Skip to content

Student debt and grants in Scotland: a summary

This post summarises the current position on grants and loans for full-time students in higher education in Scotland, and the background to it.

Background

(i)  Fees and other payments

The Scottish Government funds the whole tuition cost for almost all first-time, full-time Scottish and EU students in Scotland, from the government’s cash budget. Therefore no-one from any background has to borrow for part or all of their fees. Scottish students will only need a fee loan (as students in other parts of the UK get, to defer fee payments) if they go to study elsewhere in the UK.

Between 2001 and 2006, young students entering degree-level  HE full-time were liable to pay the graduate endowment, a single payment of £2,000 at 2001-01 prices, after finishing university. The income from the GE was in theory ring-fenced for student bursaries. Graduates could either pay it in cash or add the liability to their existing student loan (or take out a first-ever loan) to defer the payment. Because of exemptions for HNC/D students, including those on “2+2” models, mature students, disabled students and single parents, slightly under half of all the full-time students the SG supported were liable to pay the GE, which was bringing in around £23m p.a. by 2007 (more here).  When the current Scottish Government says it brought in free tuition, it is referring to its abolition of the endowment in 2007. It is also often, in practice, describing its decision not to use devolved powers to copy either of the fee regimes which have applied in England since 2007.

(ii) Living cost grants

Scotland has relatively low student maintenance grants (here called bursaries).  Most living cost support is offered instead as student loan.

Between 2010 and 2012 inclusive, the means-tested grant for younger students, Young Student Bursary, was frozen in cash terms (see here for more on how its value changed from 2001 onwards). In 2013, the Scottish Government cut its total spending on maintenance grants by around £35m, or one-third.  The maximum YSB was reduced from £2,640 to £1,750, and it was withdrawn more quickly as income rose.  The government lowered the income at which maximum YSB was payable from £19,300 to £16,999. Many students lost £900 a year and some much more. The Scottish Government argued they could make up the difference by borrowing to fill the gap. Older students get the lower-rate Independent Student Bursary. This was introduced as a lower-rate grant by the Scottish Government in 2010, and then also scaled back in 2013.

In 2015, the Scottish Government added £125 back on to some grants, costing it around £5m,  and in 2016 it reversed most of the  cut to the threshold for maximum grant, raising it to £18,999 (likely to have cost it a bit under £2m a year).

The current system

The resulting living cost model in 2016-17 is in the table below.

Young Independent (ie mature)
Bursary Loan Bursary Loan
0-18,999 1,875 5,750 875 6,750
19,000-23,999 1,125 5,750 6,750
24,000-33,9999 500 5,750 6,250
34,000 plus 4,750 4,750

A particular feature of this model is that it is built round those from the lowest incomes, especially mature students, taking out the highest loans.  Until grants were abolished in England in 2016, Scotland was the only part of the UK taking this approach. It means that someone at a low income who wishes to take out their full entitlement to living cost support over four years faces a debt of £23,000 plus interest if they are younger, and £27,000 plus interest if they are older. Grants are higher in Wales and Northern Ireland (where students also only have to borrow for the first £3,900 of their fees: true for Welsh students anywhere in the UK, for NI ones in NI).

Actual borrowing

Figures on annual borrowing by income are published annually by the Student Awards Agency Scotland (SAAS). The latest are here (see Table A6). They show that Scottish students borrowed a total of £0.5bn in 2015-16.

Around 70% of Scottish students take out a loan in any given year, and almost all those who borrow, borrow the whole amount they can.

The table below is adapted from the official statistics.  I’ve added two columns. One shows average borrowing across the group as a whole, i.e. including borrowers and non-borrowers.  The other shows the percentage who don’t borrow in each income group.   It’s reasonable to assume from other research that there are more non-borrowers in the higher income group because students’ access to family resources tends to rise as family income increases. It is likely that even within this group, non-borrowers are more prevalent at higher incomes:  it is quite plausible that at, say £60,000+, non-borrowers are in the majority.

The net effect of lower income students having higher loan amounts and making more use of loans is that students in the highest income range borrowed in practice around half as much per head (around £3,000) as those in the lowest income band (over £6,000). Another way to look at this is that Groups 1 to 4 below accounted for only 43% of all students, but took out 54% of all debt.

Borrowing by income band 2015-16

  Total students Borrowers Average borrowing (active borrowers) Average borrowing  (all) % Non-borrowers
1 No income details: receiving max bursary 10,055 9,360 6,660 6,201 7%
2 Up to £16,999 23,895 19,105 5,890 4,711 20%
3 £17,000 to £23,999 8,955 7,220 5,760 4,648 19%
4 £24,000 to £33,999 8,980 7,265 5,610 4,542 19%
5 £34,000 and above 2,965 1,850 4,650 2,901 38%
6 No income details: receiving no bursary 70,010 46,830 4,650 3,112 33%
Note:  I’ve removed EU students (14,705) from the figure for total students, as these students can’t borrow. I have assumed that they were contained in Group 6, as very few can claim means-tested support. That may not be exactly right, but it should be near enough. Most students with an income over £34,000 will be in Group 6, which covers those who chose not to submit income details, generally because they are above the threshold for bursary. Group 1 by contrast will be those who had no relevant income to declare and got the highest bursary level.  Group 5 is a small group whose income details SAAS knows, although they are over the bursary threshold. I have excluded here a very small group of low-income students separately shown in the SAAS table who anomalously have no income but don’t get full bursary: there’s something odd going on with this group (it may be that many don’t complete a full year).

Caution: final borrowing

Separate figures are published each year for students’ final borrowing. The most recent Scottish figure is £10,500. These figures are widely quoted but have to be handled with care. The average will be brought down by the large number of students in Scotland on one or two year courses, and – as shown above – any average will conceal variation by income. More on that here.

Conclusion

The Scottish system is not debt-free in the absence of fees: indeed Scottish students are borrowing a substantial amount as a group each year. The Scottish approach relies heavily on loans to cover the state’s role in providing low-income students, in particular, with living cost support. Grants are now so low that those from the lowest incomes are taking on the most of that living cost debt.  Equally, at high incomes, many students will be borrowing nothing.

Defending existing policy in Scotland means defending this outcome.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Modelling a different approach to fees and grants in Scotland

It seems common to assume that we’re faced with a straight choice on tuition fees, where the state either funds the whole of everyone’s tuition costs, or all students have to take out a loan for £9,250 a year.  It’s a perspective stuck in a binary choice between whatever-we-do-now-in-Scotland and whatever-they-do-now-in-England.  Other options are of course available. That’s what this post is about.

A starting point

The simplest alternatives for fees are:

  • setting the fee level at some number more than zero, but less than £9,250;
  • means-testing fees;  or
  • some combination of these two.

There’s a fairly common argument that any move away from free tuition inevitably means Scotland would end up where England is: the slippery slope perspective.  That puzzles me, because we would only end up there if that was what the politicians we elected chose.  It seems to assume we can trust them to keep it free, but not to maintain any alternative position. On this thinking, tuition fees are an addictive drug, from which governments must be kept away at all costs.  I’ll come back to that at the end.

If you reject this straitjacket, and believe that the Scottish political system is capable of managing other things, what might the alternatives be, and what would they mean for students?

Note:  I’m not advocating a specific model here, but demonstrating one different way things could work, as a starting point for a less constrained debate.  There are more radical ideas out there (here’s one), involving more fundamental change. All I want to show is the space  for alternative outcomes, just within the current broad general approach.

Living cost support matters

There’s no point designing a student funding system which only looks at one part of the story.

If you are only interested in thinking about fees in isolation, and don’t care much about living cost support, especially for low-income students, then look away now.  You and I are never going to have a mutually rewarding  conversation about this.

A lot of people in Scotland at this point suggest that living costs are a secondary (or even non-) issue, because students can always live with their parents and/or work their way through.  I disagree for the reasons set out in Footnote 1.

The modelling below assumes decisions on living costs are as important as fees, should be interlinked, and looks at combined effects.

Fees: upfront or deferred?

Even the most vigorously ideological advocates of tuition fees recognise that students tend not to have money at the moment.  It is very rare to find anyone, not even Milton Friedman, advocating unassisted upfront charges  (Footnote 2 describes the UK government’s short experiment with this).

Thus, in no part of the UK do first-time students now have to find the cost of their fees from their or their families’ existing resources.  With a few exceptions in Scotland and elsewhere, higher education is free at the point of entry for all  first-time undergraduate students in the UK, because at minimum they can take out a government-subsidised student loan to defer the full cost of their fees until they are earning above a certain level.

It’s hard to over-stress this point for readers in Scotland, where it still seems to be widely believed that students in England have had no choice but to find £9,000 a year from their families. Had that been the case, the system there would simply have collapsed. It is precisely because  fee costs are deferred that debt is so high in England.

So the model assumes that any fee is matched £ for £ by a government subsidised loan and that, as in other UK systems,  the fee loan (a)  would be repayable contingent on earnings, in the same way as maintenance borrowing is now, and (b)  is added to any maintenance loan to form a single debt, so that no-one is paying off two loans in parallel.

Debt aversion

Regardless of any evidence from England (Footnote 3),  it is possible that debt aversion should be a major concern for Scottish policy makers. In that case, however, we should already be worrying.

Living cost support for low-income Scottish students is now provided largely through loans, because grants are relatively low.  Here’s the figures for support from the Student Awards Agency Scotland (SAAS) for 2016-17.

Young student Independent student
Household income Bursary Loan Bursary Loan
£0 to £18,999 £1,875 £5,750 £875 £6,750
£19,000 to £23,999 £1,125 £5,750 £0 £6,750
£24,000 to £33,999 £500 £5,750 £0 £6,250
£34,000 and above £0 £4,750 £0 £4,750

The model below illustrates how in a system which includes some fee-paying,  low-income students can still have less debt than in one with free tuition, while protecting the value of their total living cost support.

The scale of student debt in Scotland and its distribution

Around £500 million is now borrowed each year by Scottish students.  At the moment, annual borrowing is skewed towards those from lower-income households, for two reasons:

  • they borrow more on average, and
  • they are more likely to make use of student loans.

As a result,  over half of all student loan is taken out each year by students declaring a household income below £34,000, although fewer than half of students fall into that group.

The current statistics don’t allow us to differentiate amongst those with incomes over £34,000.  But looking at earlier data, there’s a good chance that there’s a similar skewing of debt within the higher income group, towards middle-income households and away from the highest income ones.

The model removes the current built-in assumption that the highest debts should be taken on by those from the lowest incomes, again while protecting current total spending.

What could be different?

Put simply, we could move the debt around, so that more of this £500 million is taken out by students from higher-income backgrounds, and less by those from lower-income ones.

That means finding more to spend on grant, by spending less on fee subsidies, and expecting those at higher incomes to borrow some of their fee cost.

Mechanisms

There are in essence three ways to get this effect.

  1. Means-test the fee.
  2. Apply the fee to everyone, but then have a separate means-tested fee grant which immediately wipes it out for lower-income students.
  3. Apply the fee to everyone, but then build a means-tested off-setting amount into the  living cost grant, before making any other increases.

 

Different mechanisms would have different implications for practical administration, public understanding/presentation, student behaviour and the detail of public finances. But they would all provide an identical boost to the amount of grant provided at lower incomes for the same level of fee.

One basic model

The model below asks students from the highest income households to borrow one half of the average cost of a university place in Scotland. So students from these households would be offered a government-subsidised fee loan to cover a fee of £3,500 a year. The cash released from tuition fee subsidies would be put back into grants.

How much would a £3,500 fee raise?

SAAS currently supports around 140,000 students, of whom around 15,000 are from the EU.  I’ll concentrate for now on moving the public subsidy around between the 125,000 Scottish students.  A separate section below considers EU students.

If Scottish students from the highest quarter of student-providing households by income were liable for the fee, it would notionally release around £110m a year from tuition fee subsidies (125,000 x 25% x £3,500).

I can’t say what the income cut-off would be, because the Scottish data on students is now too aggregate to show that.  Looking at figures for years before 2013, when more detail was provided,  I’d guess it would be somewhere between £50,000 and £60,000 of household income (around the highest 15-20% of all households in Scotland by income, after equivalising for a family of 2 adults and 2 children).

How could the money saved on fee subsidies at high incomes be used to bring down debt at lower incomes?

I’ll spend the money on substantially increasing means-tested maintenance grants, on which  we now spend only £55m a year.

It would cost c£30m to switch £1,000 of living cost support from loan to grant for those on the Young Student Bursary.

I’ll spend a further c£40m on giving independent students (for example, those over 25, or who are parents, or married/in a partnership) the same  bursary as young ones, and bringing down their debt, because these students in Scotland are on a much less generous grant and a higher loan, and there’s really no good way to justify that.

I’ll also spend c£15m on  a new £1,000 grant for students from households between £34,000 and £45,000, because these families, who are not awash with cash, are expected to find much more out of pocket help for their children in Scotland than is the case in the rest of the UK and that’s a concern: more here.

The net annual effect on individual students at different incomes would be:

  • Young students with incomes below £34,000 would gain £1000 in grant and lose £1000 in debt, with no change in the total value of their living cost support.
  • Those with incomes between £34,000 and £45,000 would gain £1,000 in grant and therefore £1,000 in total living cost support, with no change in debt.
  • Nothing would change for those between £45,000 and the fee liability point.
  • Those liable for fees would have £3,500 more debt a year and no change to living cost support.
  • For mature students it’s a similar picture, except that they would gain more grant and lose more debt.

Using this approach, all students are now offered the same living cost loan (£4,750), with cash grant used to do any additional income-based targeting

I’ve spent approaching £90m. I assume that due to things I’ve failed to take into account, income wouldn’t be as high and expenditure would be  higher, so my spending plans may still be a bit ambitious on this level of fee.  But they will be in the right general area.  If students from the wealthiest quarter of households were expected to borrow £3,500 a year of their tuition cost, it seems likely that we could nearly treble our spending on maintenance grants.

The effect on the new fee payers

Total debt for those at high incomes would come to a maximum of £8,250 a year.  Two points about that. First, in practice many of these students would only have a £3,500 annual debt, because  living cost debt take-up is lower in this group, presumably because many have all their living costs met by their parents.

But the second is the more important. Low income mature students are already expected to incur £6,750 in living cost debt – and most do.  If you have managed in recent years not to be outraged at the reality of a £6,750 annual debt for most mature students with no income, you are not in strong position to be outraged now at a theoretical maximum debt of £8,250 a year for students from high income families which many won’t actually incur.

Is this a good model?

This would be a pretty clunky way to do things.  The Scottish system already incorporates large step changes in entitlement, and it’s not an ideal approach. However, because the data comes packaged that way, it’s hard to model anything without copying that.

The point of this model is not to advocate it in this precise form, but to bring out what scale of change would be possible for a particular form of fee liability.

A more radical, and carefully argued and well-evidenced, rearrangement of fee and grant subsidies has been proposed for Wales by the Diamond Committee. The Welsh Government has accepted the recommendations and recently finished consulting on the detail of implementing it. The change has cross-party support, and support from the NUS Wales and Universities Wales. Anyone interested in this debate should read that report (here),  as a further example of the range of possibilities.

What about other objections to fees?

If your objection to fees is that higher education is a public good and therefore students shouldn’t have to contribute on principle, I have bad news. Scotland crossed that ideological bridge a long time ago and is now £500m a year into that territory, because all student loan debt is a form of student contribution, whether it’s for living costs or fees.  Moreover,  there is no realistic chance that the Scottish Government is going to reduce its reliance on student loans to underwrite the higher education system.  £500m is roughly the annual cost of the whole FE system, or 1p on the basic rate of income tax.

A separate objection to fees is that they create a “weakest to the wall” market in higher education.   That’s not a necessary effect in the model above, in which the SFC continues to decide where the funded places are, and fully funds the fees of three-quarters of Scottish undergraduate students and half the cost of the rest.  It is entirely possible to seek a fee contribution from some students (or even all) in a system as planned as the current one, without moving to a quasi-voucher market.

Another objection is that fees change the nature of higher education, converting what should be a purely educational relationship into a purchase, and positioning students as consumers (some people are for this, but many are not). Around half of students in Scottish universities already pay fees, including many on full-time undergraduate courses (overseas students, rUK students).  Many others are already taking out large loans to pay  for their living costs. Would asking some, or even all, Scottish undergraduate students to borrow to cover some of the cost of tuition create a dramatic cultural shift from where we already are?  That’s debatable at best, I think.

But even if  you believe that all the things above would be unavoidable and undesirable,  is the price now being paid to avoid them defensible? In order to shelter everyone from any fee at all, we have designed a system which means student debt has to be shouldered disproportionately by those from lower incomes, while people from the most well-off backgrounds are routinely leaving university debt free.  It’s the least well-off students bearing most of the cost of these principles.

