Memorandum submitted by Alissa Goodman
and Luke Sibieta, Institute for Fiscal Studies[1]
1. INTRODUCTION
AND SUMMARY
When Mr Brown was the Chancellor of the Exchequer,
he chose to make a series of high-profile announcements about
education spending in Budget 2006, Pre-Budget Report 2006 and
Budget 2007. Budget 2007 announced the details of the Comprehensive
Spending Review settlement for the then Department for Education
and Skills, and the intended level of education spending for the
whole UK. This note will analyse the implications of this settlement.
Below we present a summary of the key-points.
— UK education spending is set to grow
by an annualised average rate of 2.4% over the three years between
2007-08 and 2010-11, after adjusting for economy-wide inflation.
This represents a slow-down compared with the equivalent growth
rate over Labour's second term (6.0%) and that achieved to date
during its third term (4.9%).
— This is likely to be a tight spending
round, with the overall spending envelope only growing by an annualised
average of 2.0% in real-terms. Thus the settlement is relatively
generous compared with what most other departments have already,
or are likely to receive. An exception to this is the NHS, which
is likely to receive a more generous settlement than education
(as it typically has done under both Labour Governments since
1997 and Conservative Governments since 1979).
— Education spending is projected to
increase by about 0.2% of national income over the course of the
current Parliament. This means that Labour's manifesto pledge
is on course to be met, but by less of a margin than that delivered
over Labour's first-two terms of office.
— The Comprehensive Spending Review
settlement will allow the Government to go further towards meeting
the pledge made by Mr Brown in Budget 2006 to "increase state
school spending per pupil to that currently seen in the private
sector." State-school spending per pupil is set to increase
from £5,270 in 2007-08 to £5,800 in 2010-11 (both 2005-06
prices), leaving a further £2,200 per pupil to be found to
meet the pledge to increase it to £8,000 per pupil (which
was average private school per pupil spending in 2005-06).
— One must remember that meeting this
pledge will not close the contemporaneous gap between spending
per pupil in the state and private sector. In-fact, the latest
figures show that the contemporaneous gap is widening with spending
per pupil in the private sector growing at a faster rate than
that in the state sector
— If state school spending per pupil
grows after 2010-11 at a rate of 3.2% in real-terms (the same
as that over the period covered by the forthcoming Comprehensive
Spending Review, and greater than the 2.4% increase in the overall
education budget due to a combination of resources within the
education budget being targeted at schools and an expected decline
in pupil numbers) then the pledge is unlikely to be met before
2020-21. This would represent a lengthening, rather than a shortening,
in the time it takes to state school spending per pupil to catch
up to the private sector level.
— The way education spending announcements
have been made over recent years has not been ideal. There were
occasions when new education spending plans have been presented
in the form of cumulative totals over a number of years, without
accounting for economy-wide inflation. This something we have
previously described to the Committee as a "highly misleading
presentational device." On occasions it has also proven difficult
to disentangle what elements of announcements represented new
spending and what was a re-iteration of previous announcements.
2. CSR SETTLEMENT
FOR EDUCATION
The forthcoming Comprehensive Spending Review
(CSR), now expected to take place in autumn 2007, will set out
departmental expenditure allocations for 2008-09, 2009-10 and
2010-11. Some departmental allocations have already been announced
(as has the total spending envelope from which funds will be shared
out), including that for education.
We will thus discuss in turn the details of
the education spending settlement and how education spending is
expected to evolve over the period covered by the CSR. This will
be set in the context of how education spending has grown in previous
years and the known or likely settlements for other departments
in this year's CSR, including that of the NHS.
We will also discuss the implications of the CSR
settlement for the then Chancellor Gordon Brown's pledge "to
ensure for 100% of our children the educational support now available
to just 10%."
OVERALL EDUCATION
SPENDING
Total UK Education Spending
In Budget 2007, Mr Brown announced the details
of the CSR settlement for UK education spending; Table 1 below
shows the details of this. It shows that education spending in
the UK will rise from £75.4 billion in 2007-08 to £80.9
billion by 2010-11, after adjusting for economy-wide inflation.
