Chapter 4: Student Numbers
The final section of my report will consider a number of issues about overall student numbers which impact directly or indirectly on fair access. The most important is the continuing debate about the extent, if any, to which the admission of students from socially deprived communities 'displaces' other students because the total number of places in universities for Scottish domiciled students is capped (or, to be more precise, the budget for funded places is limited). The implications for this student number cap of the UK's decision to leave the European Union and the indirect impact of decisions on students fees and funding in England will also be considered.
The 'cap' and displacement
The debate about 'displacement' has great political salience. Along with the criticism of SIMD as an appropriate measure of progress towards fair access, the suggestion that well qualified Scottish students are being 'displaced' by the admission of more socially deprived students are the two most powerful arguments used against current fair access policies. The issue has been frequently raised in the Scottish Parliament by opposition MSPs. It is also reflected in letters sent by aggrieved parents to Ministers and to the Principals of the more selective universities. Concern about 'displacement' is likely to grow. As with the argument about SIMD it deserves the most serious consideration, and must be confronted if fears about 'displacement' are not to erode the impressive political and public consensus that currently exists about the need for fair access to higher education.
Consideration of 'displacement' takes two forms: an examination of the available data to determine whether and, if so, to what extent it is happening; and a discussion about the equity of the current pattern of admissions to universities.
Is 'displacement' taking place?
The 'headlines' are:
- First, that the percentage of Scottish (and non-UK EU) acceptances - typically, those covered by the cap on funded places - that are Scottish has increased from 87.9 per cent in 2016 to 89.6 per cent in 2018. This is likely due to the decline in EU applicants to Scottish HEIs over the same period;
- Secondly, while the percentage of Scottish acceptances which are from SIMD20 areas has tended to increase, the percentage from the next two SIMD quintiles (who, it is has often been suggested, are the main losers as universities recruit more SIMD20 entrants) remained stable or declined slightly. However, it is important to emphasise that absolute numbers have increased in all SIMD quintiles since 2016 (Chart 6). Between 2015/16 and 2017/18, the number of SIMD20 full-time first degree entrants rose from 4,015 to 4,650, while the number of entrants from other SIMD quintiles increased from 24,605 to 25,070 (admittedly at a slower rate). One possible factor might be that participation levels among the more affluent are close to saturation;
- Thirdly, entry rates have continued to rise across all SIMD quintiles since 2016 (Chart 4). Although the main reason for this improvement is a decline in the estimated number of 18 year-olds living in Scotland, it suggests that, in aggregate, there has been no 'displacement'.
Chart 6: Scottish and EU domiciled acceptances to Scottish HEIs by deprivation quintile (SIMDQ1=SIMD20), 2016 to 2018
What the data tells us at subject level
However, these 'headlines' must be heavily qualified in a number of respects. First, only one year of data (2017/18) is available in which the direct impact of CoWA inspired targets can be observed. Percentage changes in earlier years reflect the natural evolution of universities' admissions policies on widening participation. Secondly, some subject areas have a small number of entrants that fluctuate year by year and reflect other changes, such as the introduction of new courses. Thirdly, SIMD20 entrants are concentrated in particular subjects. So their percentage share varies considerably, and their overall share is also influenced by any changes in the number of places in different subjects.
With these qualifications in mind, the data shows that between 2016/17 and 2017/18 there were increases in the percentage of full-time first degree entrants from SIMD20 areas in the great majority of subjects - 15 out of 18. The only subjects in which their percentage declined were engineering and technology, agriculture and education. In comparison the percentage of SIMD20 entrants increased in a bare majority of subjects, 10 out of 18, between 2015/16 and 2016/17.
The most striking - apparent - turnaround was in mass communication where their share declined by 3.5 percentage points between 2015/16 and 2016/17 but increased by more than 5.1 percentage points between 2016/17 and 2017/18. However, this underlines the need for caution before making categorical statements about year-on-year changes in percentage shares. Mass communication is a small subject which typically recruits only small number of SIMD20 students. As a result there have been substantial year-on-year fluctuations in percentage shares, which have also affected other SIMD quintiles.
In law, a similar pattern can be seen - an increase of 4.4 percentage points between 2016/17 and 2017/18, following a 2.9 percentage point decline the previous year. Before that decline, which was mirrored by an even greater increase in the percentage of entrants from the top SIMD quintile while the percentages of entrants from the three middle SIMD quintiles remained the same, there had been a slow but steady increase in the percentage of SIMD20 entrants dating back to 2012/13, before the CoWA had been established.