Investing in grants vs other things

One of the arguments often made for fees is that access to HE remains socially skewed and it would be better, and fairer, to subsidise HE students less and spend more on levels of education which everyone uses.  The model above doesn’t address that, because it doesn’t release any cash, it just moves it around between existing students.

The model also therefore doesn’t deal with the relative under-supply of places in Scotland compared to other parts of the UK.  A thousand extra places fully funded for fees and grant would cost around £10m. Nor does the model offer universities any additional funding per student: increasing university spending has often (though not always) been behind fee rises.

To deal with these issues as well in any serious way would mean a higher fee, and/or one which was less heavily means-tested, and/or ceasing to provide EU students with free tuition (recalling that none of the sums above included them).

SAAS funds just under 15,000 EU students. The total current spending on them is around £100m a year (15,000 x £7,000: they cannot claim maintenance grant). We don’t know yet what the Scottish Government will decide to do about this group.

In 2010-11, the SG was actively seeking ways to charge these students at least something (here). My assumption is that, once EU law ceases to apply and once the current commitment to the 2017 and 2018 entry cohorts has been met, the SG will return to the issue of how it can reduce its spending on this group in some way, so that some or all of the cash is available for other things. At a time other things are under pressure, the sum at stake simply looks too large at first sight to be affordable as a voluntary symbolic gesture.

Where next?

One of the great campaigning coups of the past 20 years has been the success with which  so many  people have been persuaded that free tuition is essential to widening access and that defending it must be given absolute priority over improving (or even just protecting) levels of student grant.   Thus grants in Scotland were cut by a third in 2013 with the support of NUS Scotland, and no outcry beyond the parliamentary opposition parties.   Grants are important too, it is sometimes conceded, but not so important we should give an inch on free tuition to spend more on them.  According to this view, the only proper way to increase grants is by finding the cash from some other budget, or more tax, and until that happens it is better to put up with what we have than to raid the fee budget.

I don’t expect any real shift in policy here or even in what people are prepared to debate. The SNP, the Scottish Greens, Scottish Labour and the Liberal Democrats all supported free tuition in the 2016 Holyrood elections (though at least the Conservatives, Labour and Liberal Democrats also mentioned increasing means-tested grants, and the SNP said it would “work to improve” them). The Conservative offering on fees was much more cautious than the model above, limited to something like the old graduate endowment, and would have raised a relatively small amount, and not for several years.

The appearance of any proposal like the one discussed here tends to trigger Spanish Inquisition-like questioning of Scottish opposition politicians about whether they will rule out tuition fees (grants don’t get a look in), with a moment’s hesitancy being taken as political death.  This – for the avoidance of doubt – is an absolutely brilliant state of affairs for people whose parents can fund them through university but much worse news for people whose parents can’t.

This positioning of tuition fees as a box which must never be opened even a crack benefits one section of society. It’s the one I know best, and it has always been good at identifying high-minded arguments in defence of its own interests. But rarely so successful as in this case at persuading other people that they must leave their barricade neglected, and come and defend its one instead.  It’s been a rather one-sided vision of solidarity so far.

But here we are. The maths of a more even sharing debt among students in Scotland is really pretty easy. The politics look as impossible as ever.

 

 

Footnote 1: Living costs don’t matter as much as fees because …

Students can live at home: (a) no, they can’t all do that,  (b) even for those who can, it will not always be a particularly good idea and (c) even when it’s a good option, those students still need to be fed, to travel (especially, often, travel) and to have clothes, books and so on, and it’s not reasonable to expect families on low incomes to absorb these costs unaided.

Also, that students can work is not a killer argument against the equal importance of living cost support.  There’s a growing literature on the impact of working, especially in term-time.  It’s not all discouraging: some types of working, at some level, for some people, appear to be fine. But the overwhelming message is that those students who don’t take on paid employment, especially in term, will tend to get more out of higher education, academically and in other ways.

But there’s a more fundamental problem with saying that fee loans are a problem, but maintenance loans aren’t, because people can work. It confuses the income and expenditure sides of the equation.   Logically, you might as well say fee loans wouldn’t be an issue, given a high enough level of grant, because people could work to pay their fees. Unless you accept the second of these arguments, you can’t use the first.

Footnote 2: Actual upfront fees – the 1998 reforms

In 1998, the UK government introduced an upfront yearly fee of £1,000. It was means-tested (this is generally forgotten), so that – roughly – the top third by income paid the whole amount, the middle third some of it, and the lowest third by income, nothing.  The dedicated fee loan had not yet been invented (though living cost loans were boosted with the idea people might choose to use some of the extra amount borrowed to cover the cost). The change was very unpopular with those who had to pay, and the way it was discussed obscured that many paid nothing or only part.  It was also accompanied by the abolition of grants, but that attracted much less fuss, as did their reintroduction in 2004.

The 1999 Scottish elections were dominated by 1998 fee regime and the sense that fees must be an immediate cost to families persists in Scotland still. However, when fees ceased to be means-tested in 2006, the UK government also enabled them to be deferred using a government subsidised loan. In passing, this means that nowhere in the UK since 1962 have first-time full-time low-income students been expected to find the cost of an upfront fee with no form of government help. But you could be forgiven for not knowing that, from the political rhetoric.

Footnote 3: Debt effects in England

Researchers looking at the statistics, and interviewing students, have discovered a high degree of willingness  (not necesssarily enthusiasm, just willingness) to borrow among young students of all backgrounds in England.  Participation rates there, including for those from disadvantaged backgrounds,  have increased at least as quickly as in Scotland.  This is not the same as saying no-one, anywhere has ever been deterred, and there’s more evidence that older, especially part-time, students are more debt averse.  But the last 20 years of data from England (and comparisons with Scotland) on participation levels and access have generally not been as helpful to advocates of free tuition as they might have hoped.

 

Long-term and short-term trends in education spending in Scotland: who’s right?

There was some debate at First Minister’s Questions this week about trends in school spending.  This post looks at the claims made (labelled below for ease of reference).

In brief, the increase claimed by the FM for last year relies on ignoring inflation (and rising pupil numbers). Her claim of a real terms budgeted increase for the year just started depends on including in the calculation the new additional activity required under the terms of the Pupil Equity Fund: like-for-like spending will see a real terms fall.

The Labour leader’s claim of a fall in spending of 7% since 2010 if anything appears to slightly understate the position.  The claim of £1.23 billion fall needs more definition to be able to make sense of.

What was said

According to the First Minister:

I think that I am correct in saying that the outturn figures for local government spending will show that spending on education has gone up (A).

 Data published on 27 June shows that councils are planning to spend £144 million more on education this year than they planned to spend last year—that is 3 per cent in cash terms and 1.3 per cent in real terms (B). Of course, that includes the planned spend on the pupil equity fund of £120 million that I spoke about.

According to Kezia Dugdale:

Her own Government’s figures show that, this year, spending on education is going down again in real terms (C).

The SNP has cut spending by hundreds of pounds on every single pupil, and it has cut spending on each secondary school pupil by more than £1,000. That is a 7 per cent cut by this SNP Government since 2010 (D).

I have come to the chamber time and time again to tell the First Minister that her Government has taken £1 billion from our schools. I was wrong. New figures show us that it is at least £230 million more than that—£1.23 billion has been taken out of schools on the SNP’s watch (E).
Sources

 

The First Minister’s comments are based on this news release, which draws on financial information routinely collected from councils, usually issued at this time of year as only a statistical news release. The decision this year to have an accompanying ministerial quote is unusual and therefore worth noting: it has been used as a hook to publicise the SG’s planned school reforms. The year on year change in education spending was actually slightly better last year, when there was no ministerial comment.

The first substantive point in this year’s statistical news release is that “Scotland’s local authorities’ provisional outturn total net revenue expenditure is £11.875 billion. This is a decrease of £0.119 billion (-1.0%) compared with 2015-16.”  I don’t think this was reported anywhere. The highlighting of the education number appears to have distracted from the wider picture on local government finance.

Kezia Dugdale’s figures come mainly from research from SPICe, which doesn’t appear to have been published, but more detail has been released to the press, as reported by The Herald.  Some of the  Labout numbers can however be checked using the same data series as  was drawn on by the FM.

A

“Outturn” – what councils actually spent on education – rose between 2015-16 and 2016-17. The increase was £86m, or 1.76% in cash terms.  The Treasury GDP deflator for year on year change between 2015-16 and the FY2016-17 was 1.98%, implying that the cash increase last year was a little below inflation. Pupil numbers in publicly-funded schools rose by 0.6%.

Taking all these figures together the implication is that cash spending on education rose in 2016-17, but that spending per pupil in real terms fell by 0.8%. So claim A is correct, but less impressive than it sounds.

B

Planned spend is what councils budget to spend at the start of the year. The press release states that education budgets for 2017-18 are “£4.970 billion, an increase of £0.144 billion (+3.0%) from that budgeted in 2016-17”. The FM’s analysis of the change as 1.3% in real terms is correct: the Treasury inflation estimate for the year ahead is 1.63%. Compared to a 2.89% cash increase (the change in more detail) that gives 1.26% in real terms.

A further 0.6% rise in pupils numbers is projected. Real terms funding per pupil is therefore due to rise by 0.7%.

However, as the FM notes, the Pupil Equity Fund accounts for almost the whole rise (£120m out of £144m). This matters, because although the PEF is undoubtedly money for schools, it comes with clear guidance on its use: it is intended for new activity. As the SG’s operational guidance puts it (emphasis added by me):

The Pupil Equity Funding must enable schools to deliver activities, interventions or
resources which are clearly additional to those which were already planned

So the like-for-like change in budgeted spend – what schools can spend on what they were already doing – is low: just £24m. That is 0.5% in cash terms, or a real terms fall of 1.2%.  If I factor in the projected rise in pupils numbers, the real terms fall in like-for-like spending per pupil in the coming year is nearer 1.7%.

For claim B, a lot therefore hangs on the inclusion of the PEF in the year-on-year comparison.

C

Kezia Dugdale’s claim that education spending is going down this year therefore holds if looking at like-for-like, but not when including the money for additional PEF activity.  It holds for last year unambiguously for real-terms spending, though not for cash.

D

Labour hasn’t (as far as I can see) published from the Scottish Parliament Information Centre (SPICe) analysis on which Kezia Dugdale’s further figures are based.  However, the same outturn figures used above are published by the SG going back to 2008-09 and can be used to look at longer-term trends. Here’s what they look like, pulled together:

Untitled

So I get a real-terms fall in total spending over the period covered of nearer 8% rather than 7%. Taking the figures just from the high-point of 2009-10, as Labour seems likely to have done, gives me an 8.7% real terms fall.  It may be however that SPICe has picked up a nuance in the figures which affects the detailed calculation – the broad picture is similar.

I’ve included funding per pupil, as pupil numbers have increased slightly over the period: the percentage fall per pupil in real terms across the whole period above is 8.3%.

Labour have provided more detailed figures  to the press . The Herald  quotes a real terms fall per secondary pupil from £8,033 in 20101-11 to £6,892 in 2016-17 –  a 14% drop, at the same time as the exam system was being bumpily overhauled.  That might explain why secondary teachers have been particularly vocal in recent years about the pressure they feel under. The Herald reports that the funding per head in primary over the same period stayed steady in real terms, at £4,826.

E

All the figures above are for revenue (current) spending only. I make the reduction between 2010-11 and 2016-17 in revenue spending to be around £0.5 billion a year in real terms.  So I can’t make sense of the Labour figure of a £1.23 billion drop since 2010-11, unless

  • it includes capital, which these figures don’t cover, or
  • it is using the now-common device of adding up how much extra would have been spent over the whole period up to 2016-17, if the real terms figure in 2010-11 had been maintained. But in that case I get a cumulative “gap” figure closer to £2 billion.

I haven’t yet managed to find a more detailed explanation of this figure that’s available to the public, so I am at this point stuck.

 

Conclusion

Numbers will always get thrown around at FMQs and the FM is hardly the first politician to look for the most positive story. What matters more is that whatever they say in public, ministers understand in private what the reality is – and it is clearly of a non-trivial fall in education spending over the first part of the decade, a modest recovery in 2015, and then a continuing smaller underlying real terms fall, even still, on existing activities, as the graph below shows.

Education spending

It looks to me as though the Opposition could, if anything, have gone harder on the real terms fall in revenue spending, but there may be reasons for being more cautious than the figures above suggest.  It would however be easier to make sense of their claims if they showed more of their workings.

The rise, fall and rise of student debt in Scotland and the contribution of different administrations to debt gaps within the UK

Scottish students tend to leave higher education with lower levels of debt than those elsewhere in the UK.  Some of that will be due to the higher proportion of people here who only stay in the system for one or two years (mainly HNC/D students).  Some of it may be down to a higher proportion living at home.  But policy divergence has also evidently been a large part of the story.

Differences in debt are sometimes presented as specifically an achievement of the past decade, but the figures below suggest it’s been a function of devolution more generally.

The changing picture on final debt across the UK

Using this week’s figures on final student loan from the SLC, it’s possible to chart how debt has changed in Scotland over the past 17  years.  It turns out to be a tale in three parts: rise, fall and rise again. It’s also possible to unpick exactly how levels of student debt in Scotland have diverged from those in the other UK nations

Disclosure: I was working as a civil servant on policy in this area between 2000 and 2004, specifically the implementation of the graduate endowment and Young Student Bursary. Readers will want to be aware of  my background in reading this.

This chart shows the average final debt of students as they entered repayment in each year from 2000. There’s a time lag, so for example the 2017 figures relate largely to those who left HE in 2016. If they completed a degree, they will have entered no later than 2013.

UK loan graph

The divergence between Scotland and other UK nations set in quickly after devolution in 1999.

It’s possible to identify three distinct phases in student loan change in Scotland.  First there was a phase where debt rose, then one where it fell or barely changed, and one last one where it rose again.

The chart below shows the annual change in debt each year in each nation, separating into the rise/fall/rise periods for Scotland. 2000 is the earliest year covered by the recent SLC data.

Loan graph UK 2

 

2000-05

From 2000 to 2005 final loan in Scotland rose each year.  In the early years most students leaving had studied under pre-devolution arrangements. As time passed, more leavers had been under the “Cubie” arrangements brought in for new entrants in 2001 by the Labour/Liberal Democrat coalition. This

  • reintroduced a grant for younger students called Young Student Bursary (grants were abolished UK wide just before devolution)
  • abolished a means-tested annual upfront fee of £1,000
  • introduced a single post-graduation payment (of around £3000 at current prices) for around half of students (in effect young students on courses of degree length: HNC/D students and all mature students were exempt).
  • reduced the amount of maintenance loan available to those from higher income families.

The first one-year HNC students under these rules appear in the chart above in 2003, HNDs in 2004, 3 year degree students in 2005 and 4 year students not until 2006.

Over 2000-05  Scotland peels quickly away from the position in the rest of the UK, finishing at about half the debt level. Debt in the other UK nations rose much more sharply between 2001 and 2003, before leveling off. At this stage, all three other nations are very similar: Wales and Northern Ireland were following the English model (Wales didn’t have powers to deviate, Northern Ireland had other pressing concerns).

2005-2011

Having begun to flatten out by 2005, average final debt levels in Scotland then fell every year until 2011 (except in 2009, when they rose very slightly).  In 2005, 2006 and 2007, those entering repayment were graduates of the Labour/Liberal Democrat scheme (or in a few cases, one of the pre-devolution ones).  Grants had also been increased in 2005, reducing debt.

In 2008, those entering repayment benefited from the incoming SNP administration’s abolition of the graduate endowment in 2007. Debt continued on its the falling trend (other than the modest rise in 2009) up to 2011.

Over the same period debt in the other UK nations pulled away further, particularly in England and Northern Ireland, which both moved to a higher £3,000 fees (plus grant) system for new entrants from 2006.  Most of these entered repayment in 2010.  By 2009, the use of new powers in Wales shows in its less quickly rising line (grants were increased and fees limited).

The abolition of the graduate endowment contributed to a continuing downward trend in Scotland. However, the effect wasn’t enormous: average final debt fell by just under £600 (-8%) between 2007 and 2011.  Most of the increased difference between Scotland and the UK by 2011 is accounted for by the decision by the Labour/Liberal Democrat coalition in Scotland in 2004 not to adopt the new £3,000  fee model brought in in England in 2006.

2011-2017

From 2011, final loan increased every year in Scotland. A finding that I hadn’t expected is that it rose by almost exactly the same absolute value in all three devolved nations (£5,150 in Scotland, £4,590 in Northern Ireland and £4,915 in Wales, though quite erratically). Only England, adopting a £9,000 fee for new entrants from 2012 broke away.

The rise here was due to the Scottish Government increasingly turning to loans to fund living costs, particularly after 2013, when substituted loan for one-third of grant and generally increased total loan entitlements.

 How did loan end up so much lower in Scotland?