The bottom half of Table 1 shows that this settlement
implies that UK education spending will grow by an annualised
average of 2.4% per year over the three year period between 2007-08
and 2010-11, after adjusting for economy-wide inflation. This
is lower than the equivalent growth rate over Labour's third term
to date (4.9%) and much lower than the equivalent growth rate
over Labour's second term (6.0%). It is also slightly lower than
that seen during Labour's first term (2.8%), during which time
Labour made the decision to keep to Conservative spending plans
set out in Kenneth Clarke's November 1996 Budget for 1997-98 and
1998-99. However, it is still higher than that seen under the
Conservatives between 1979 and 1997 (1.5%).[2]
Table 1
UK EDUCATION SPENDING OVER THE CSR AND PREVIOUS
GROWTH RATES
| Total UK Education Spending
|
| Nominal | Real (2006-07 Prices)
|
| 2005-06 | £67.1bn |
£68.8bn |
| 2006-07 | £71.5bn |
£71.5bn |
| 2007-08 | £77.4bn |
£75.4bn |
| 2008-09 | £81.1bn |
£76.9bn |
| 2009-10 | £84.8bn |
£78.3bn |
| 2010-11 | £90bn | £80.9bn
|
| Average Annual Increase over CSR07 (2007-08—2010-11)
| 2.4% |
| Average Annual Increase over Labour 1 (1996-97—2000-01)
| 2.8% |
| Average Annual Increase over Labour 2 (2000-01—2004-05)
| 6.0% |
| Average Annual Increase over Labour 3 to date (2004-05—2007-08)
| 4.9% |
| Average Annual Increase under Conservatives (1978-79—1996-97)
| 1.5% |
Sources: Budget 2007, HM-Treasury (2007) Public Expenditure
Statistical Analyses 2007 (and previous years for figures before
1988-89) (http://www.hm-treasury.gov.uk/media/E/B/pesa07_complete.pdf)
Figure 1, shown below, shows the annual increases in education
spending, after adjusting for economy-wide inflation, between
1996-97 and 2007-08, compared with those projected to occur during
the period covered by the forthcoming CSR.
It shows that the relatively slow growth in Labour's first
term can largely be accounted for by the slow growth in 1997-98
and 1998-99 during which Labour had made the decision to stick
to Conservative spending plans. Since then, there has been strong
year-on-year growth in education spending, particularly since
2000-01. It is clear that the annual increases over the period
covered by the next CSR represent a slow-down compared with the
recent much larger real-terms increases in education spending.
Figure 1
ANNUAL INCREASES IN EDUCATION SPENDING, AFTER ACCOUNTING
FOR ECONOMY-WIDE INFLATION
Sources: Budget 2007, HM-Treasury (2007) Public Expenditure
Statistical Analyses 2007 (and previous years for figures before
1988-89) (http://www.hm-treasury.gov.uk/media/E/B/pesa07_complete.pdf)
Comparisons with Settlements for Other Departments
It is certainly the case that the CSR settlement for education
implies that education spending will grow at a slower rate (2.4%)
than it has done under previous spending reviews. However, education
spending is planned to increase at a faster rate than overall
spending, with total managed expenditure only growing by an annualised
average of 2.0% between 2007-08 and 2010-11, after adjusting for
economy-wide inflation. Thus, the settlement is relatively generous
when compared with what most other departments are likely to receive.
In-fact, most of the settlements that have already been announced
are relatively tight. For instance, the combined budgets of the
Home Office and the new Ministry of Justice will receive a real-terms
cut of 0.7% on average between them over the period. Nine smaller
departments will receive between them a real-terms cut of 5.0%
on average over the period.
We note that the Committee has previously shown an interest
in relative comparisons between health and education spending.