Business administration, a relatively large subject, has followed another pattern. The percentage of SIMD20 entrants increased by 3.3 percentage points between 2016/17 and 2017/18, and exceeded the percentage shares of the next two SIMD quintiles (although, to demonstrate the complexity of these shifts, SIMD20 entrants already had a bigger share than entrants in the next SIMD quintile back in 2012/13).
In allied medicine, another large subject with more than 4000 Scottish domiciled entrants, SIMD20 entrants increased their percentage share by 3.1 percentage points between 2016/17 and 2017/18. But all five SIMD quintiles have very similar percentage shares, reflecting perhaps the comparative lack of interest in this subject among high-performing school leavers from more privileged social backgrounds.
In biological sciences, SIMD20 entrants appear to have gained ground between 2016/17 and 2017/18 by increasing their share of places by 2.5 percentage points, apparently at the expense of SIMD quintile 3 (which, to demonstrate the complexity of these fluctuations, had markedly increased its percentage share the previous year).
Finally, in medicine the percentage of SIMD20 entrants increased by 0.2 percentage points between these two years. But this gain was actually less than the year before the CoWA inspired targets came into operation.
Two tentative conclusions can be drawn from this data. The first is that it is difficult to conclude that any 'displacement' can be attributed directly to the impact of COWA inspired targets. It seems more likely that the gradual narrowing of, still very unequal, percentage shares of full-time first degree entrants between the five SIMD quintiles is the natural effect of universities' widening participation and admission policies. In other words, any 'displacement' that has taken place is a reflection of a growing consciousness across Scotland about the importance of fairer, and necessarily more equal, access to higher education and, in particular to universities.
The second is that it is also difficult to conclude that fair access policies (including both institutional practices and national targets) can fully account for the increasing percentage share of SIMD20 entrants that can be observed in most subjects. Fluctuations in the social mix of new entrants occur naturally, and can be magnified in small subjects. Some subjects naturally attract more SIMD applicants than others. As a result changes in the comparative numbers of students in different subjects can influence the overall percentage shares of different SIMD quintiles. Demographic factors are also important, and have not been fully evaluated. The only safe conclusion, based on the available data, is that a small amount of 'displacement' may be taking place in certain subjects, but not on the scale suggested by public debate on this issue.
Of course, this does not address the issue of the overall student number cap. But, even if the cap were to be raised, a narrowing of percentage shares would likely still tend to occur. In other words, applicants from the most advantaged social groups, who currently benefit from the unfair distribution of university places, would inevitably experience some degree of comparative reduction in the proportion of places they comprise. But it is difficult to characterise this as any form of discrimination.
The overall allocation of public expenditure is a matter for decision by the Scottish Government which necessarily has to balance the undoubtedly strong claims of higher education against those of schools, the National Health Service and other public services. However, if the cap was increased, its impact on 'displacement' would be likely to be twofold. The first effect, much emphasised in public discussion, is that the increased availability of places would help to reduce concerns about 'displacement' because fewer applicants would suffer an absolute reduction in their chances of securing university places. The second effect, barely mentioned, is that the narrowing of the respective percentage shares of entrants across the top four SIMD quintiles would probably accelerate because it would no longer be constrained and inhibited by a keen awareness of the restricted availability of university places for Scottish domiciled students. So the relative reduction in opportunities for more socially advantaged applicants could well increase.
Equity in university admissions
Despite the tentative nature of the conclusions that can be drawn from the data, there can be little doubt that there is a strong, and perhaps growing, perception that Scottish domiciled applicants from more socially privileged backgrounds are being 'squeezed out' by Scottish domiciled applicants from socially deprived communities, specifically those living in SIMD20 areas, as well as by applicants from the rest of the UK and from outside the European Union.
Universities argue, with some justification, that reducing the number of places available currently filled by entrants from the rest of the UK and outside the EU would not make more places available to Scottish domiciled students because of the cap; and, more contentiously, that reducing these uncapped places would deprive the universities of additional funding that can be used to benefit all their students. However, while this argument has generally been accepted in the case of international, i.e. non-EU, students whom universities have been free to admit since at least the 1960s, the apparent discrimination between Scottish and other UK applicants has probably intensified the perception of unfair 'displacement'. It may appear that it is easier for an English applicant from a socially advantaged group to be admitted to a Scottish university than an equally privileged Scottish applicant because of the cap - despite the awkward fact that the former would be charged £9,250 for tuition while the latter would enjoy free tuition (or that Scottish domiciled school leavers from socially privileged backgrounds are at liberty to apply to English universities). Nevertheless, the power of perceptions with regard to 'displacement' has to be acknowledged.