These charts unpick the process by which average loan in Scotland has departed from the levels elsewhere in the UK.  It becomes clear that it has happened in these stages:

  • active policy-making by the Labour/Liberal Democrat coalition – introducing  the Cubie package – accounts for Scotland having around half the average debt of the the rest of the UK by 2005.
  • passive policy making by the Labour/Liberal Democrat coalition – declining to follow the example in England, Northern Ireland and Wales of £3,000 fees – accounts for a further widening of the gap after 2009.
  • active policy making by the SNP administration – abolishing the graduate endowment – has some effect but much less than the others here.
  • passive policy making by the SNP administration – declining to follow the example in England of £9,000 fees – has a large effect on the difference with England, but none with that for Wales and Northern Ireland.

The difference in final debt levels is generally claimed as an example of the success of the current government in Scotland. However, it becomes clear that much of that difference is due to the position it inherited from previous administrations and that its own active contribution to that difference is a relatively small part of the story.

Its most substantial contribution to differences in debt has been not to follow the £9,000 regime in England.  This is a wholly devolved area and no political party or civic Scotland body in Scotland has advocated this model, at least in public. So the main challenge facing the government in not following England has been in finding ways to pass to other parts of the budget any negative impact through Barnett of reduced spending on English universities. (It’s not clear how large a task this has been, however, as the SG is keen not place a cost of free tuition, some of the money saved in England may have been re-spent on other devolved areas, and no savings elsewhere have been specifically attributed to tuition fee policy).

Put briefly, these comparisons bring out that the difference in debt levels between Scotland and the other devolved nations is largely an achievement of governments elected in Scotland prior to 2007, and that the current government’s contribution to the difference with England is largely down to a decision not to use devolved powers to do something no-one here has ever asked it to do.

 

Footnote

Figures underlying the graphs here UK debt for blog

 

Student loan for Scottish 2016 leavers up 13%

The Student Loans Company has published the average final debt for students who left HE last year: link here.

The figure for Scotland is now £11,740, an increase of 13.3% on the previous year.  The longer term trend is more striking: see chart below. Average final debt has roughly doubled in cash terms (the real terms rise will be less dramatic, but still substantial) since the current Scottish Government entered office in 2007  promising students that it would “dump the debt”.

Screenshot 2017-06-15 at 09.36.39

Source: SLC

The rise since 2012 is due to the changes to student funding implemented in Scotland in 2013 still working their way through.  What’s pushing up debt is the substitution of loan for around one-third of grant, and the general use of loan to increase living cost support across the board, but especially at middle-to-high incomes.  There’s potential for a further step up next year, when the first cohort on four year courses who have studied wholly under the 2013 reforms will be included.

The figure for Scotland remains lower than elsewhere in the UK. That’s partly due to there being no fee debt for those staying here, but the size of the gap with other UK nations is exaggerated by the higher proportion here who leave after doing a one or two year HNC/D.   That will bring down the average.  Roughly, it appears to mean that the Scottish average doesn’t represent the average after 4 years (let alone the 1+4/2+3/2+4 models used by half of those moving from college to university). It’s closer to the average over 3 years.

The next nearest UK nations for debt levels are Wales (£19,280) and Northern Ireland (£20,990). With its £9,000 fee regime pretty much fully rolled out, England now sits £32,220: this figure will rise further, given grant cuts for new entrants from last autumun, but that won’t show until this group leaves in a few years’ time.

In comparing Scotland and Wales, in particular, it’s worth remembering that in Scotland low-income students borrow above average each year, while in Wales the opposite applies.

So there are a few reasons these figures don’t provide a good guide to the reality of final debt for low-income students leaving university in Scotland, or comparing with other parts of the UK.  That won’t stop them being quoted in support of the Scottish status quo.  But their limits shouldn’t be forgotten.

More here on the impact on the numbers of HNs and the skewing of debt towards lower incomes in Scotland.

 

 

 

Comparing entry rates to university: the (data) gap that could easily be reduced

The First Minister was interviewed yesterday by Andrew Neil. One journalist reported that “Sturgeon says difference in university access rates for poorer pupils between Scot & Eng down to different sets of figures.”

I didn’t see the interview, but this sounds plausible.  The Scottish Government has become increasingly confident and robust over the past year in dismissing UCAS figures as a basis for making cross-country comparisons in entry rates to university for young people from disadvantaged backgrounds. UCAS figures show 19.5% of 18 year olds from most disadvantaged 20% of areas in England entering university, and 10.9% for Scotland. It’s worth noting that this gap is not uniform across young people: it disappears as levels of disadvantage fall.  For those at the other end of scale, there’s hardly any cross-country difference in university entry rates at 18.

There are two possible problems being invoked by the FM here. One is a technical point about different ways of measuring area disadvantage. For England UCAS uses a measure called POLAR3, which looks particularly at education disadvantage: for Scotland, the figure above uses a different measure called SIMD. However, I haven’t seen an argument from the Scottish Government that specific differences between POLAR and SIMD explain the gap above.

It’s much more likely this argument is about the coverage of UCAS data.  UCAS doesn’t include all the 18 year olds who go to college and from there to university, which is a much more common route here than south of the border. The Scottish Government leans on this point increasingly hard, and it is an absolutely true observation about the coverage of the UCAS figures.

There’s a substantial argument to be had about how analogous college-to-university entry is with direct entry. In particular, slightly under half of those who do it (in total, at all ages) get full credit (“advanced standing”) for their time at college. The rest mostly go back to square one and into the first year of a degree.  A few get partial credit (“advanced progression”), such as doing two years at college and then starting in second year at university.  Anything other than full credit means repeating one or two years, depending if students have done an HNC or an HND. That means five or six years of full-time study to get an undergraduate degree which would have taken a direct entrant four years. That in turn carries extra living costs, and means later labour market entry, putting these entrants at a relative disadvantage.

These cases also raise a reasonable question whether their English analogues are 18 year old direct entrants, or those south of the border who also took a couple of years doing something else before starting their degree course, and who also are not in the age 18 UCAS figures there. Of course, we shouldn’t only be interested in school leavers.  But there are particular reasons to monitor how far access to starting a degree programme straight from school is socially biased.

Does the SG have the data on how many people entered college at 18, from the most disadvantaged 20% of areas (SIMD20), and how many of those got full credit? This information isn’t published, but other material suggests it could be calculated.

The SFC publishes how many college-to-university movers got full credit (Table 22 here): it was 3,999 in 2014-15.  The SFC notes that after 2014-15 numbers will have been rising: but it appears not very sharply. The “national ambition” for 2016-17 for those with full credit is 4,100.  Elsewhere it has published that 71% of all movers in 2013-14 were under 25, and 23% of them were from SIMD20 (para 10 here). So it feels like it should at  least be possible to isolate the number who moved at an age consistent with having entered college at 18, were from SIMD20 and got particular amounts of credit.

Without those particular figures being public, there is still a way to test how likely it is that the UCAS gap at 18 is closed by college-to-university movers who get full credit, from the figures we do have.

UCAS reported 1,345 SIMD20 age 18 Scottish entrants in 2016. To make that 10.9% into something like the 19.5% in England would require that figure to be a little over 1,000 higher.

1,000 19 and 20 year olds from SIMD20 areas getting full credit would equal around one-quarter of all those getting full credit from all ages and all backgrounds.  That feels high.

If I assume that those aged 19 or 20 make up one-third of all people at all ages moving with full credit, and that they are half as likely again to be SIMD20 as the entire college-to-university population (which isn’t an obvious assumption, given disadvantage tends to delay engagement with higher education), I can get this group to 11.5% of the full credit group (33% x (23% x1.5)). That only half closes the gap, and these to me feel like quite hopeful assumptions.

So it’s not immediately obvious from what we do know that “UCAS leaves out colleges” is necessarily a powerful rebuttal to concerns about how Scotland compares to the wider UK in giving young people from disadvantaged backgrounds access to university-level study on equal terms.

Including those who don’t get full credit would come close to closing the gap: although that would still be using the same optimisitic assumptions.  But in any case, as argued above, it’s open to question whether those who had to go back to the start of the process after a year or two at college should be included in any comparison of age 18 university entrants, even if they went straight to college at 18.

It’s frustrating that while the Scottish Government has been keen to stress what’s missing from the UCAS data, it hasn’t moved with equal speed to use the information it holds to fill that gap, even with a few footnotes and caveats. At the moment, “ignore this data, it doesn’t tell you the whole story, but we aren’t saying what whole story is” is being too easily played as a get out of jail free card.   It’s time to move from general dismissal to the actual numbers.

 

 

 

 

 

Scottish Government special advisers: a quiet re-jigging round HE, and expansion in number

The Scottish Government has quietly re-jigged its Special Advisers. Kate Higgins returns to the education brief for early years, and further and higher education.

Ms Higgins had a general education brief from April 2015 (more here) until  June 2016, after the last Scottish elections and John Swinney’s appointment as Cabinet Secretary for Education. At that point, she was moved to Rural Economy and Connectivity (see here). Education was passed back to Colin McAllister, one of senior special advisers, who had held the brief before her.  Mr McAllister retains a general education brief, and so must be assumed still to be the lead adviser on schools policy.

The changes also include the appointment of a new special adviser, giving 13 special advisers,  4 women and 9 men.

The new appointee is Stewart Maxwell, formerly Convenor of the Scottish Parliament’s Education Committee, who lost his seat in the 2016 Scottish Parliament elections. Back on 13 January he announced on Twitter that he was at end of his first week as a special adviser, implying he had started around 9 Janury.  I have been periodically checking for an official updated list ever since.

Searching  at the time, or indeed month later (I last recorded a search  on 8 February), only produced an old list from September 2016, under “Transparency data” on the SG site.  But doing so today has produced a new list (dated 17 January).  This shows that Mr Maxwell now has responsibility for “Business, the Economy, Skills and Fair Work. Business and Economy outreach”, some of which is responsibility transferred from Jeanette Campbell, who retains “Communities, Social Security, and Equalities”.

It took nearly a fortnight to update the government website to reflect the expansion in numbers and reallocation of roles and it then took several weeks more for the updated list to turn up on a search.   I am pretty sure that back in February I didn’t just Google, but also searched the SG site directly (if so, it would have been the SG’s  new “beta site”, where the January list can now be found), but in fairness to the SG,  I am not completely certain about that.

More importantly, searching “special advisers” in the PQ section of the Parliament’s website as of today still produces an answer on 23 September 2016 as the most recent.    So in a break with convention, the Parliament appears never to have been informed of the changes, including the new appointment. By contrast,  in April 2015, June 2016 and  September 2016, a written PQ was used to set out the revised SpAd team on or before the day the SG updated its website.

I don’t know what conventions now exist here but, whether or not any do, it would be desirable to get back to making Parliamentary announcements at the time changes are made to the team of special advisers, rather than relying on Twitter and quiet website updates.  Otherwise, what should be easily found and scrutinised public information about the location of power and influence in government becomes the currency of the grapevine and those in the know, reinforcing the sense of government as a game of insiders and outsiders.

The Scottish Government recently declared itself a “global leading light in the campaign for more open and accessible government”.  Going backwards in terms of openness and accessibility in relation to special advisers suggests that there’s still a bit of work to do making good that commitment.

How far is Scotland from meeting its 2021 widening access targets?

This post looks at what today’s data from the Scottish Funding Council suggests about the scale of change needed to meet the Scottish Government’s targets for widening access to university.

The 2021 targets

Government’s targets come from last year’s report by the Commission on Widening Access, which recommended that (emphasis added):

• By 2030, students from the 20% most deprived backgrounds should represent 20% of entrants to higher education.

Equality of access should be seen in both the college sector and the university sector. To drive progress toward this goal:

• By 2021, students from the 20% most deprived backgrounds should represent at least 16% of full-time first degree entrants to Scottish HEIs as a whole.

• By 2021, students from the 20% most deprived backgrounds should represent at least 10% of full-time first degree entrants to every individual Scottish university.

• By 2026, students from the 20% most deprived backgrounds should represent at least 18% of full-time first degree entrants to Scottish universities as a whole.

• In 2022, the target of 10% for individual Scottish universities should be reviewed and a higher level target should be considered for the subsequent years.

[Note added after original post: after writing this I came across a target of 15.5% of HEI entrants by 2019-20, used by the Scottish Funding Council – see para 56 here. The SFC attributes that to the Commission, but no such target appears in the Commission report.]

The distinction between higher education and universities (or HEIs, a category comprising universities plus 3 other specialist HE providers) is important,  because “higher education” includes the substantial amount  of HN-level HE provided in colleges in Scotland, which already have intakes which are relatively well-balanced by social background.  It is in the universities where unequal access by background persists.

The remit of the Commission was specifically to think about access to university. However, the Commission’s report looked more widely, and the newly-appointed Commissioner was keen to stress last week that he saw his brief as extending across all higher education:

The major thing that I can do is not to focus too strongly on just the universities’ contribution and to give greater recognition to what colleges can and do deliver.

Sir Peter Scott, 25 January 2017

The 2021 targets above however are clearly about universities, so it’s worth looking at the data released today by the Scottish Funding Council.  That’s because it is the nearest thing we have to a statistic measuring the access targets’ focus: the percentage of undergraduate entrants to universities/HEIs from the most disadvantaged areas.

The SFC statistics are not a perfect reflection of that target. They are only for full-time entrants under 21. Adding older entrants  would almost certainly boost the share of students from disadvantaged areas, but the overall effect is likely to be modest. There are far fewer full-time entrants of 21 and over, and any greater skew amongst them  towards the most deprived areas isn’t likely to be large enough to affect the overall average much.

On the other side, the SFC figures exclude entrants from outside Scotland, who tend to be more advantaged than the Scottish student body as a whole. Although it is not explicitly limited in this way, I assume here that the target is concerned with the percentage of disadvantaged Scots as a proportion of all Scottish students. If not, then the challenge is much greater, and the gap much wider, than shown below. I am encouraged in focussing on Scots by the Access Commissioner’s interpretation of his remit:

I see the primary responsibility of my role—and I think that this was the prime focus of the commission—as being to focus on fair access for Scotland-domiciled students. I do not think that I have any remit to make access fairer for students who come from England or Wales to attend Scottish institutions, although I accept that the social composition of those students changes the flavour and affects the culture of at least some Scottish universities.

(Source as above)

The SFC data

The SFC statistics are reached via the link at the foot of this page.

Here’s the total proportion from the most disadvantaged 20% of areas (SIMD20) since 2006-07, with the 2021 (all age) target added for comparison.

screenshot-2017-02-02-at-18-07-19

[Note added after posting: for the separate SFC target mentioned above, imagine a line almost as high, 15.5%, but 2 years sooner, in 2019-20.]

In 2013-14, some additional HEI places ring-fenced for widening access were released (more on the effect of that here). That explains the increase from that year. But intervention on a completely different scale appears likely to be needed over the next 5 years, to reach an average of 16% across all HEIs.

The target also includes a minimum of 10% for individual institutions by 2021. Figures for the last 4 years are shown below for each HEI.  2012-13 is marked in yellow, as the last year before additional places were released.   Those institutions marked * received some of these places.

screenshot-2017-02-02-at-18-08-07

Only 4 HEIs have been above 10% for  each of the last 3 years (Dundee, Glasgow Caledonian, Glasgow School of Art and the University of the West of Scotland).

7 HEIs have never crossed this line: of these Aberdeen, Edinburgh, RGU and St Andrews are all below 7%. The remainder have hit 10% at least once in the past 3 years, though some are more consistently close than others.

There is, noticeably, a lot of volatility at institutional level. Many HEIs show no clear trend, despite the emphasis placed on widening access in recent years. Of those receiving extra places, Dundee, Stirling and Glasgow School of Art stand out for increasing the share of their students from disadvantaged backgrounds and sustaining that, over the whole period these were  distributed.

In short, getting everyone over the 10% line by 2021 looks very hard. Keeping them there looks harder still. The institutions with furthest to go also tend to be those with younger intakes, where adding older entrants is least likely to help.

Getting to 2021

There hasn’t been much acknowledgement of what demanding targets the Commission set, and the Government has accepted. However, at last week’s Committee hearing the new Access Commissioner said:

I suspect that some of my responsibility might be to manage down expectations about what the commissioner can deliver.

That seems absolutely fair. At the same session, the Commissioner clarified that his is a 3 to 5 day a month appointment (although in practice he expected to be in Scotland 5 days a month, and to be working a similar amount – unpaid presumably – beyond that), with no assigned budget, at least as yet.   It’s a much more low-key arrangement than the Office of Fair Access in England, no doubt deliberately so. Sir Peter also stressed that he wanted to take time to understand the situation here better and was not yet ready to move into “pro-active mode”.  His first  annual report is due in a year’s time, that is, early 2018 – less than three years before applications will be submitted for entry in 2021 [added: and less than a year before applications begin for 2019].