Naturally, since the settlement for the NHS has not been announced,
we cannot perform an exact comparison. We can instead do a comparison
under the settlements already announced and likely settlements
for those not announced. Before we do so, it is worth pointing
out that historically NHS spending has tended to grow faster than
education spending. For instance, over the eight years covered
by spending reviews to date (1999-00 to 2007-08), NHS spending
is set to grow at an annualised average rate of 7.2%, after accounting
for economy-wide inflation, during which time the equivalent growth
rate in education spending is set to be 5.3%. NHS spending also
grew faster during the period of Conservative Government between
1978-79 and 1996-97, when it grew at an annualised average rate
of 3.0%, after accounting for economy-wide inflation, compared
with 1.5% for education.[3]
The first set of bars in the chart below sets out how the
overall spending envelope will evolve over the period covered
by this CSR and the settlements already announced, as described
above.
Figure 2
POSSIBLE CSR ALLOCATIONS UNDER SPENDING COMMITMENTS MADE
SO FAR
Source: C. Emmerson, "Public Spending", IFS Post-Budget
Briefing 2007 using HMT Figures, (http://www.ifs.org.uk/budgets/budget2007/public_spending.ppt).
We first assume that official overseas development assistance
(ODA) will increase by 11.2% to stay on track to meet the Government's
objective for ODA to account for the UN's recommended level for
developed countries of 0.7% of national income by 2013. We then
assume that social security and debt interest payments will increase
by an annualised average of 2.0% over the period covered by the
CSR, the same as that seen between 2006-07 and 2007-08. If we
further assume that no extra funds are allocated to meeting the
Government's challenging 2010 Child Poverty target and the rest
of non-NHS public expenditure grows by 1% on average in real-terms
(which is in-line with what these departments are receiving between
2006-07 and 2007-08), NHS spending will be able to grow by 3.4%
over the period covered by this CSR. This is the same as the pre
new Labour long-run average increase that the NHS received between
1950-51 and 1996-97.[4]
An alternative scenario is that the NHS receives a real increase
of 4.4% per year, which was the most optimistic scenario for NHS
spending over this period (in terms of lowest cost to the taxpayer)
envisaged by Derek Wanless in his 2002 study for the Treasury
which examined the cost to 2020 of delivering a world-class, free
at the point of use, taxpayer financed, health care system. However
a settlement of 4.4% per year for the NHS would, under the assumptions
underlying the calculations in Figure 2, imply an average real
annual increase of just 0.4% per year for other non-NHS departments.
Either way the settlement for education is likely to be less generous
than that for the NHS, but is on average more generous than what
most other departments have or are likely to receive.
Education spending as a proportion of national income
Labour made manifesto commitments in the 1997, 2001 and 2005
general elections to increase education spending as a proportion
of national income, over the course of each parliament. Figure
3, below, illustrates UK education spending as a proportion of
national income up until 2007-08 (shown by the solid line) together
with the projected share beyond 2007-08, given the CSR settlement
for education and assuming real GDP growth of 2½% per year[5]
(the dashed line).
Over both of Labour's first two terms of office, it is clear
that education spending at the end of the parliament was significantly
higher as a share of national income than it was at the start
of the parliament (although the cuts in education spending as
a share of national income in 1997-98, 1998-99 and 1999-2000 meant
that education spending as a share of national income was lower
on average during new Labour's first term in office than it was
during the Conservative Government's parliament from 1992-93 to
1996-97). In slight contrast, if the next general election were
held in 2009 or 2010, education spending is only likely to have
grown by 0.2% of GDP over the course of the current Parliament.
Figure 3
UK EDUCATION SPENDING AS A PROPORTION OF NATIONAL INCOME
Sources: HM-Treasury (2007) Public Expenditure Statistical
Analyses 2007 (http://www.hm-treasury.gov.uk/media/E/B/pesa07_complete.pdf).
HM-Treasury Website for Deflators and GDP Numbers, http://www.hm-treasury.gov.uk/economic_data_and_tools/gdp_deflators/datagdp_index.cfm
The actual change, and thus whether the pledge is met or
not, will depend on the exact outturn figures for real GDP, inflation
and (cash) education spending over these years. If real GDP growth
were lower than 2½% on average over this period, then (assuming
that inflation was in-line with the Treasury's projections) the
proportion of GDP devoted to education spending would grow by
more than 0.2% over the course of this Parliament.