A particular argument that has gained currency is that, if there are more SIMD20 entrants and overall student numbers remained capped, the applicants most likely to be squeezed are those from the next most deprived SIMD quintile. This is a familiar argument about the so-called 'squeezed middle' caught between those with excellent grades from prosperous and well educated families who will always manage to find places and applicants from socially deprived communities who are favoured by fair access policies. It is an argument that is given additional rhetorical force because it is often implied that the applicants who are suffering most from 'displacement' are from working-class backgrounds who have worked hard at school and got decent grades.
There are two responses to this argument:
- First, there is little evidence from the data that it is a 'squeezed middle', or applicants from the next most deprived SIMD quintile, who have been most 'displaced'. To the extent that there is evidence of 'displacement', the data reveals no particular pattern of relative loss of advantage. There is at least as much evidence that it is applicants from the top SIMD quintile who are being 'displaced' as a result of fair access policies as applicants from the three middle quintiles;
- Second, even if applicants from the second most deprived SIMD quintile were being squeezed out by the drive to recruit more SIMD20 students, the responsibility would lie squarely with the universities themselves. The contextual admissions policies being developed by all universities are sophisticated enough to allow universities to make appropriate allowance across the whole range of applicants, especially as in many universities a very substantial minority (or even majority) of applicants have at least one contextual admissions 'flag'. Fair access does not end with SIMD20.
In the end there is no alternative to confronting directly the issue of 'displacement'. If it is important that fair access to higher education is achieved (and there is no - public - dissent to this objective), then it logically follows that it is possible that applicants from currently (and unfairly) over represented social groups will suffer some modest reduction, certainly in comparative and perhaps in absolute terms, in their chances of securing university places - especially to the most competitive subjects and the universities with the greatest prestige. There is no way round that logical possibility. The only way to remove it completely would be to abandon the drive to fair access. Exactly the same considerations apply to the current debate about the underrepresentation of women across wide tracts of national life, or the struggle to rectify previous discrimination against black students in US universities. In practice, the general incidence and severity of any disadvantage suffered by applicants from socially advantaged groups are slight - although that is not to deny there may be individual 'hard cases'. But in the end, the principle of securing fair access for all is more important than the need to avoid at all costs 'displacement'.
The implications of Brexit
When, or if, the UK leaves the European Union, the Scottish Government would no longer be obliged as a matter of EU law to provide free tuition to non-UK EU students alongside Scottish domiciled students. In the event that the Scottish Government chose not to continue to provide free tuition to non-UK EU students, they would also cease to be included within the cap on student numbers. It has been argued that, in that eventuality, the cap should not be reduced pro rata but maintained at its current level, creating more head room for the recruitment of extra Scottish domiciled students. These additional funded places could then be used to mitigate the effects of any 'displacement'.
After Brexit other EU students could be treated as international students and be charged the same fees as other international students. If the number of other EU students entering Scottish higher education remained at or near its current level, institutions would stand to gain significant additional income. If at the same time the student number cap was maintained at its current level, they could enjoy a win-win situation - no reduction in public funding and additional fee income from EU students.
However, this enticing prospect has to be qualified in a number of ways:
- First, already since the referendum the number of students from the EU (excluding those from the rest of the UK) on first degree courses in Scottish universities has declined, although not dramatically. While they made up 9.5 per cent of all full-time first degree students in 2017/18, their percentage of first year students was 8.8 per cent. Between 2016/17 and 2017/18 full-time first degree entrants from the EU (apart from those from the rest of the UK) declined from 4,370 to 3,865, although it was still higher than the year before, which suggests there may have been a temporary post-referendum blip. In the longer-term a collapse of EU student demand seems unlikely given the quality of Scottish universities and the attractions of studying in an Anglophone country;
- Second, the Scottish Government has already announced that 2020/21 entrants from the rest of the EU will still be eligible for free tuition. In England the UK Government has now given a similar guarantee. Even if EU students become liable to pay international student fees from 2021/22, the full effect of this change would not be felt until 2024/25. This lengthy transition also poses difficulties with regard to the student number cap. The budget for funded places could well be adjusted for other reasons over this period in successive public spending reviews, which would make it difficult to assess whether any promise to maintain the cap post-Brexit had been kept;
- Third, colleges and universities will have to make a strong case for maintaining the current student number cap, and current level of public expenditure, after Brexit. Inevitably there will be arguments within Scottish Government that to ignore the removal of other EU students from the cap and the removal of the obligation to offer them free tuition, in effect, would offer a 'windfall' to higher education, which would at a minimum need to be assessed against competing claims from other public services. If the number of other EU students held up, an even stronger case will need to be made to persuade the Scottish Government to ignore the additional income produced by the fees they would then be liable to pay. A case based on advancing fair access is likely to be more persuasive than one based on mitigating 'displacement' of applicants from more socially advantaged backgrounds;
- Finally, there will be strong political pressures to make special arrangements for other EU students, to demonstrate Scotland's solidarity with the rest of Europe. Universities themselves may also wish to make such arrangements to highlight, and safeguard, their European links. Given the uncertainties that surround the terms of any UK withdrawal from the EU, and of any future partnership, it is difficult to assess what form any special arrangements might take. There could also be legal complications. European law allows Scottish universities to charge entrants from the rest of the UK fees (because national Governments can discriminate between their own citizens) but obliges them to treat other EU students in the same way as Scottish domiciled students (because discrimination against citizens of other EU states is not allowed). After Brexit European law would no longer apply and a new legal basis would need to be found for continuing discrimination against students from the rest of the UK compared with students from the rest of Europe (not only EU member states?). Currently there are 21,430 full-time first degree students from the rest of the UK studying in Scottish universities, whose fees represent a significant source of income.