So, time is tight, and these targets are tough. A system which increased the share of the most disadvantaged from 8.7% to 10.8% over a decade, with the help of dedicated additional places, is now expected to increase from 10.8% to 16% in just 6 years [or even more demandingly, to 15.5% in 4 years, according to the SFC], with no promise of further expansion. Over half of institutions will need to increase substantially (in some cases more than double) their performance over the same period. Even if including older entrants might provide a boost of, say, a couple of percentage points to the national average, the only sensible response to this ambition is – blimey. And to wish everyone involved good luck. It’s not that it can’t be done, but it needs to be acknowledged what a huge task it represents.

Displacing the privileged vs squeezing the middle: a bit more evidence

The discussion in Scotland about widening access to university remains dominated by the idea that any “displacement” will be a problem mainly (or even only) for the most advantaged.  As the journalist Kevin McKenna put it this weekend (“Scottish Higher Education Should be for all, not the gilded few”):

Conservatives have their own weasel word for this process; they call it “displacement”.  What they really mean is that the privileges affluent children have enjoyed for centuries may be put at risk by giving privileges to poorer children. We will hear more people talking about “displacement” in the next few weeks as we begin to digest figures released last week by Ucas, the university admissions body…..

As university places for Scottish students are capped as a result of the SNP and Labour’s successive policies of free tuition, it means that, inevitably, there will be a squeeze on places going to middle-class students if more poor children gain access. ….

There are vast, sprawling neighbourhoods in Scotland’s major cities where children, no matter how bright they may be, will rarely have the opportunity to make the most of their academic abilities by attending our top universities.

Within a few miles of these places there are more arboreal districts where children enjoy an eye-watering assortment of advantages in the race for university places. These range from hours of expensive private home tuition to flat financial inducements for every higher A pass gained…..

Thus, our universities are diminished by choosing from a very narrow gene pool. Their lecture theatres are not filled with the brightest and the best, just the brightest and the best of a narrow social stratum.

McKenna is right that the UCAS data is worth careful attention for what it shows about changing patterns of participation. That’s what this post looks at.

Not everyone is either very rich or very poor

The data reminds us that between McKenna’s “vast sprawling neighbourhoods” and “arboreal districts” are a lot of places which are neither acutely deprived nor especially privileged. It adds to the evidence that it is the people from these in-between places, rather than the more arboreal ones, whose access to higher education is now under most pressure.  It turns out, quite predictably and no doubt for all the reasons that McKenna’s article identifies, that those at the very top are proving to be resilient occupiers of university places.

What the UCAS data does

The latest UCAS data release includes an“experimental data explorer” which enables anyone to graph trends in a variety of factors relevant to equality in university access. Setting the parameters to Scottish institutions and SIMD provides data just on Scottish-domiciled students at Scottish universities. It tells a story of dramatic change, due to policy intervention.

Offer rates

The offer rate is the proportion of applications made from a particular group which resulted in an offer from a university.

Here it is for Scottish 18 year olds, applying to Scottish universities.screenshot-2017-01-30-at-00-42-13

This shows that in 2013, applications from SIMD 1 and 2 (the 40% most disadvantaged areas) went from having the lowest offer rates to doing as well as those from SIMD 5 (the least disadvantaged 20%).  In 2014, this continued, so much so that SIMD 2 applications had the highest offer rate of any group.  In 2015 and 2016, their offer rates fell back, as these did across the board, but remained as high as for those in SIMD 5 and SIMD 4.

SIMD 5 offer rates have held up pretty well over the same period: they are still as high as they were in 2012.  Much the same is true for SIMD 4, though  there are signs of a slightly sharper drop in the past year: they are now a little below their 2012 level.

It is applications from SIMD 3 areas – the 20% in the middle – which now sit at the bottom of the table, and have seen the clearest decline over the period.

Actual vs expected offers rates

Behind this, UCAS provides an even more revealing graph. This considers how the actual offer rate for a particular group compares with what would have been expected from the overall level of offers that year, based simply on candidates’ grades and choices of course. As I recall, this statistic was first produced in response to the concern that some ethnic minority groups were getting fewer offers than their grades and choices would have predicted.  It therefore functions as a rough measure of how far the system favours, or not, certain groups.

The first chart below shows what UCAS calculates the average offer rate would have been in theory in each SIMD quintile, just based on grades and courses applied for. It shows that the likelihood of getting an offer would be expected to decrease as disadvantage rose, reflecting decreasing exam scores.

screenshot-2017-02-01-at-16-54-30

The graph below shows how the actual offer rate (the figures in the very first graph shown above) depart from this.   A positive score shows that a group has done better than their grades and course choices would predict

For Scottish 18 year-olds by SIMD, it looks like this.

screenshot-2017-01-30-at-00-42-26

This is a powerful image of positive discrimination.  From 2013, it shows SIMD 1 and 2 applications becoming much more likely to receive an offer, despite lower results. By 2014, they see the largest positive benefit of any applicants.

This doesn’t mean there was displacement by SIMD 1 or 2 entrants during this period. The improvement for SIMD 1 and 2 was due to the release of additional places ring-fenced for widening access from 2013-14 onwards. As these charts show, this meant in practice a concentrated effort on applicants from SIMD 1 and 2. To fill these extra places, universities clearly dropped their entry requirements for these applicants, but not for others, and this produces the graph above.  The UCAS site makes it easy to look at the pattern for individual institutions, and this confirms that this effect is due to changes in the data for those universities which received the extra places (see here).  You’ll rarely see the impact of a specific policy intervention illustrated so unambiguously.

The falling lines for SIMD 3 and SIMD 4 show that ordinarily these groups would be expected to benefit from a fall in the average entry requirements, because they include a fair number of people with less strong results. But in this case, they didn’t, as these lower entry requirements were limited to more disadvantaged groups. But the larger issue for these groups has been the absence of any growth in places more generally, as general demand has risen. SIMD 3 in particular are the only group to have started and ended the period less likely to receive an offer than expected.

SIMD 5 meantime turn out to have had a historic advantage in terms of higher-than-expected offer rates, which has fallen since 2012 but persists (in contrast to SIMD 4). Life is not much disturbed in McKenna’s arboreal districts. They are a resilient group, due to a combination of stronger exam results and – this graph suggests – some additional advantage they enjoy in obtaining offers (which could be something as simple as getting better advice on the tactics of applications).

Looking ahead

Further extra places are not promised (there was some controversy a year ago over the future of the targeted places scheme).

From here on, it seems more likely that any growth in participation by those from disadvantaged areas will involve constraining admissions from another. From the graphs above, who looks most vulnerable?

The latest UCAS statistics suggest that far from there being a single homogenous “advantaged” group, serious advantage is more resilient than the moderate form.

To loosen the grip of the most privileged on the university system, to “displace” them, would almost certainly require policies as focussed on that group, as recent policies have been on SIMD 1 or 2: limits on the numbers who can be recruited, financial penalties on universities for over-recruitment, substantial financial incentives for such students to go elsewhere (to do the apprenticeships increasingly promoted as equaivalent to degrees, for example).  Without such policies, which it’s hard to see any politician enacting,  a government might be a champion of the poor, but it is unlikely to disturb the wealthiest much.

The assumption made by Kevin McKenna, and others, still too often seems to be that in a system that isn’t growing,  adding more people from very disadvantaged backgrounds  will some how force out those from the opposite end –  like a single strike on a Newton’s cradle. But that doesn’t take into account how effectively serious socio-economic advantage defends itself. The UCAS data here suggests we should imagine something more like a pouch of liquid in a vice. There will be one group protected by government intervention (rightly) and another by their families, squeezing a less well-protected group. This will typically contain first or second generation students from modestly-funded state schools with average catchments, perhaps limited family knowledge of HE, capable but not clutching straight As.  It would be ironic if people concerned about social justice were so transfixed by the idea of getting one over the rich, that they failed to notice it was these people who face the greatest squeeze.

 

 

Footnote 1: Offers are only part of the story

This data is dramatic, but is only one part of the story that gets someone into university. Offers can be rejected, exams not yet sat can disappoint (although in Scotland, more offers have traditionally been based on existing Highers results), people can change their mind just at the point they are due to start. The offer rate is however worth examining in isolation because it is the part of the system which most directly reflects decisions at university level.  The data explorer does also provide information on placed applicants per 10,000 population by SIMD, but this is affected by variations in the application rate in each group over time.

Footnote 2: All-age data

For those at all ages, UCAS only provides data on the offer rate (the first  graph above). That shows less dramatic change once older applicants are included, implying that for applicants through UCAS, the policy intervention was concentrated on school leavers. This data will include some of those who started at college on an HN-level course, and then moved to university. However, only some of these cases will go through UCAS. Including this group might well alter the picture again, as a number of the ring-fenced access places were for articulating students.

screenshot-2017-02-01-at-18-54-29

Footnote 3: Are there alternatives to funding more places? 

This piece for The Herald discusses in more detail what other safety valves there are in the system. One ought to be a temporary fall in the number of 18 year olds, although that is not having as much effect as might be expected. The most substantial change would be EU/EEA students ceasing to be entitled to SG-funded places, post-Brexit.

Baby boxes – what’s happening in the rest of the UK (and elsewhere)?

In the discussion of the launch of the baby box pilot in Scotland last week, some references were made to schemes also being launched in England.  This post gathers up information about what’s happening there, and in some other places, to help put the Scottish scheme in context.

It reveals that the explosion of interest in baby boxes would make a great case study for someone, into how ideas gain attention, how different people and organisations play a role, and the use of evidential claims. And that a mixture of enthusiastic support and scepticism about some of the claims made isn’t unique to Scotland.

Warning: this is very long. It’ll be of interest mainly to people fascinated by the inner workings of the policy process and/or interested in infant health; it may be a heavy read for others. You can skip to the end for some reflections on what we learn that might be useful for the Scottish project.

Background

Baby boxes have been used in Finland since the 1930’s, but it’s only in the last few years that there’s been a surge of international interest. Many articles and interviews trace this back to a BBC piece from June 2013, which received huge attention worldwide, becoming one of the most shared and read stories the BBC had ever published, as discussed in another BBC piece here.  Reflecting still later again on their coverage of baby boxes, a BBC editor said:

“What the original baby box story showed beyond doubt was that a story about parenting and public health policy can, in certain circumstances, go viral. I had not seen that coming.”

Flipboard 11 April 2016

A follow-up BBC report in July 2013 noted that the article had prompted lots of immediate interest and approaches to the Finnish Government, and  this in turn appears to have led the Finnish Government to send a box to Prince George, prompting further coverage.   The original BBC article seems to have  been a simple product of journalistic curiosity and research.  The BBC ran a piece following up its original report in April 2016.

England

I’ll mention first, as an outlier, very specific research being undertaken at Durham University on using (clear plastic) boxes as sleep spaces.

The most publicised baby box scheme in England is one being piloted at Queen Charlotte’s and Chelsea Hospital, London, part of Imperial College Healthcare NHS Trust. There, 800 boxes are being given out to new parents first come, first served.  Parents’ use is being monitored for 8 months and they are required to sign up to “the Baby Box University”, an on-line information resource (more on this below).  The boxes and on-line resources are being supplied for free to the Trust by the Baby Box Co., a US-based company (again, more about them below).  The educational element provided via the Baby Box University is persistently highlighted by the company as an important component of what it offers.

The publicity for this scheme also mentions schemes taking place at a number of other sites in England, all also using the Baby Box Co/Baby Box University model.  This article gives the best overview of how the English initiatives are linked and Imperial College’s co-ordinating role. Further detail on some of the other sites here: ColchesterSandwell and West BirminghamNorth Middlesex (not all the places listed as possible pilots seem to have started yet).  The Baby Box Co. has said:

The Baby Boxes which are now being delivered en masse to UK parents free of charge through community partnerships account for a much greater percentage of our operations [note: by comparison with commercial sales in the UK]. The products included in these Baby Boxes are unique to each hospital. For example, the contents included in the North Middlesex University Hospital Baby Boxes are slightly different than the items included in the Hillingdon Hospital Baby Boxes or the Sandwell and West Birmingham Hospital Baby Boxes. We believe that enabling hospitals to have input over content selection is significant, as is empowering them to create an exterior design which reflects their special community of patients.

The Dad Network 5 Septmeber 2016

In local English media coverage, a strong emphasis is placed on improving  infant mortality – higher than average infant mortality in the areas taking part is a common theme.

It’s not clear whether the other sites are also getting their supplies for free at this stage, but it seems plausible that the company is investing in all these trials as a step towards increasing its presence in the UK. The annual report for Sandwell Child Death Overview Panel describes Sandwell as having been chosen by the company as one of its “starter sites in the UK” and other articles suggest an intention to expand the business here considerably (as does the very long list of English NHS accounts it follows on Twitter).

Jay Hemingway, manager for UK Baby Box company said: “The idea is that every child has the same start in life and we want the boxes to be in every NHS hospital by this time next year.”

Edmonton Gazette and Advertiser 8 September 2016

With plans to extend the initiative throughout the UK, Ms Clary said the diversity of the country’s population would be taken into account. ‘It won’t be a standardised Baby Box that’s the same across the UK”.

Independent Nurse 1 July 2016

There are around 770,000 births every year in the UK.  The wholesale cost of boxes isn’t easily found, but the Scottish Government’s basis for budgeting for its boxes seems to be around £100 per box, so complete adoption implies a total new annual commitment of around £75m for the NHS across the different parts of the UK. However, there might be off-setting savings, as argued here by Scott Johnston, Imperial College Healthcare NHS Trust head of midwifery:

When asked whether the cost of the scheme may be prohibitive for some Trusts, Mr Johnston said this was not a concern; the programme may actually bring benefits by engaging more parents with services early, thus saving costs later on. He added: ‘I think it’s more about the logistics. Within our service we have about 750 births per month, so actually storing [the boxes] and distributing them can be a bit of a challenge. But I can say, as head of service, it’s definitely worth it. It’s something we’ll get over.’

Independent Nurse 1 July 2016

Abstract millions don’t mean much: one way of looking at this is that £75m is the equivalent of 2,500 FTE midwives or health visitors (or put another way, every 300 boxes equals one FTE person). If this were an intervention justified on its impact on SIDS (cot death) and governed by the NICE guidelines on cost-effectiveness (for England) it looks as though it would normally be required to be likely to reduce such deaths by a little under 15%.  It’ll presumably be calculations about the opportunity costs (ie what the money would be spent on instead), what the trials reveal about overall benefits to children and parents,  and potential savings, which determine how far trials in England lead to a publicly-funded roll-out there. The decision seems likely to lie with individual local NHS Trusts.

Note: The Baby Box Co

It is impossible to look properly at the rapid growth of interest in baby boxes without recognising the part played by the Baby Box Co. There are other suppliers around the world, but this company has been by far the most active in promoting boxes to public bodies and charities as part of more complex interventions. Most of the rest appear to concentrate on mail order to individuals.

The Baby Box Co. was started in California in 2013-14,  is headquartered in Los Angeles, and since last year is also a UK-registered private limited company (it also has offices in Canada, Singapore, Ukraine and Australia).  In this interview, one of the founders explains how she was inspired by seeing the 2013 BBC story about Finland. There’s a summary of their activity here:

While we sell Baby Boxes direct to consumers as a baby shower gift or new parent present, we focus most on partnerships with hospitals and other institutions such as nonprofits, government agencies and insurance companies. Through a single program with a large institution, we can get our Baby Boxes to thousands of new parents, so it really is the most effective method of distribution for us. We are currently working with 20 states through government agencies, hospitals, insurance companies, tribes, corporations and nonprofits to distribute our boxes. Organizations implementing Baby Box Co. programs include Cook Children’s Hospital System in Texas, Alberta Ministry of Health and Human Services in Canada, Mountain Park Health Centers in Arizona and Queen Charlotte’s & Chelsea Hospital in the UK among many others! We are also working globally with 12 nations on significant intervention programs and ship directly to consumers in 52 countries.

Working Mother 18 April 2016

It has a clear strategy for growth:

…  in the crowded space of parenting-related ventures, Jennifer has also been adept at growing The Baby Box Co., not only by selling the box to top hospitals and medical institutions, but also sealing partnerships with top players in the space, including organizations like Every Mother Counts, Children International, Room to Grow, Baby2Baby, and others…… For 2016, she estimates Baby Box Co.’s earnings to be 4.5 times higher than those of 2015 and she’s looking forward to growing the business more.

Axial Made in the Middle “2016 Growth 100” [undated]

“We are on track to have a million Baby Box units in circulation by the end of 2016,” Clary said. She estimates that the company will have five million units in circulation by the end of 2017.

Business Insider 1 July 2016

I can’t quickly find any information on the company’s most recent turnover or staff numbers, or where they manufacture, but they have evidently had a huge impact.