If real GDP growth were higher than 2½% on average over
this period, then (again assuming that the inflation projections
proved accurate) the change would be less than+0.2%. However,
if real GDP were more than 2½% and proved to be sustainable,
then the public finances should be in a relatively stronger state
than currently envisaged. This could permit the Chancellor of
the day to "top-up" the level of education spending.
For instance, the 1998 CSR settlement implied that education spending
would grow by an annualised average of 5.7% between April 1999
and March 2002, after adjusting for realised economy-wide inflation.[6]
However, the latest outturn shows that it actually grew by 6.6%
over this period. Part of the reason for this is that expenditure
on cyclical social security benefits and debt interest payments
turned out to be lower than had originally been expected, permitting
Mr Brown to sanction higher levels of education spending.
In conclusion Labour's manifesto pledge seems on course to
be met, but by less of a margin than that delivered during Labour's
first two terms of office
SCHOOLS SPENDING
In Budget 2006 the then Chancellor, Mr Brown, stated that
"Our long-term aim should be to ensure for 100% of our children
the educational support now available to just 10%." He clarified
this aim in quantitative terms by pledging to increase spending
per pupil in the state sector to that then being spent per pupil
in the private sector. According to Treasury figures, this meant
increasing funding per pupil in the state sector from around £5,000
per pupil to around £8,000 (the private sector level in 2005-06).
Here we set out the implications of the CSR settlement for the
achievement of this aspiration, followed by a discussion of pledge's
significance.
Progress Towards Achieving Pledge
Figure 4 shows a pie-chart where the full-circle represents
the amount spent by the private sector per head in 2005-06, £8,000.
It is estimated that in 2007-08 state spending per head will be
about £5,270 in real-terms (to the nearest £10). In
Budget 2007, Mr Brown announced that by 2010-11 state school spending
per head would reach £5,800 in real-terms. We estimate that
of this £530 per head increase:[7]
— £80 will come from falling pupil numbers.
— £120 from extra capital spending
— £320 from extra current spending
Figure 4
MATCHING PRIVATE SCHOOL SPENDING?
Source: Budget 2006 and 2007, IFS calculations.
Note: Extra current and capital are those implied by the announcements
made in Budgets 2006 and 2007. The extra amount implied by lower
pupil numbers is the increase in spending per pupil that would
occur if overall spending remained constant in real-terms, but
full-time equivalent pupils declined as projected.
This leaves a remaining real-terms gap of £2,200 in
2010-11 (2005-06 prices). If the gap were to be closed entirely
in 2010-11, it would require about an extra £16.6 billion
in real-terms school spending. Unless the Government decided that
it wanted to close the gap immediately, this does not have to
be raised immediately through taxation or borrowing, so it makes
sense to try to understand how quickly state spending per head
would need to grow in real-terms beyond 2010-11 to meet this aspiration
in specific years, as is shown in Table 2.
Table 2
REQUIRED GROWTH RATE IN STATE SPENDING PER PUPIL TO MEET
THE PLEDGE IN VARIOUS YEARS
| Year in which pledge is met | Required Growth Rate
|
| 2015-16 | 6.6% |
| 2020-21 | 3.3% |
| 2025-26 | 2.2% |
Source: IFS calculations.
To meet the aspiration in 2015-16, state school spending
per head would need to grow by an annualised average amount of
6.6% in real-terms after 2010-11. To meet it in 2020-21 would
require a lower equivalent growth rate of 3.3%, whilst to meet
it in 2025-26 would require a still lower equivalent growth rate
of 2.2%. To set these figures in context, the figures announced
in Budget 2007 imply that state school spending per head will
grow by 3.2% between 2007-08 and 2010-11. If this growth were
continued beyond 2010-11, the aspiration would not be met until
about 2020-21.
Significance of Pledge
It is important to remember what meeting this pledge would
mean in practice. As we have pointed out before to the Committee,
one thing it is unlikely to achieve is to close the contemporaneous
gap. This is because spending per pupil is highly unlikely to
remain constant in real-terms.