These qualifications underline the complexity of the position of other EU students after Brexit. The safest conclusion is that any Brexit 'bounty' is likely to be limited - and late. Even if the current student cap is maintained, and all EU students become liable to pay international student fees without any special arrangements being made, it would not be sufficient to change substantially the terms of the debate about 'displacement'.
Augar review of student funding in England
The Independent Panel for the Post-18 Education and Funding Review in England led by Philip Augar made far-reaching proposals for reforming student fees and funding in its report published last month. The proposal most likely to have an impact on Scotland is the panel's recommendation that the maximum fee should be reduced from £9,250 to £7,500 a year. If implemented, this change would have two implications for Scottish universities. First, their income from charging English students would be reduced, which could have a significant impact on some universities. In 2017/18, at the University of St Andrews there were 520 English first degree entrants, compared with 570 that were Scottish, while the University of Edinburgh enrolled more English first degree entrants (2,045) than Scottish (2,025). Second, English universities would become comparatively cheaper for Scottish students. Although far fewer Scottish students go south of the Border for their higher education than English students attending Scottish universities, it is possible that applicants from more socially advantaged backgrounds, who fear 'displacement' by SIMD20 students, might find it a more attractive proposition.
However, the main impact of the Augar review could be to vindicate Scotland's policy of free tuition. Moving towards the English model of high fees, backed by student loans and accompanied by targeted initiatives on widening participation, has been the private passion of some people in Scottish universities. But it has been clear for some time that this model, even setting to one side principled objections to its more aggressively 'market' tone, is unsustainable both for students (and graduates) and for tax-payers. Ever since the Office for National Statistics (ONS) decided at the start of this year that a much higher proportion of student loans should be counted as current public expenditure, radical overhaul of the English model has been inevitable. An overall student number cap is likely to be reintroduced in England to establish some degree of control over public expenditure on higher education, either directly or indirectly by restricting which types of student can be admitted to universities. If this happens, the argument that exceptionally in the UK Scottish domiciled students from more socially advantaged backgrounds are at risk of 'displacement' as a result of fair access and widening participation policies, will lose much of its salience.
Conclusions and recommendations
According to the available data there is limited evidence so far that 'displacement' of applicants from more socially advantaged backgrounds by SIMD20 applicants has been on a significant scale. But it would be a mistake to play down claims of 'displacement' - for two reasons. First, although there is limited evidence that the CoWA inspired targets have produced 'displacement' (perhaps because only one year of data is available), the impact of universities' own access and participation policies seems to have led to a gradual, but discernible, narrowing of the participation gap between entrants from different SIMD quintiles. Secondly, whatever the facts, there remains a powerful perception of 'displacement', which is reinforced by anecdotal evidence and individual cases. So this is a key issue that must be confronted.
Recommendations in relation to student numbers
The Scottish Government and the SFC should monitor closely the statistical evidence about 'displacement', so that national debate is based on the most reliable and up-to-date data rather than perceptions and anecdotes.
Universities should monitor the impact of fair access targets on opportunities for applicants from other social groups at subject and course level, with the same intention in mind.
Robust and honest arguments should be developed to explain and justify the need to secure a fairer distribution of opportunities across all social groups, concentrating on opportunities for the most disadvantaged (along the same line as the arguments on gender equality).
Any resources released by Brexit should be retained within the higher education system, but should primarily be used to promote fair access.