The Baby Box University is integral to the company’s model. Where there are specific projects, on-line educational material has been developed with each partner area. In the pilots underway using the Baby Box University model, some initial interaction with the online material is needed before the box can be taken up, but the larger aim appears to be to encourage continued engagement with the site, which the company describes as providing access to expert advice and research. The article  here also states that the platform can be used to support interaction between health care workers and parents, and that  free product offers and bonus draws are also included, which may indicate some sort of revenue stream for the company via sponsorship/advertising arrangements with product suppliers (the site can be accessed for free by anyone willing to register). Other articles mention the platform being a way for mothers in an area to communicate with one another.

The Baby Box Company is not directly involved in the Scottish pilots, but is following developments in Scotland.  Between 12-14 January, its Twitter acount followed 5 MSPs and a Scottish local councillor, retweeted this STV story about a closure threat to a Scottish charity providing supplies for families in need, and also retweeted the tweet below saying that the boxes were “shown to reduce cot death” and describing some parties as having a politically-motivated bias against the SG scheme. (Note: a Labour MSP responded to say that Labour supports the baby box initiative, which is correct, but largely missed in coverage and debate; a commitment to a pilot also appeared in the 2016 Scottish Lib Dem manifesto.)Screenshot 2017-01-13 at 08.52.39.png

The Scottish Government is already committed to national implementation  from summer 2017. The contract for that is not yet out to tender. It’ll be a decent-sized contract (for around 60,000 boxes a year: the budget allows for £7m annual cost to be covered by health funding). It’s not clear how many organisations will be capable of bidding – this company is clearly one.

Ireland

A baby box scheme started in partnership with the Baby Box Co in Limerick in September 2016. As in the English pilots,  local health professionals stress the educational element, noting that they have been given the chance by the company to produce their own educational materials, including videos. Further positive comment from local health professionals here. Around 5,000 boxes are expected to be handed out over a year.

There has been a lot of press coverage, much of it in identical terms and so probably reproducing the news release,  all of which brings out the issue of infant mortality (on which it often reads quite similarly to the English coverge).  I wasn’t able to find anything saying whether this was a short-term pilot or a permanent commitment, whether it was being evaluated, or how it was being funded.

Canada

Canada has several baby box programmes, including what seems to be the largest. All are recently started, so research and evaluation isn’t yet available.

Welcome to Parenthood is a scheme covering 1,500 families, closely tied in with a new programme of extra support and mentoring (see here and here) and funded by a $500,000 research grant from  Alberta Human Services, agreed in 2014. The scheme went live this year, and  is led by an assigned  researcher (Karen Benzies, quoted below). The Baby Box Co appears to be the supplier for the boxes.

Mentors must record their interactions in a journal briefly describing time with the parent and baby, to help researchers in their evaluation of the pilot program.

Benzies said the goal of the project is to evaluate the impact of the various support mechanisms on the developmental outcomes of children and the health of mothers and families, in general.

CBC 4 January2016

Benzies isn’t jumping to any conclusions before the investigation is complete and the data is analyzed.

“As a researcher, I’m always skeptical,” she said.  “We need evidence to say that this is the right thing to do to improve outcomes for children and families. The success for us and for society is around healthy parents and healthy children.”

CBC 9 January 2016

The largest scheme anywhere seems to be in Ontario. It covers  those having a baby between 1 August 2016 and 1 August 2017 (here), estimated to be 145,000 women.  News reports here and  here. At the end of this interview,  the scheme is described as “being funded by the province”, but another piece suggests a more complicated fundingstructure:

[It] includes more than 60 agencies across the province, from midwife practices to family resource centres. Non-profits, charities and the organizations themselves will contribute resources, says Jennifer Weber, chief education officer at Baby Box Co., some through in-kind contributions such as transportation services, product storage and more.

TVO 2 August 2016

The article below gives some further useful detail, including that some contents are provided by sponsoring companies:

In addition to The Baby Box Co.’s education department, which assists communities all over the world with Baby Box program development, the Ontario Baby Box program is organized by The Children’s Aid Foundation, The New Moms Project, and The Mary Berglund Community Health Centre Hub. A network of primary health care facilities spread throughout the province are supporting these groups with distribution to ensure the Baby Box program is accessible to all Ontario residents…..

Contents for the Ontario Baby Boxes are still being finalized, but CEO Jennifer Clary has confirmed that Pampers, which provided the diapers and wipes for Alberta’s Welcome to Parenthood Baby Box program, is supporting Ontario families as well. “We are thrilled Pampers is continuing their partnership with The Baby Box Co. and are so grateful for their contribution of diapers and wipes. ….

The Baby Box Co.’s Chief Education Offer Jennifer Weber also expressed excitement over the company’s partnership with Vroom, a Bezos Family Foundation project.

PR Newswire 10 May2016

From the pieces I have found, it’s not clear what evaluation is planned in Ontario and what is expected to happen after the 12 months covered.

The situation in Ontario is complicated by the parallel presence of a separate local company, Baby Box Canada, offering free boxes of items (but the box cannot be slept in), the cost of which is covered by sponsors: see here.

A much smaller initiative (21 boxes) in a remote Ontario community preceded the current larger one: this was reportedly established by local health professionals who contacted the Baby Box Co for assistance, having earlier seen the 2013 BBC report. The goal was “to guide families toward local services and provide parents with basics that are difficult to attain in Ignace” (from TVO article above).

Separately Nunavut, which has a high rate of infant mortality, is also piloting boxes: here. 800 will be handed out, around one year’s worth of births in this very northerly territory. This is a government initiative, but being funded by donations from companies in Ontario.

There’s been some questioning in Canada about how far these schemes have departed from the Finnish original, and are too commodififed and not enough about support: see here.

The long article from which the extract below comes provides a particularly careful summary of the debates around baby boxes. The article quotes a number of those involved in Canadian projects cautioning about the relationship between lower infant mortality and boxes in a contemporary developed countries, while stressing that Canada still contains substantial numbers of disadvantagaged households.

….  The Ontario baby box initiative’s strength, then, is its commitment to community engagement and providing reliable information regarding infant health, particularly concerning safe sleep practices. …..

Benzies [in Alberta] questions the focus on infant mortality in Canada and the efficacy of some attempts at replicating the Finnish program: “We’ve done an amazing job in [providing] neonatal intensive care, reducing mortality,” she says. “Where we need to focus our efforts is morbidity.” That is, the likelihood of disease, illness and injury to infants. Like Clary, Benzies urges parents to carefully research baby box programs that have sprung up in Finland’s wake — albeit, many decades later — and if they choose to participate, go into such programs with the understanding that stashes of baby supplies can’t address the more systemic issues that affect infant health, such as health care access, poverty and infant care education. “They need to understand why people want you to sign up for something and what the expectations are for that.”

….In her practice [in Ignace], Graff says she sees social barriers to infant and maternal health more often than high-risk pregnancies requiring a neonatal intensive unit. Such barriers include housing concerns, low breastfeeding rates and a lack of resources that might be available in larger cities to deal with postpartum depression and other mental illnesses.

The Ontario baby box initiative aims to bridge the gap in some small way, taking the lead from communities – remote towns, new immigrants, young parents – to ensure the most success. “We’re not saying it will cure everything, but the families, they know who they can actually turn to sooner rather than later,” Baby Box Co.’s Jennifer Weber says.

TVO 2 August 2016

United States

In the US, pre-existing government-funded safe sleeping programmes with a strong outreach and education element who were already providing  portable cots (versions of  folding travel cots) have reacted variously to the advent of boxes.

This one in Chicago has added boxes to what it offers, but not wholly replaced their existing ones, saying boxes are “perfect for families that have limited space” and that “both types of beds will be distributed to families, based on the type of bed needed. Transient families likely will receive the lighter weight baby box.”

However a long-established non-profit  organisation which operates across the US, Cribs for Kids National Infant Safe Sleep Initiative is strongly of the view that boxes are a less good option than the folding device it uses. Its unfavourable comparison of boxes to its established bit of kit  is here (with some further comparison made here).

On the claims about the effect of boxes in Finland on infant mortality, it says:Screenshot 2017-01-13 at 22.12.26.png

It also questions claims that the boxes can be used as a bed for up to 8 months (this is included in coverage of the English schemes, for example, including on the Imperial College website), suggesting that 2 to 4 months will be more common,  before babies can no longer use it safely. That looks like an important point to clarify for policymakers motivated by SIDS concerns, because it has implications for how much of the most risky period for cot death is covered,  and for equality campaigners, because the shorter the period for which it is useful, the less practical help it offers parents.

The Baby Box Co. has produced its own comparative summary: here. It’s not a point-by-point rebuttal, so the issues above about research and potential length of use aren’t addressed. It focusses instead on whether the devices used by bodies such as Cribs for Kids are in fact suitable for overnight sleeping, their greater cumbersomeness and higher cost.

Many organizations are transitioning from Pack n’ Play distribution to Baby Boxes. Baby Boxes wholesale for less than 50% of the cost of Pack n’ Plays, thereby allowing non-profits, hospitals, governments, and other institutions to reach double the number of new parents without increasing their program budgets. An extended reach = more lives saved and that is a huge factor in the increasing rise of Baby Boxes’ popularity.

TheBaby Box Co

Another US non-profit organisation, Babies Need Boxes was founded in 2015 and sources its boxes from The Baby Box Co and uses the Baby Box University model.  The Baby Box Co lists partnerships in a number of other locations in the US.

The views of Cribs for Kids deserve the same careful scrutiny as the case put forward by those promoting the use of boxes  – existing schemes could after all be argued to face competition from new arrivals (equally, they may be reacting to perceived pressure to switch suppliers). Even though organisations such as Cribs for Kids are non-profit, it is possible that in some cases the viability of their model of outreach might be reduced if box schemes became very popular, or they might lose funding from public sources.  Also, some of their practical concerns have to be put alongside the successful use of boxes in Finland for decades.

But the strength of Cribs for Kids’ scepticism about boxes, and the detailed way they make their case,  bears including here, not least because it’s the only reference I’ve seen anywhere to anyone doing a literature review about the Finnish case: none of the references I’ve seen to studies/proof/evidence of the boxes’ effects have provided any links or references, and attempts of my own to locate research on the Finnish box scheme also drew a blank. When reports say that the Finnish box scheme” is credited with” reducing deaths, which is a very common phrasing,  they never say who is doing the crediting.

Australia

A number of commercial Australian baby box companies turn up on a search, who are simply selling boxes and their contents on-line. But there are also pilots reported as starting in Victoria (again involving the Baby Box Co.) and Western Australia (involving an Australian charity). Both seem to be targeted on those deemed especially in need, whereas most of the pilot box schemes described above appear not to be targeted in that way. Some further Australian press coverage here.

There has been press interest  there in the claims made about infant mortality. In Queensland, Professor Jeanine Young, a neonatal nurse and midwife who devised the Queensland Health Safe Infant Sleeping guidelines, reportedly “said the company [the Baby Box Co.] was making money by playing on parents’ fears over sudden infant death syndrome (SIDS)” and Fair Trading officers were reported to be investi­gating claims. Prof Young is separately involved with a “Pepi Pod”programme in Queensland targeting particularly high-risk cases for SIDS, which includes use of a plastic box bed from New Zealand (more on the Pepi Pod programme in the New Zealand section below).

“I have a real problem with this,” Prof Young said. “It is not appropriate for this company to be telling people that the boxes help prevent SIDS ­because there is no evidence that this is the case.”

News.co.au 23 September2016

The same piece carries the company’s response:

But when contacted by The ­Courier-Mail, a spokesperson for the company said online education was more important, and blamed the media for reporting the link because it “makes for a simple and palatable feel-good story”.

The spokesperson said: “Baby Boxes distributed thoughtlessly are not a cure for infant mortality.”  The company is calling for Australian hospitals to go into partnership with The Baby Box Company in order to issue the boxes to new mums.

Many of the company’s press quotes emphasise the importance of education, and that the box is not a solution in itself. Its website says carefully and only that the box has “helped” Finland reduce infant mortality:

The Baby Box program has helped Finland achieve one of the world’s lowest infant mortality rates. The initiative, which enables every expecting woman in the country to claim a free Baby Box once she receives prenatal care and parenting information from a healthcare professional, is credited with helping to decrease Finland’s infant mortality rate from 65 deaths for each 1,000 children born in 1938 to 3 deaths per 1,000 births in 2013.

These comments by Jennifer Clary are also typical:

 Q: HOW DO BABY BOXES HELP TO DECREASE INFANT MORTALITY? 

A: I love the Baby Box concept but think the media has a tendency to romanticize and simplify the tradition. Kela, the Finnish social service, should be commended because they established an incredible foundation upon which to build their Baby Box program: every expecting mum in the country has to visit a healthcare facility for prenatal care and basic educational information in order to be eligible for a free Baby Box. This is a fact that frequently gets left out in media coverage, and it’s a shame as this is arguably central to Finland’s statistical success.

It’s not the Baby Box product that decreases infant mortality, it’s the Baby Box program.

What we know is that there are numerous research studies linking increased parenting education to a reduction in infant mortality outcomes, as well as to an increase in breastfeeding, positive nutrition choices, and improvements in maternal mental health. Therefore, my personal philosophy—and our corporate mantra—is to tie Baby Box distribution to parenting education and ongoing community supports to actually have an impact.

The Dad Network 5 September 2016

However, stronger claims about the link with reduced infant mortality have appeared in material from the company. These 2016 slides are credited as copyright to the company and have a reference “BBC Presentation” in the document name. They are titled “A 75 year-old life-saving tradition” and on page 6, after a more general reference to boxes “helping” bring down infant deaths, do also include the sentence, “In Finland, Baby Boxes decreased the infant mortality rate from 65 deaths for each 1,000 children born in 1938 to 3 deaths per 1,000 births in 2013.”

The process by which the nuance Jennifer Clary argues for so strongly gets lost in reporting would be an interesting study, because the experience of reading so many stories from different places in a short space of time brings out that it’s a widespread phenomenon.

New Zealand

New Zealand is a very different case.  Health professionals had been providing safe sleeping spaces (in the form of woven baskets called Wahakura) for new babies in Maori communities since 2006, because of concerns about particularly high infant mortality rates in New Zealand, with deaths concentrated in Maori communities. More recently, the emphasis has been on the use of clear plastic boxes developed in New Zealand, called Pepi Pods.

This document gives a lot of background. Pepi Pods were first used as an emergency response to the 2011 Christchurch earthquakes and then targeted on those at increased risk of accidental suffocation. They are cheaper and easier to supply in large numbers than the Wahakura. Pepi Pod programmes appear to operate under the umbrella  of an organisation called Change for Our Children, which describes itself as a “social innovation company”, which it explains here is a private profit-making company, but where the profits are used for community benefit.  It says:

PSSs are not for all babies. They are a public health response to the higher
risk of sudden infant death for babies who are more vulnerable due to
exposure to smoking, especially in pregnancy, being born before
37 weeks or weighing less than 2500 grams, or in family environments
where use of alcohol and drugs are prevalent. These babies have a
predisposing vulnerability to hypoxic challenges.

… PSS are not free baby beds for poor families. They are a central component of a comprehensive service that needs to be embedded into a SUDI prevention strategy and regional infant health plan. A Pēpi-Pod® service needs a project action group, coordinator, PSSs and bedding packs, referral processes and criteria, agencies and distributors authorised to distribute, a thorough recipient briefing, follow-up of and feedback from users, and systems for recording, monitoring, communicating etc.

Change for Children NewZealand FAQs

There is some actual research available from New Zealand.

On use after earthquakes:

  • Cowan S, Bennett S, Clarke J, Pease A. An evaluation of portable sleeping spaces for babies following the Christchurch earthquake of February 2011. J Paediatr Child Health. 2013 May;49(5):364-8. doi: 10.1111/jpc.12196.Epub 2013 Apr11.

On use in high risk communities:

  • Young, Jeanine, Craigie, Leanne, Hine, Helen, Kosiak, Machelle. Trial of an innovative Safe Infant Sleep Enabler—The Pepi-Pod. Citation: Women & Birth, 02 October 2013, vol./is. 26/(0-0), 18715192

On the relationship with overall falls in infant mortality:

  • Mitchell, Edwin A. ; Cowan, Stephanie ; Tipene‐Leach, David. The recent fall in postperinatal mortality in New Zealand and the Safe Sleep programme:  Acta Paediatrica, November 2016, Vol.105(11), pp.1312-1320

This last concludes that

The recent fall in postperinatal mortality has not happened by chance. It is likely that the components of end-stage prevention strategy, a focus on preventing accidental suffocation, the education ‘blitz’, the targeted supply of ISSDs [infant safe sleep devices] and strengthened health policy, have all contributed to varying degrees.

Change for Our Children has elsewhere said cautiously, “no claims can be made of cause and effect but the statistics are encouraging.”