To illustrate this point, the latest figures presented in
Figure 5 show that in 2006-07, private school spending per head
increased to about £8,580 per head (2005-06 prices), a real-terms
increase of 5.0% on the previous year.[8]
This compared with an equivalent growth rate of 4.0% over the
same period in the state sector. If anything, the latest figures
suggest that the cotemporaneous gap is widening rather than shrinking.
Figure 5
CLOSING THE GAP? THE EVOLUTION OF REAL SPENDING PER HEAD
IN THE PUBLIC AND PRIVATE SECTORS (2005-06 PRICES)
Sources: For public spending per pupil see Budgets 2006
and 2007, and Department for Education and Skills (2007), Departmental
Report: 2007. Figures for private sector spending per pupil
are taken from the annual census of independent schools conducted
by the Independent Schools Council from various years, http://www.isc.co.uk/Publications_ISCCensus.htm
The length of time it takes to achieve this pledge is the
most important indicator of its significance. This is because
the time-taken to achieve the pledge marks the "lag-time"
between state and private school spending per head. Reducing this
lag-time would mean shortening the time it takes for public sector
spending per head to reach what private sector spent in a specific
year.
We can assess what the current lag-time is by calculating
when this pledge, made in Budget 2006, would be met if Mr Brown
had instead made it when he became Chancellor in 1997. In 1997-98,
the private sector spent about £5,600 per head (2005-06 prices).
As is shown in Figure 5, state school spending per head is not
scheduled to reach this level until some point between 2007-08
and 2010-11, so that the current lag-time is about 11-12 years.
To reduce the lag-time, per pupil state spending would need
to reach that spent in private schools in 2005-06 around (or before)
2016-17/2017-18. However, if the current growth in spending per
pupil continues beyond 2010-11, then this level is unlikely to
be met before 2020-21, as we have shown above. If the lag-time
between private and state school spending per head is to be reduced,
then state school spending per head will need to grow by more
than 3.2% in real-terms. Reducing the lag-time by 1-2 years (and
so achieving the pledge in 2015-16) would actually require a real-terms
growth rate in per pupil state school spending in excess of 6%.
3. HISTORY AND
PRESENTATION OF
ANNOUNCEMENTS
The full CSR settlement for education announced in Budget
2007 followed a series of high profile announcements about elements
of education spending, made during the previous year, at Budget
2006, and PBR 2007. A short recap of the chronology of these announcements
is provided in Figure 6 below.
Figure 6
A TIMELINE OF RECENT ANNOUNCEMENTS ON EDUCATION SPENDING
A) Budget 2006
Schools Capital Spending Announced
|
B) PBR 2006
Total Education Capital Spending Announced
But schools element, 78% of total capital spending,
already announced in Budget 2006
Therefore only the non-schools (about 22% of total
education capital spending) was new spending
|
C) Budget 2007
Total Education Spending Announced
Since capital spending already announced only
current spending represents new spending. However
this is about 89% of all education spending.
|
While it is not possible to provide a full analysis of the
contents of these announcements here,[9]
some points about the history and presentation of these announcements
are worth making.
(a) There have been several occasions where it has proved
difficult to disentangle how much of any spending total announced
represented a new allocation of money, and how much was simply
a re-iteration of a commitment previously made. For example:
— While the education capital spending totals (announced
in PBR 2006) were announced in PBR 2006, the bulk of this total
capital spending consists of schools capital spending, which had
already been announced in Budget 2006. Our own analysis suggested
that the only new information to be gleaned from the announcement
of total education capital spending in PBR 2006 was a modest increase
in the non-schools element of education capital spending between
2007-08 and 2010/11, amounting to the equivalent of an increase
of £100 million in 2006-07 prices. (see http://www.ifs.org.uk/budgets/pbr2006/education.ppt
for more information)
(b) There have been a number of occasions where new education
spending plans have been presented in the form of cumulative totals
over a number of years, without accounting for economy-wide inflation.
