be the only Australian equivalent) can mean that a US university CEO runs what is effectively the company in a company town. Thus any story about student indiscretion (such as a football player running amok) may require a direct response by the university CEO. And any shift in the numbers of staff or students can influence the livelihood of a whole community.
Coupled with this is the “market” in which they compete the US is much more competitive. In Australia, each institution has only the local competition for domestic students and in the case of the largest 8 they often only have at most one cross-town rival. However, US institutions compete on a national level for a large proportion of their students. Although the consideration of international students is now becoming important for the US sector they have a long history of competition in a very large national pool.