Up to now, funding of Pepi Pod schemes appears to have been cobbled together locally from a number of public and private sources. However, in August last year the decision was made to make national funding available for a safe sleeping programme including Pepi Pods. The reporting isn’t clear, but this petition suggests the national programme is intended to continue a targeted approach, and is not an all-population approach. The move to national funding has involved the Minister over-ruling advice from officials, who are reported to have cited concerns about insufficient evidence and possible safety concerns. Both these points have been strongly challenged by Prof Mitchell (author of one of the articles above), a meeting with whom is reported to have been important in persuading the Minister to make  funding available. Some articles  herehere,  here and here.

In August, the likely annual cost was estimated at around NZ$1.5m (about £900,000): the emphasis in New Zealand is on the pod as a sleeping space, so the extra cost of other items may not  be relevant.

South Africa and elsewhere

The 2016 BBC story above refers to other projects, including one in South Africa using clear plastic boxes. However, these are used as baths rather than beds.

Back to the source: Finland

For those interested, there’s an English language Finnish government website which has details about their box, the wider support schemes in which it sits,  the obligations to engage with services which are required of parents in order to receive the box, and even quite a lot of detail about its tendering process: all here.  It’s referred to as the”maternity package” in what comes over as a deliberate move to reduce emphasis on the object in its own right. There’s a (non-official)  video about the Finnish baby box at the foot of this piece as well.

There was a bit of discussion last week on Twitter between Suzanne Zeedyk and Elizabeth Jarman, who are interested in child development and children’s environments,  on the design of the box, both arguing that busy patterns inside the sleeping space should be avoided. Interestingly, one commercial supplier based in the UK points out that the Finnish box has no pictures on the inside (see under “Is it safe?”), although it presents that more as a safety point.

In line with guidance from the Finnish manufacturers of the baby box (the same supplier as to the Finnish Government’s Baby Box Maternity Kit) we have not printed the inside of our box. The manufacturers of the box are chosen through a rigourous safety process, and do not print on the inside of the box.

Little Un Baby Box

The Scottish boxes used in the current pilots follow instead the Baby Box Co. model of  internal decoration.

Reflections and conclusions

This is an astonishing story. It is only three and a half years since the BBC’s original article. In that time, baby box schemes have begun around the world, some of them on a huge scale. Many more commercial companies have emerged than are mentioned here. Huge numbers of boxes have been sold, direct to individuals or to public or third sector organisations. Public health officials in many places have embraced box-based programmes. An enormous number of articles have been published. Someone should do some Google stats on the on-line incidence of references.

The impact of the BBC piece and the extraordinary energy , impact and growth of The Baby Box Co are both remarkable features of this story.  The latter’s belief in what it does is evident, and its ability to pitch what it offers, to form positive relationships with public health officials and to understand what to offer them, is impressive. And absolutely fine. In turn, citizens just need public health officials making decisions about resources to keep their analytic heads, confronted with a whirlwind of enthusiasm, claims and proffered help. Journalists too, maybe. That’s all.

What do we learn that might be useful, as Scotland also goes down this route?

I think we’ve done the right thing funding our own pilot. Although it will carry a cost, it has the advantage of not tying anything learnt into a very specific model run by one provider.  In England, the pilots are at no cost to the NHS, but if they are found to be beneficial, the integration of the Baby Box University into the scheme means that it may be difficult to find another supplier: I’m always wary of the public sector ending up in sole supplier situations. It’s not clear whether the NHS retains the intellectual property for the education materials produced by its staff (I hope so): if not, then it’s easy to see that professionals may be very unhappy to lose access to things they have made with huge local care and enthusiasm, and got used to using, after just a year.

Where we compare less well, firstly, is in having a pilot period that can’t be more than a very few months, given the full national programme is due to start this summer. Lots of the clothes will still have been too big even to try, when the SG is already drawing up the contract for the national box. The peak danger period for SIDS will not be past when the pilots finish. Many parents only emerge in a coherent state to reflect on anything some time after 3 months (if then). Everywhere else doing a pilot is running it for 8 months upwards, which seems to me a better length of time to understand how far the box’s contents are useful, how much and how long it’s been slept in, the SIDS incidence in the cohort (though with only 200, statistically – and happily – there were anyway very unlikely to be any cases) and how it may have helped parents, and aided engagement and education more generally.

We’ve also made a mistake (I think) commiting to a national scheme, regardless of what happens. I understand the arguments that there’s something of symbolic value (I really do) that transcends its measurable public health impact. But the reality is that the health budget (from which this is funded) is under huge pressure: and so are all the others it might be moved to. Right now,  it’s right to demand some more substantial benefits from a long-term commitment to anything which will cost £7m every year. Or more? The £7m budget is for a project starting part-way through the financial year, after all.  It would be preferable to let a one-year national contract and treat it as a much more extensive trial, and build in decent evaluation: if any opposition politicians, or oppositional types, or journalists, or indeed people who like and have defended this scheme, are reading this, if the SG go down that route, please welcome it. It would be a wiser approach to using scarce public funds. They could even contract for two years, so that people go on getting boxes while the evaluation is pulled together and considered. Just build in an easy exit route.

We are also less strong than others on the box as a means  to engaging people with education and human support. The publicity has been very much about “the box”.  Most recently there’s been some reference to there being wider support alongside, but what this means remains vague, compared to other places (Ontario is perhaps the next nearest for vagueness of those mentioned here: but it’s at least using The Baby Box University).

There’s nothing here like the detailed safe sleep programme developed to go with the Pepi Pods or the Alberta mentoring scheme developed to go with the box pilot there.  The wider support part still comes across as an afterthought here: it’s not clear  that the £7m budget assigned includes anything for developing new safe sleeping programmes or other new materials to support new parents.  We’re also unusual in not tying receipt of the box (as far as I can tell) to parents engaging with services in any particular way.

The model which I think emerges best from this is Alberta, with a research based, self-funded pilot, with clear aims, which keeps complete control of the design and delivery of the education and support programme and doesn’t over-stress infant mortality goals, as opposed to broader ones of physical and mental health. And New Zealand has an interesting story to tell about targeted intervention in cases where the risk factors for SIDS are known to be especially high. I think it’s a shame we’ve committed entirely to cardboard boxes, and not piloted Pepi Pods as well.  There’s still time to keep the inside unprinted like the Finns, though.

My final comment is about debating this at all. When people (like me) suggested  when the pilot was launched on 1 January that there were some fair questions to ask, not least  about the claims made about infant mortality, and the decision to commit so completely to this, without more evaluation, when budgets are so tight, one of the reactions was to see this as a depressing further symbol of the schismatic state of Scottish public debate.  People seemed depressed at the inability of everyone simply to come together to celebrate something nice.

This exercise has revealed that anywhere in the world where there’s already been a sharp drop in infant mortality since the Second World War (that includes us) and where these schemes aren’t funded by the supplier, especially where a link  to reduced infant mortality is made,  there’s been a debate. That there’s no controversy in England may be due to the fact that it appears not to be costing the NHS  anything, and the pilots are all relatively small scale (Ireland ditto?).  People in Scotland are raising points made in Canada, the US, Australia and New Zealand. If the NHS in England were, as the Baby Box Co hopes, to commit to a nationwide publicly funded programme while its NHS budget buckles, you know, I think there’d be questions there, too. It’s if we stopped hearing people asking questions about high profile (and not free) public policy choices that seem to come almost out of nowhere, that we’d  have real cause to worry about the unusual state of debate here, I think.

 

 

 

Trying to read the draft Scottish budget for 2017-18 … with difficulty

This is a post about the information provided by the Scottish Government to the Parliament about its spending plans. It’s prompted by having looked in detail so far at just four lines of the draft Scottish Budget, and finding something wrong with the way the numbers are presented in each case. That seems a high hit rate, given this should be one of the exercises over which the Government takes the most trouble.

Sources

There are two relevant sources.

The draft budget document presented to Parliament by the government, which gives figures to “Level 3” (quite detailed, but not the most detailed) alongside a general text, where the government mentions the points  to which it wishes to draw attention. The document is here.

More detailed “Level 4” figures are also provided – they are produced by the Parliament’s research unit, but using information provided by the government, and are available on the Parliament’s website, here.  There’s no general text, but – very usefully – here there is a comment against every line, explaining any change.

Here are the cases I’ve looked at.

Baby boxes

The text of the Budget document says that baby boxes are a priority within the Education portfolio, and within that come from the Children and Families budget. They are not itemised, but are mentioned in the text, under “What the budget does”.

screenshot-2016-12-20-at-12-41-04

Extract from draft budget 2017-18 (under Table6.04)

But search the Level 4 data for Children and Families in the Education spreadsheet for more detail – and there’s nothing more on baby boxes.

To find that requires going instead to the Health section (line 137), where baby boxes are listed as having a new budget of £7m within Miscellaneous Other Board Services and Resource Income.

screenshot-2017-01-04-at-19-12-54

The total for Miscellaneous Other Board Services and Resource Income and for Children and Families in the Level 4 spread sheet matches that given in the draft budget document. That means the text in the draft budget document is out of synch with the numbers it contains.

This means that though the document says that this falls within the Education portfolio (and therefore the  remit of the Education Committee), this is in fact a cost against the Health budget (and sowould normally fall to that Committee). Describing spending as falling within the wrong part of the budget should be a very difficult mistake to make, and then not to spot.

Student Support and Tuition Fee payments

I looked also at the cash funding for student grants and SAAS fee payments, which is given as a single combined figure, in the first line below of Table 6.07 (from the draft budget document). I wanted to know the official explanation for the £26.4m rise and then fall shown. The document text is silent on this.

screenshot-2017-01-02-at-22-58-06The Level 4 document doesn’t break this line down further (and never has done, as far as I know) but would, helpfully, be expected at least to provide a specific comment against the relevant line. However, turning to the Level 4 spreadsheet, the relevant line (141) doesn’t have the same figures. It shows the same (lower) one for both 2016-17 and 2017-18 (only these two years are covered), and so no change needs to be explained. It’s possible that SPICe lifted the 2016-17 figures from what they held last year: but the commentary column at minimum ought to have come from the SG. For consistency with the draft budget document, it ought to pick up the change that shows between the two years.

screenshot-2017-01-02-at-23-06-23

Which version of the numbers gives a true picture of planned spend in this area over the two years? Parliament can’t tell, because the budget document and the Level 4 data are inconsistent.

(The answer is in fact likely to be a third set of numbers – see below – but being able to reconstruct that, more or less, from yet another source, if you happen to know about it, is a poor substitute for having the correct ones in the official budget documentation.)

Cost of student loans

I also wanted to know what the explanation was for the dip and then rise in the cost student loans shown in the Budget document: £175.6/125.6/175.6m – also Table 6.07 above. It’s not reflected in the pattern of the value of loans being issued (£491.3/491.3/560m – also Table 6.07). Again, no explanation is included in the accompanying text.

But again the Level 4 table (above) is no help, as it shows instead a steady £175.6m in each year. Again the budget document and the Level 4 data are inconsistent.  So what’s going on with the lower 2016-17 loan cost figure presented in the budget document to Parliament? It’s impossible to say.

University funding

I also looked at the Level 4 explanation for more on the fall in university current (ie non-capital) funding (“Higher Education Resource”).

screenshot-2016-12-30-at-18-53-06

Here the figures are consistent between the two documents, and the drop is explained as being offset by the move of all postgraduate funding to loans rather than some grant (previously provided via the SFC), and surplus income from the fees charged to rUK students.

This however raises a different internal consistency issue. The extra £26m for student support in 2016-17 discussed above is in fact almost certainly partially accounted for by an in-year  £20m transfer made by the SFC to SAAS (see para 36 here. from the February 2016 SFC funding circular).  In 2016-17, for some reason the Scottish Government decided to under-budget for student support and fee costs: it knew as early as February 2016 (before the budget was passed by the Scottish Parliament) that SAAS would need a large cash injection during 2016-17 to cover its costs in relation to rising student numbers, and planned for this to be taken from the SFC allocation.

The SAAS figures for 2016-17 in the draft budget document appear to be shown after that transfer. However, the SFC university funding figure appears to be shown before it (£1027.2m is  also the figure which was presented to Parliament in the official 2016-17 budget last year: Table 6.06 here). But if that’s the case, the same £20-odd million would be showing in two places for the current year. That would be a substantial error.

This really isn’t good – if money has transferred between one body to another, it needs to be treated consistently. Also, if this is what explains what’s happening in 2016-17, what’s the plan for 2017-18 regarding transfers from the SFC to SAAS – because if a larger one than this year is planned, it means the real terms fall in university funding is larger than shown here.  Also, there was quite a big transfer from the SFC to SAAS in 2015-16 too (para 22 here) – where does that fit in?  Parliament is entitled to greater clarity about all this.

Conclusion

Commentators (myself included) can be pretty critical of the opposition parties at Holyrood for not being more effective. But if the budget document  – one of the single most important things the government puts before the Parliament – contains these sort of inconsistencies, they really are up against it. Even more so, if I haven’t just been especially unlucky, and there’s more of this to be found  in other lines.

I hope the current review being undertaken by the Presiding Officer will look at this as a case study. It suggests at minimum that the government is under-resourcing the production of this material for the Parliament, because nothing here should have survived the sort of systematic, careful cross-checking these documents (and the Parliament) deserve.

So why were student grants cut in Scotland in 2013-14?

It’s the simplest and most unanswered question.

In the summer of 2012, Ministers decided to reduce grant levels substantially for all low-income students – new and continuing – from the following autumn. And we still don’t know why.

It was inevitable this would have one of two effects on students from lower-income families: they would either have to borrow more than before to make up the gap, or make do with less. The extra debt/lost income was non-trivial: around £35m a year of cash support for people from low incomes would be lost, or around 40% of existing means-tested grant.

So this was an odd choice of a target for cuts, for a government which had repeatedly expressed concern that people from low-incomes are put off going to university  by debt, and emphasised its creation in 2013-14 of a “minimum income guarantee” and a much higher universal minimum loan, to meet NUS concerns about immediate hardship.

The decision to visit cuts on grants has never been explained (it’s also unclear why the SAAS budget took a much larger hit than the SG budget as a whole that year: see here). Instead, the SG’s discussion of the 2013-14 changes has concentrated entirely on the way these provided more money up-front (using extra loan). But topping up the package with more loan doesn’t logically imply cutting grant at the same time (see Footnote 1 for a red herring that sometimes turns up here).

At the same time as cutting grant, the SG continued to spend many more times its grant budget on fee subsidies, channelled partly through SAAS and partly direct to universities by the SFC. Within the SAAS budget alone, fee subsidies account for between two and three times the grant budget. All this spending has been sheltered from cuts.  It’s not a matter of opinion that spending on fee subsidies has risen, at the same time as spending on grants has been heavily reduced. It’s just a statement of fact: see Table A1 here.

What might explain these choices? It’s hard to see the policy  commitment to free tuition as irrelevant.  In a context where savings were being demanded from the SAAS cash budget, free tuition policy put a protective barrier round spending on fees. That left grants as the only available shock-absorber.

Recently, a comment in the annual SAAS statistical report was spotted by Patrick Harvie MSP, which made explicit the link between protecting fees and cutting grants (here: emphasis added below).

“The types and value of support students received changed substantially from 2012-13, within the aim of protecting free tuition. For example, the total amount of support provided in bursaries and grants reduced by over a third, offset by a substantial (61 per cent) increase in authorisations for student loans.”

That seemed like a straightforward admission.

However, when the SAAS wording was quoted in a PQ a couple of weeks later, the Scottish Government responded that it was, in fact, wrong and revealed that it had now asked SAAS to amend the report, to remove the words shown above in bold.

This got some press coverage (here), which then  attracted the attention of the UK Statistics Authority (here).  In response to this, the SG’s Chief Statistican has today released a statement (I don’t have a link, just a screenshot I was kindly sent).

screenshot-2016-12-21-at-20-51-36

This explains that the reference to a link between  free tuition policy and the 2013-14 changes, including – as seen above – explicitly the cuts to grants,  has been deemed inaccurate because it “was not consistent with previously published policy statements”. How well these policy statements themselves did or did not reflect the actual decision-making process is left to one side (interestingly, the original SAAS wording turns out to date from relatively soon after the change – it is not just a recent piece of revisionism: see footnote 2 below).

In any event, previous SG policy statements about the 2013 changes have carefully avoided acknowledging that the grant cuts even took place, let alone provided a rationale for them. On this particular point, therefore, there is no previous policy statement to be inconsistent with.  We have just lost the nearest thing we have ever had to an official explanation of why spending on student bursaries for the least well-off was selected to take a £35m hit in 2013.  It’s an odd thing for a government never to have to explain.