This has the effect of generating very large numbers, whose meaning
is difficult to interpret. For example:
— In Budget 2006 the Chancellor announced in his
speech "£34 billion of new investment over five years",
a figure which can only be arrived at by adding together total
planned capital spending in nominal terms for each of the five
years 2006-07 to 2010-11. In last year's note to the Committee
we likened this presentation of the figures to a worker on approximately
male full-time average earnings (£25,000) being given a nominal
pay freeze, but is told he is being awarded a "total of £125,000
new income over five years". (http://www.ifs.org.uk/publications
.php?publication_id=3667)
— In PBR 2006, the Chancellor made a similar claim,
announcing "In the next four years a cumulative investment
in education alone of £36 billion" Again the presentation
of such cumulative totals mainly serves to make new spending allocations
look very large, and is a practice that was largely abandoned
after the Government's first Comprehensive Spending Review in
1998, when such a presentational device was widely criticised.
For instance, the Treasury Select Committee stated in its report
on the 1998 Comprehensive Spending Review that, "There is
thus no cash bonanza of the type which newspaper headlines might
suggest, but a steady increase in real resources. We recommend
that, for the sake of transparency, in future the Government should
refer to annual increases over the previous year rather than a
cumulative total."
(c) Many of the recent education spending announcements
made were not based on publicly available DfES or HMT series,
against which ready historical comparisons can be made. For example:
— Spending per pupil numbers given in Budget 2006,
PBR 2006 and Budget 2007—which form the basis of the Government's
aspiration for spending per pupil to match the private sector,
are a composite of capital and current spending per pupil—where
capital spending per pupil also includes PFI spending. By contrast,
the publicly available DfES series on spending per pupil that
were available at the time of these announcements were for current
spending only. It is only since the DfES Departmental Report 2007,
published in May 2007, that the new series has been made.
July 2007
1
The authors would like to thank the Education and Skills Select
Committee for commissioning this memorandum, and co-funding from
the ESRC-funded Centre for the Microeconomic Analysis of Public
Policy at IFS (grant number M535255111) is also gratefully acknowledged.
The authors would also like to thank Carl Emmerson for providing
many useful comments and suggestions. Any views expressed, as
well as any errors, are those of the authors. Back
2
One may be more concerned with measures of education resources
per head rather than overall levels of spending. Thus when comparing
the relative generosity of spending across such long time frames,
one may wish to take account of long-term trends in student numbers.
For instance, between 1979 and 1997, the number of pupils in maintained
schools decreased from 8.4 million to 7.3 million, with little
change since 1997 (Source: http://www.dfes.gov.uk/trends/index.cfm?fuseaction
=home.showIndicator&cid=1&iid=2). Back
3
C Emmerson, C Frayne and G Tetlow, (2007) Challenges for public
spending, in R Chote, C Emmerson, A Leicester and D. Miles
(eds), The IFS Green Budget 2007, Commentary no 102, IFS,
London, 2007 (http://www.ifs.org.uk/publications.php?publication_id=3841). Back
4
Ibid. Back
5
This is the Treasury's cautious assumption for real GDP growth
used for its public finance projections. Back
6
Rather than projected economy-wide inflation at the time of the
1998 CSR, which implied a growth rate of 5.1%, subsequent inflation
was lower than expected. Back
7
Note that these numbers do not add up to £530 due to rounding. Back
8
Note that the ISC census has this year adopted a new method for
calculating average fees. The figure the Chancellor chose to use
in Budget 2006 to measure equivalent spending per pupil in the
private sector was the annual average day fee for day schools.
Under the new ISC method this was £8,320 in 2006-07, whilst
under the old method it was £8,580 (both expressed in 2005-06
prices). However, the estimate under the new methodology is only
available for 2006-07, and we have thus chosen to continue using
those calculated under the old methodology. Back
9
For a more detailed analysis, see the following briefing note
we wrote based on previous evidence, http://www.ifs.org.uk/publications.php?publication_id=3667,
or the following presentation we made following the Pre-Budget
Report in December 2006 http://www.ifs.org.uk/budgets/pbr2006/education.ppt Back
|