 

Footnote 1: a red herring

In the past, bits of the cash budget were sometimes sacrificed/traded with Whitehall  for additional student loan, to stretch resources (33p of cash used to “buy”£1 student loan). But with Scotland entitled to consequentials from the massive amounts of student loan being released in England from 2012-13 onwards, it’s hard to see how the Scottish Government can have needed to “buy” even more.  Even with the recent rise in the use of loans here, we still only use about half the amount of loan England does,  pro rata.  Also, even if the SG were trading cash for loan, it still doesn’t follow that the cash had to be swiped from the poorest students – not least given that quite a bit of the extra loan was for those further up the income scale.

Footnote 2: a curiosity

The Chief Statistician’s statement helpfully adds that the offending wording was also in the  two previous SAAS reports, issued in autumn 2014 and 2015.  I’m kicking myself I didn’t spot it in either of those, and it means that when I have said in the past that the fact of grant cuts hasn’t been admitted by the SG, I have been wrong. SAAS is  an SG agency and it has been stating  that the cuts happened for the past 2 years. The curiosity is that it took two years for anyone government to be moved to address what is now described as an  error substantial enough to be worth post-publication correction, in the government’s principal annual publication on student funding.

Footnote 3: a theory

Just looking at what was happening in the total post-school budget in 2013-14, it stands out that as SAAS took a large hit (budgeted as around £24m, though actual spending on grants fell by more), funding for universities rose by around £40m: Table 5.06 here. In the period after the 2011 election, the Scottish Government appeared especially keen to foster good relations with the universities,  to reduce the likelihood of criticism of its policies, and to reassure them that compared to universities south of the border they wouldn’t be disadvantaged by free tuition, as fees rose sharply in England in 2012. So if – let’s say – there was a political desire to use 2013-14 budget to show university funding getting a large boost, but an instruction from the centre that this had to be found within the portfolio, and given that the large amount devoted to fee subsidies was untouchable as a matter of policy, and – uniquely in the portfolio  – cuts to grants could be back-filled by loan, of which the SG probably had more than it knew what to do with …..  you can begin to see how student grants might become vulnerable – as long as you weren’t particularly concerned about the resulting extra debt/lost income for those from low incomes.  This may not be the explanation for what happened, but in the absence of any other being offered, it’s the most plausible I can construct.

 

 

Trends in student grant across the UK, 2012-13 to 2015-16: England’s last hurrah, and Scotland still a cause for worry

This post uses data published yesterday to look at recent trends in student maintenance grants across the UK – claimant numbers, percentage of students getting one, total spending, and average amounts.

Grants will never attract the same attention as fees, but these numbers tell us about changes in cash support affecting students from the lowest income families, and also, with caveats, something about trends in the recruitment of this group.  Comparing the situation across the UK brings out how different policies can have substantially different effects. How far Scotland is a guide to what lies ahead for England is a hanging question here.

A piece on here in October drew attention to the surprising fall, of almost 10%, in the number of low-income Scottish-domiciled students claiming a means-tested maintenance grant since 2012-13, concurrent with large cuts to student grants which took effect in 2013.

Figures published yesterday by the Student Loans Company for the other UK nations now allow the trends in Scotland to be looked at in a wider context. For consistency with the earlier post, 2012-13 to 2015-16 is kept as the period of interest.

In summary, things still  don’t look any better for Scotland. There have been upwards trends in lower-income students in Wales and England since 2012-13 and Northern Ireland has seen only a small drop. However, the relatively positive picture in England is a swan song: grants have been abolished there for new entrants from this autumn, with consequences still to be seen.

Important note: these figures are sensitive to rule changes on qualifying thresholds.   Wales and Northern Ireland saw no change to their thresholds over the period, or the run-up. In 2012-13, England reduced the upper income cut-off from around £50,000 to £42,620: that took a few years to work through the system, taking out a group of middling-high income households. Grant rates were cut by  around one-third in Scotland in 2013-14  for all students (not just new entrants), but the upper income cut-off stayed round £34,000:the cut-offf or maximum grant fell from just over £19,000 to £17,000.

The tables underpinning these graphs are all here:grant-tables-dec-2016.

Numbers claiming

The population of 18 years olds is falling in every UK nation, but the number of students entering full-time HE has still risen everywhere. The graph shows the change in total claimant numbers relative to 2012-13:  the absolute numbers are so much higher in England than in the devolved nations that the raw figures cannot sensibly all be put into the same graph.

screenshot-2016-12-02-at-10-35-26

The slight English fall in total grant claimants (-1.2%) reflects the tightening of the rules on qualifying levels of income. However, it conceals an increase of 5.5% in the number of those claiming the maximum grant (incomes up to £25,000). This was faster than for students as a whole (+3.6%), which looks like good news for access.

The similar overall fall in Northern Irish grant claimants  (-0.75%) can’t so obviously be explained by changes in the grant rules. It has happened at the same time as a 4.0% increase overall in numbers: this looks like less good news for improving access.

The rise in Wales also can’t be easily attributed to changes in the grant rules. It reflects instead a general growth in student numbers, although in contrast to Engand the rise has been slower among grant claimants (+3.5%) than among students as a whole (+8.3%).  The rise has been faster for those on the maximum grant (+4.4%, up to £18,300) than for those on incomes between that and £50,000 (+2.5%). There may be some purely technical reason for the particularly sharp increase in the numbers with incomes too high to claim a grant, but that’s not immediately evident from just looking at the grant system.

Only Scotland has seen a sharp drop (-9.5%) in grant claimants over the period, despite the cut-off point for receipt of any grant remaining the same, and all-income student numbers from Scotland rising at a similar rate as in England and Northern Ireland (+3.5%). This continues to look like a concerning (and now we can also say anomalous)  pattern.

The Scottish data doesn’t allow a breakdown by grant level, except in the final two years: between 2014-15 and 2015-16, there was  a fall in those on full grants (-4.6%, incomes up to £16,999) and on partial grants (-6.5%, up to £33,999).

Proportion of students getting a grant

The effect of different upper income cut offs on how many students benefit from a grant shows clearly in this graph.  England’s and Northern Ireland’s cut-off points are very similar  (around £42,000),  Wales has the highest (£50,000) and Scotland the lowest (£34,000).

screenshot-2016-12-02-at-10-51-56

As long as the upper income threshold stays the same over time, changes in the percentage of students who get a grant within a nation should be a rough measure of changes in the compositon of the student body by income.

The graph shows that the proportion receiving a grant is not only lowest in Scotland but also that the percentage of the student body receiving a grant has fallen most here.

The percentage of students getting a grant has also dropped slightly in other places. But only in Northern Ireland is this due to an absolute fall in the number getting a grant.  In England, as seen already, this is because the rules have been tightened at higher incomes, while in Wales, again as already seen, there’s been growth but not as quickly as at  higher incomes.

Spending on grant

This is shown by comparison with 2012-13, as again the absolute number is so much higher for England that a single graph doesn’t work.

The largest rise has been in England, despite the reduction in the threshold, reflecting the growth  in numbers at lower incomes. Spending in Wales has risen, but not by so much, reflecting the less rapid growth in low-income numbers. Northern Ireland has held roughly steady.

screenshot-2016-12-02-at-11-04-34

Scotland unsurprisingly stands out: the effect of the very different policy choice made here to withdraw substantially from targeted support for low-income students becomes very clear.

Average grant paid

Again, the figures here show how Scotland has the lowest grants, and the effect of the  cut in 2013-14. In Northern Ireland average payments have held pretty steady. Wales has seen a slight rise. England’s steady and more substantial rise reflects how the composition of the grant-taking group has shifted towards those entitled to a full grant.

 

screenshot-2016-12-02-at-11-22-40

Of course, just as England was developing a positive story to tell on grant, and very probably the particularly strong recruitment of students from the lowest incomes, it pulled the plug. Policy-makers in England who want to tell a good story on widening access must be hoping that Scotland’s rapidly-falling numbers on low-income grants since 2012-13 are not a sign of what lies ahead for them.

 

 

 

 

Average student borrowing across the UK in 2015-16: not as different as you might expect

Yesterday the Student Loans Company published its annual student funding statistics for England, Northern Ireland and Wales. The Scottish figures were published in October. Links to all the data are at the foot of this post.

The average annual loan taken out in each nation in 2015-16 is shown below. The figures cover  borrowing for maintenance and, excluding Scotland, for fees.  The relatively narrow gap between Scotland and the other  devolved nations reflects the relatively low levels of maintenance grants here, and therefore higher dependency on loans to fund living costs.

The average will understate borrowing at low-incomes in Scotland, where low-income students borrow more in the absence of access to as much grant,  but will overstate borrowing levels at low incomes in the other nations, especially high-grant Wales.

The SLC figures do not show borrowing by income, but the Scottish ones do, so they are also included here.

screenshot-2016-12-02-at-04-57-00

There are more non-borrowers in Scotland, mainly from higher incomes: around 30% of Scottish students did not borrow in 2015-16, compared to fewer than 5% in the other nations. So the average across all students, including those who borrow nothing, would show larger differences between Scotland and the rest  – but would also be an even more unreliable  guide to the relative position of those at lower incomes.

Scottish degree students tend to study for a year longer. These annual figures bring out that over the course of a degree many students, particularly those from low incomes, are likely in practice to emerge with similar or more debt in Scotland, compared to Northern Ireland or Wales.

Source: Table 4D for England, Northern Ireland and Wales. Table A6 for Scotland.

Note: maintenance grants have been abolished in England for new entrants from this autumn. These figures pre-date that change.

 

 

Small untruths matter too

This is a post about government news releases, an occcasional topic on this site.

Earlier today the Scottish Government put out a news release about some pilot childcare projects: here. It’s sensible, worthwhile government business.

The ministerial quote included the sentence “As highlighted in research from Heriot-Watt University published yesterday…”    .  The main text added further down:

Research produced by Heriot-Watt University for the Joseph Rowntree Foundation (14 November) identified that “reinforcing and extending the improved provision for good quality, flexible, subsidised childcare across the working year” is one of the “most significant measures” at tackling poverty in the UK…

This was a reference to some research the BBC had covered the day before: here.

It caught my eye because yesterday someone I follow on Twitter had been trying – quite hard – to find this research, which wasn’t linked to the BBC report.  By some collective effort, this report from 18 August this year was tracked down, which includes the quote above.

The Scottish Government had clearly seen the full research, as the quote above is not included in the BBC report. They may even have stimulated the press interest in it, ahead of today’s news release, explaining why the BBC suddenly covered it yesterday. Nothing wrong with that: it is respectable relevant research, which appears to support what they are trying to do.

But what was with using the official voice of government to make out that it had only come out the day before, not three months ago, and doing so not once, but twice?

It’s hardly the largest crime ever committed against truth. But in a small way, it signals a casual attitude towards factual accuracy in Scottish government news releases, just when defending the line between what’s true and what is not seems as important as it has ever been.

And that’s why I’ve recorded it here. Just to notice. Because noticing the small things is always the first line of defence for the bigger ones.

 

 

 

The mystery of Scotland’s disappearing low income students

The Scottish Government published its annual student support statistics yesterday, covering 2015-16. Link here: http://www.saas.gov.uk/_forms/statistics_1516.pdf.

Some things were predictable: the highest loans are still being taken out by those from the lowest incomes (Table A6), reflecting the limited amount of non-repayable grant (bursary) now available to these students.

More unexpected was the further fall in the number of students receiving a means-tested grant, either Young Students Bursary or Independent Students Bursary.

The table below shows how numbers on income-related maintenance grants have changed since 2012-13. In 2013, grant levels were reduced and means-tested grants were substantially restructured, from four schemes to two. However, the old schemes were rolled into the new ones, so it makes sense to include that year as a starting point for comparing effects.

2012-13 2013-14 2014-15 2015-16
All 51,515
YSB 33,150 32,310 30,480
ISB 17,400 16,985 16,135
Total 51,515 50,550 49,295 46,615
Year-year change -1.87% -2.48% -5.44%
Change over period -9.51%

Correction: this post originally had an error in the bottom line of the table, repeated in the  text, which gave -10.51% as the total change.  That is now corrected. Other figures for year-on-year change are right.

A 9.5% drop in grant recipients over three years is pretty remarkable and certainly wasn’t predicted by the government when it launched the 2013 reforms. Since 2012-13 the number of Scottish domiciled students supported by SAAS has increased by 3.5% (Table  A2).

The drop in numbers receiving income-related maintenance grant has been especially large this year.

The SG may be aware of an essentially technical explanation for this downwards trend, but if so it is not sharing it.  Asked about these figures yesterday, the Minister for Further and Higher Education reportedly only “replied that 126,000 full time students were receiving support from the government”, leaving the specific fall in those from lower incomes unacknowledged.

Assuming this is a real effect, in the absence of any explanation otherwise, logically one or more of the following must apply:

  • low-income students are becoming less likely to be recruited relative to others; and/or
  • they are becoming less likely to declare their low-income status; and/or
  • they are becoming less likely to claim grant.

Other data has shown increases in the numbers entering HE from more disadvantaged postcodes (SIMD 1 and SIMD2, the most deprived 40% of areas and target of access policy). So if it were the first point above, that would suggest that it’s possible to increase entry from SIMD 1 and 2, while still reducing those in the system from low incomes.  That seems possible, as the link between low-income and SIMD is far from cast-iron.  So it is possible that as widening access policy concentrates on SIMD 1 and 2,  some of those benefitting are not from households with  low enough incomes to receive a grant, while students from low income families in SIMDs 3 to 5 are now not doing so well relative to others.  That’s no more than a theory – but one suggesting these figures are worth some attention from the Commissioner for Widening Access, whose appointment is due any time now.

If the second two points are relevant, then it’s possible that the amounts of cash support available at low incomes have now fallen so low that some low-income students don’t think it’s worth asking their families to go through the means-test to get them. Bear in mind that the grant at incomes between £24,000 and £33,999 is £500 (for the year).  Those students averse to taking out any debt  (around 20-25% of those at low incomes) won’t benefit from the higher loan they could get by submitting income details: maybe some don’t see the grant alone as worth all the complication.  The fall in bursary recipients has – interestingly – been steepest in the £24,000 to £33,999 band; it’s been next-sharpest in the nil income group (mainly mature students entitled to £875 pa).

So the SG may have inadvertently set up a live experiment into how low non-repayable means-tested student support has to go, before claimants are put off applying. If that’s part of the explanation for these numbers, then that figure seems to be somewhere around £1,000.

The possibility that the SG has managed to cut grants so hard that some people are put off claiming them should certainly be on the agenda of the review of student funding announced today. More generally, what’s driving the fall in YSB and ISB numbers since the system was reformed should be of central interest  – however uncomfortable that may be for the Scottish Government or for NUS Scotland, which strongly supported the changes.

As further context, no other part of the UK has seen this steady fall in grant recipients since 2012-13 (Wales has seen a steady rise, NI and England have been more up and down – but not so down overall: see here) [Update: more on UK comparisons, following the publication of updated data here.]

It was already known that a pretty substantial minority of low-income Scottish students rejected the SG/NUS assumption that they would happily borrow to make up grant cuts (see here). Today’s figures raise the further possibility that if means-tested grants are reduced to a low enough level, even those may be rejected. It may turn out that a low-grant/high loan package simply does not work  at all, in any of its elements, for some of those who need it most. At the very least,  the hyping in 2012 of a new of “minimum income” which would benefit all low-income students looks increasingly to have been based on a shoogly set of assumptions about how quite a few of its target audience would respond.  Let’s hope the new review can do better.

How good is SIMD as a basis for setting HE access targets?

Note: updated on 9 Sept for reference to Access Commission recommendation about use of area deprivation measures.

 

In March this year, the Scottish Government’s Widening Access Commission recommended that (emphasis added):

By 2030, students from the 20% most deprived backgrounds should represent 20% of entrants to higher education. Equality of access should be seen in both the college sector and the university sector.

This has undergone a subtle but significant transformation in the hands of the Scottish Government. Yesterday it confirmed in its Plan for Scotland  that (emphasis added again):

We have set the Government and our universities, along with the wider education system, the challenge of ensuring that by 2030, 20% of university entrants are drawn from the 20% most deprived communities.

I originally thought this move from “background” to “communities” could be traced back to the SNP Manifesto, but in fact the recommendation to use a measure of area deprivation was in the Commission’s report (at page 67), although with a caveat about it being less appropriate for universities in north-east Scotland.

The Commission’s target left open that deprivation might be defined in relatively personal terms – for example family income or employment status, or (getting to the heart of how disadvantage functions down the generations) having no family history of higher education, especially at university level. I’ve spent much of the summer reading research about access to higher education: family background recurs as one of the single most significant factors influencing young people’s decisions.

The Scottish Government’s alternative approach of “deprived communities” keeps things instead within the established practice of using the Scottish Index of Multiple Deprivation (SIMD) classifications as the basis for measuring progress on access. SIMD is already mildly controversial in the HE sector, with the argument running that area measures are too detached from individual circumstances, and in particular are not so good at picking up deprivation in rural areas (although the Scottish Government has tried to answer this).

The release of detailed SIMD data last week allows some further testing of the relationship between SIMD classification and levels of participation in higher education, by comparing areas’ general SIMD ranking with their detailed data on entry to university. This is still  looking at areas rather than individuals, but it is still useful as a way of identifying how well the portmanteau of measures which make up SIMD specifically predict low levels of entry into university within an area.

The answer is that there’s a clear link, but very many exceptions.

Each dot on the graph below represents one of the almost 7,000 small “datazones” into which Scotland is divided for the purpose of SIMD.  How far a dot is along the bottom line shows where that datazone is ranked in SIMD terms. Those to the far left are the most deprived, those to the far right are the least deprived. How high up a dot is shows what proportion of young people in the area went straight university (there’s more detail on the HE measure used in the post below). The pale horizontal line is the mid-way cut-off: dots below it are areas in the bottom 50% for HE entry, those above are in the top 50%.

scatterplot-simd-hesa

It’s immediately clear that while there are relatively few areas with above-average university entry in more deprived areas (though there are some, including some well above), there are many areas with relatively low entry rates towards the right-hand side of the graph. In some very high SIMD ranked areas, HE entry rates are at or close to zero. This looks like evidence that there’s a substantial presence of households less confident or well-placed to get involved in university education right across the SIMD spectrum.

The table below puts some numbers on the relationship between SIMD ranking and  university entry.  SIMD is divided into quintiles, as used by the SG in the contest of access. SIMD 1 and SIMD 2 make up the bottom 40% of areas: these are a particular focus for widening access.  University entry rate is divided into quarters: 1 is the bottom quarter of areas by entry rate, 2 the next up, 3 the next, and 4 the top quarter.

The table shows that in SIMD 1, over 90% of areas have below average entry (62.8% plus 29.8%). In SIMD 2, however, one-quarter of areas (20.3% plus 5.3%) are above average.

Even more strikingly,  1,157 areas in SIMD 3 to 5 have below average levels of university entry. That means one-third of all the areas with below average entry rates are in these higher quintiles.

table-simd-hesa

This analysis reinforces the arguments against linking access targets too closely to SIMD.  Universities Scotland came out this week committing to lower entry requirements for some students, stressing the need for “case by case” judgements. If the SG target is for deprived areas however, the pressure will be greater to use that as the basis for reduced offers. The analysis above suggests that that would lead to some extreme rough justice.

The Scottish Parliament Education Committee met for the first time today and spent some time looking at widening access. The Official Report is not yet available, so I don’t know if this shift in language and its potential implications were picked up. But these figures suggest that the practical effect of a high-profile access target which is area-based could yet come back to bite MSPs hard in their constituencies.  The time to ask some searching questions about what this change in wording will mean would be now.

Footnote:

The Access Commission’s target refers to “higher education” rather than university. However, its remit was specifically to widen access to university, and all the staging-post targets it suggested are specifically for  university entry (also, more than 20% of college entrants already come from SIMD1). So the SNP Manifesto and SG target’s reversion to “university” makes sense.

A picture of education inequality

 

Here, thanks to the underlying data provided as part of last week’s new version of SIMD, is one picture of what education inequality looks like in modern Scotland.

entry-rate-17-21-simd

The graph above takes the 6,976 small areas (“datazones”)  into which Scotland is divided to calculate SIMD scores, but looks only at the single indicator which measures the proportion of all those aged 17 to 21 in each datazone who started in full-time degree-level study. The figures are available for all but three datazones.

The data used are available here (under “indicator data”) and cover the period 2012/13 to 2014/15.  The full description of the indicator is at page 48 here.

Interestingly the SG explains that “study at degree level has been chosen as this level provides the highest gains in future earning potential and reduces double counting of students that progress from HND to degree”.  This means that in its own measure of disadvantage, the SG chooses to exclude college-level HE (which it often elides with university-level study) not just on technical grounds, but also on grounds of lack of equivalence. Worth noting in passing.

What does this graph tell us?  Most obviously, that the position is generally skewed towards lower levels of entry.

The thick black vertical line marks the median value – that is, the mid-point of the distribution, where one-half of areas are to the left, and the other half to the right. It has a value of 7% (0.07/1). The dotted line marks the point at which three-quarters of all areas are to the left, and just one-quarter are to the right: its value is 11.3%. At the green line, just one-in-ten areas are to the right: it has a value of 16%.

So this graph shows how unequally distributed early, direct entry to university in Scotland remains by area (it tells us nothing about individuals’ circumstances).  The 10% of areas with the highest values sit between 16% and 43% (excluding an outlier at 63%).  Meanwhile, half the country lies at 7% or below, and three-quarters is below 11.3%.  The bottom quarter (not marked) all lie at 4% or below.

You could argue, rightly, that chopping up Scotland along different lines, or using a smaller number of larger areas, would produce a slightly different result (using larger areas ought to reduce the number with very low or high results, for example). But the broad picture?  It wouldn’t change much.

Yesterday the Scottish Government confirmed in its Plan for Scotland  that:

We have set the Government and our universities, along with the wider education system, the challenge of ensuring that by 2030, 20% of university entrants are drawn from the 20% most deprived communities.

This graph is one illustration of quite how big a challenge that represents.

How far “deprived communities” as measured by SIMD as a whole map on to those with the lowest direct young entry rates to university can also be worked out from this data – a post on that may follow.

Technical note

The full technical definition of the indicator is at  page 48 here.
NB These percentages are unique to this measure, and not comparable with ones quoted elsewhere. For example, they are not an “age participation rate”. They record instead that in half the areas no more than 7% of all local 17 to 21 year olds started a full-time degree – but some of those 17 to 21 year olds may already have been at university. The value of these numbers lies in allowing us to compare the variation in entry rates across the country.

School leaver destinations: HE still popular in 2014, but less “sticky”

The SG’s annual statistics on school leaver destinations were published yesterday. The information is collected via surveys in September and March and, among other things, provides  a snap shot of how well initial entry into HE holds up for this group after a few months.

Two figures are provided: an initial percentage going into HE and a “follow up” percentage, which catches those who are still there the following March. Very few HE courses can be completed in that time, so the drop between the two figures is likely to reflect those who started a course but decided not to complete it, for whatever reason.

A single “HE” destination figure is used, which combines entry into college to do an HNC/D with entry into university. The figures don’t allow anything to be said about either of those categories separately.  College entry will account for around one-third of the figure.

The figures (see table at foot of post) show:

  • A very similar percentage of school leavers went into some form of HE in 2014 as in the previous year (38.8% vs 39% the previous year)
  • But the percentage still there a few months later was noticeably lower (36.8% vs 38.2% last year)
  • The attrition rate from HE was therefore 5.4% (-1,106, more than double the equivalent number in the previous year), much higher than in the past couple of years, but very similar to 2010/11.
  • The absolute number initially entering some form of HE was the highest for any year shown (and probably for any year), as the total number of school leavers rose.  The number still there a few months later is a little below the previous year.

 

As the table below shows, there was a large rise in those moving into employment between the initial and follow-up surveys – over 3,000 more were in employment  by the time of the follow up.

FE numbers also see a (much larger) fall, so it’s likely that the figures reflect more young people leaving education for employment, for whatever reason – there will be push and pull.

The drop in FE is large enough to mean that the proportion still in either FE or HE by March was last lower in 2010/11, and the 9.3% drop in combined FE/HE participation between initial and follow-up is much the largest in the period covered.

So these figures show that interest in going into HE (and FE) held up in 2014, but that both were considerably less “sticky” than in the previous couple of years, for some reason.

You’d want to be incredibly wary of pinning any causal relationship against that. But simply as a change, it’s interesting to observe. The evidence that in 2014-15 young people were more likely to fall out of post-school education early, compared to the recent past, is therefore no more than a straw in the wind at the moment, but one we probably shouldn’t let blow past entirely unnoticed.

The table below is  Table 2  of the SG statistical release, with some additional calculations added – those are marked in bold.

2010/11 2011/12
Initial Follow Up2 Change Initial Follow Up2 Change
Destination Category
Higher Education (%) 36.3 34.4 -5.2% 37.8 36.1 -4.5%
Higher Education (nos) 19382 18320 -1062 18804 17909 -894
Further Education 27.1 24.6  -9.2% 26.6 24.8  -6.8%
HE & FE 63.4 59.0 -6.9% 64.4 60.9 -5.4%
Training 5.4 3.3 4.5 3.6
Employment 19.2 23.8 24.0% 19.8 23.9 20.7%
Voluntary Work 0.5 0.5 0.4 0.5
Activity Agreement1 0.5 0.6 0.9 0.7
Unemployed seeking 9.5 10.2 8.1 8.1
Unemployed Not Seeking 1.2 1.6 1.3 1.8
Unknown 0.3 0.9 0.4 0.6
Positive Destinations 89.0 87.2 90.1 89.6
Number of Leavers 53,394 53,255 49,745 49,610

 

2012/13 2013/14
Initial Follow Up2 Change Initial Follow Up2 Change
Destination Category
Higher Education (%) 37.1 36.9 -0.5% 39.0 38.2 -2.1%
Higher Education (nos) 19161 19009 -152 20052 19594 -458
Further Education 27.7 24.5  -11.6% 26.3 24.3  -7.6%
HE & FE 64.8 61.4 -5.2% 65.3 62.5 -4.3%
Training 4.8 3.1 4.0 2.5
Employment 20.4 24.6 20.6% 21.7 25.5 17.5%
Voluntary Work 0.5 0.5 0.4 0.4
Activity Agreement1 1.3 0.9 1.0 0.7
Unemployed seeking 6.9 7.6 6.2 6.5
Unemployed Not Seeking 1.1 1.6 1.1 1.5
Unknown 0.3 0.3 0.3 0.3
Positive Destinations 91.7 90.4 92.5 91.7
Number of Leavers 51,647 51,515 51,416 51,293

 

2014/15
Initial Follow Up2 Change
Destination Category
Higher Education (%) 38.8 36.8 -5.2%
Higher Education (nos) 20367 19260 -1106
Further Education 27.6 23.4 -15.2%
HE & FE 66.4 60.2 -9.3%
Training 3.8 2.7
Employment 21.4 27.8 29.9%
Voluntary Work 0.4 0.5
Activity Agreement1 0.9 0.7
Unemployed seeking 5.4 5.7
Unemployed Not Seeking 1.1 1.6
Unknown 0.5 0.6
Positive Destinations 93.0 92.0
Number of Leavers 52,491 52,337
1. In April 2011 the Scottish Government rolled out the use of Activity Agreements.
2. Leavers who moved outwith Scotland, were deceased or who had returned to school between the initial and follow up survey were excluded.

FM: “We must have a debate based on facts” (although they may need to be corrected later)

what I want is, Facts. … Facts alone are wanted in life. Plant nothing else, and root out everything else. .. Stick to Facts, sir! (Thomas Gradgrind, Hard Times: Charles Dickens)

…I simply pointed out what the figures actually say…. I am simply setting out factually for the chamber what the figures actually say. I think that that is the appropriate thing to do…. the numbers from our most deprived communities are up 10 per cent—up 10 per cent for applications and up 10 per cent for entries. That is simply a fact, and it is a fact that is in these figures….instead of arguing over the facts—and we cannot argue over these facts, because they are what they are ... (FM at FMQs Thursday, 9 June 2016)

The louder he talked of his honor, the faster we counted our spoons. (Ralph Waldo Emerson)

The fact the FM was particularly keen to discuss on 9 June – she quoted the figure at three different points  – was that

when we look at the figures for people of all ages we see that the numbers from the most deprived areas who are both applying to university and being accepted are up in 2015 compared with 2014, in both cases by about 10 per cent.

This post (from the day of FMQs) explained why that 10% was not a real increase and the actual figure was likely to be much lower.

This number was taken from a new data release by UCAS. It was held up as more meaningful than the 7% fall between 2014 and 2015 in the number of 18 year olds entering university through UCAS from the most deprived areas, which had emerged  unexpectedly from the same set of figures. The FM described this figure as having “dropped slightly” and as “a slight decline”.

Jump forward to Tuesday 14 June and there is this (government inspired) written parliamentary answer:

Jenny Gilruth (Mid Fife and Glenrothes) (Scottish National Party): To ask the Scottish Government what its position is on the additional data published by the Universities and Colleges Admission Service (UCAS) on 10 June 2016 clarifying that the scope of its undergraduate scheme for providers in Scotland increased in 2015 to include courses previously recruited through the postgraduate UCAS Teacher Training scheme and there has been variability in the recording of very late acceptances from cycle to cycle. (S5W-00737)

Shirley-Anne Somerville:

The Scottish Government welcomes the publication of this additional data by UCAS clarifying the scope of the previously published figures, including those referenced by the First Minister at First Minister’s Questions on 9 June 2016. We also welcome that the figures continue to show that, even accounting for the issues UCAS subsequently clarified, the number of people of all ages accepted through UCAS to Scottish universities from the 20% most deprived areas in Scotland increased from 4,020 in 2014 to 4,075 in 2015, an increase of 1.4%. The increase since 2010 is 17.6%.

Note: the UCAS update increases the reported fall in 18 year olds placed applicants, to 7.5%.

It’s right that the Scottish Government moved quickly to correct the record, but it is odd to suggest that the additional information provided by UCAS on the Friday was new. The SG’s own press office was aware in January 2015 of the issue with the increase between 2014 and 2015, saying then:

Today’s UCAS publication suggests a 10 per cent rise in Scots-domiciled applications. However, much of the rise is due to the inclusion of teacher training courses at Scottish universities in the UCAS undergraduate scheme for the first time this year. The comparable year-on-year figure is a rise of one per cent as per the figures noted above.

UCAS had also highlighted the issues in its end of cycle report in December 2015:

In 2014, there were fewer late acceptances to Scotland recorded in the UCAS data for some Scottish providers, meaning that comparing acceptances with 2014 may not give an accurate measure of change. Also, a large set of teacher training courses at providers in Scotland were recruited through the UCAS Undergraduate scheme for the first time in 2015, having previously been recruited through UCAS Teacher Training. These two factors are estimated to account for around 3,800 of the 4,400 increase in acceptances to providers in Scotland in 2015 compared with 2014.

The Scottish Government’s press notice of 22 December 2015 welcoming the report  noted:

Record number of Scots were accepted to university – 34,775, an increase of over 900 (3 per cent) after taking account of changes to the coverage of UCAS data

In her contribution to FMQs, the FM said:

I have studied the figures in some detail, as people would expect me to have done

For the most senior member of the government, with all the extraordinary and unenviable responsibility that entails, people would surely expect that studying numbers in detail would include taking advice from people within the government machine who know how to read the numbers in question and can flag up any potential mis-reading. The press and therefore presumably also the government received advance copies of the latest UCAS report under embargo.

It has to be assumed from this case either that the system for briefing the First Minister before the most closely watched and widely reported event of the parliamentary week does not operate with the checks and balances that would have alerted the First Minister to the problem with her key “fact” in this case; or else that the collective institutional memory of the SG in this area is less than 6 months old, or resides in so few people that there’s very little to prevent such a large error making it all the way to the FM’s script.  Any of these options have implications well beyond this one issue.

 

On 9 June, the First Minister said,

I am not saying that the figures are wrong. I am simply setting out factually for the chamber what the figures actually say.

when asked to comment on the (now clarified)  7.5% single year fall in the number going straight to university at age 18 from the most deprived 20% areas in Scotland. That remains a bona fide, UCAS confirmed, surprising fact and as such it deserves some further attention.

 

Ironic footnote

In the same FMQ session, the FM was critical of the Labour Education spokesman for releasing set of figures over the weekened which wrongly suggested that the number of women taking science and computing at highers level had fallen. Here’s how the FM approached this, placing further rhetorical emphasis on the concept of “facts” and introducing a further one – “distortion”:

I think that the question is whether Iain Gray did that deliberately, or whether the Labour education spokesman did not know that highers were being reformed. Frankly, I am not sure which would be worse….

She added:

I hope that Labour and the Scottish National Party can be allies on the education agenda, but we must have a debate based on facts, not on distortions.

Let me underline what Labour did at the weekend. It compared the numbers of girls going into STEM subjects in 2007 with the figures for 2015. It took 2007 as the baseline, when young people sat only highers. It then went to 2015 and counted only the old highers; it did not include the new highers or the revised highers that are replacing the old highers. Labour then went to the media on the basis of that information and said that there was a fall in the number of girls studying those science subjects. That was flatly wrong; it was a distortion of the reality. Frankly, it was a disgrace.

If we are going to move forward and build consensus and alliances on improving education for our young people—as I am determined to do—and if Labour wants to be part of that, let us stop the distortion and do that on the basis of facts.