Education under the Coalition

In this article Anthony Heath and Anna Mountford-Zimdars summarise the evidence on the effects of Coalition policies on education. They find that, while the government implemented its core policies – the pupil premium and the promotion of academies and free schools – it is not clear whether its reforms have had any effect in raising educational standards and reducing class inequalities. Their earlier assessment of education under the Labour government, 1997-2010, is included in the special journal edition which is free to download until the election.

  1. Introduction

In their 2010 manifesto the Conservatives had emphasized raising school standards and closing the attainment gap between pupils from the richest and poorest backgrounds, two themes which had also been present in the previous Labour manifestos. Their partner in the coalition, the Liberal Democrats, had rather different emphases, focussing on cutting class sizes in schools and famously promising to scrap ‘unfair university tuition fees’. In the coalition agreement, the key elements became:

  • To fund a significant premium for disadvantaged pupils from outside the schools budget by reductions in spending elsewhere.
  • To promote the reform of schools in order to ensure that new providers can enter the state school system in response to parental demand.

There followed a major programme of reform, not all of which had actually figured in the coalition agreement. The most eye-catching reforms which the coalition undertook in office were to encourage private providers to establish Free Schools and to extend the Academies programme . There were also, later in the life of the coalition, major reforms to GCSEs intended to strengthen standards and make them ‘more challenging’, notably by ending the modular system and moving to assessment at the end of the course, changing subject content, and restricting the number of ‘equivalent’ qualifications which count towards school performance tables.

In some respects these were continuations of Labour policy, particularly the emphasis on school choice and extending Labour’s Academies programme, although the GCSE reforms diverged from previous Labour policy. It is also worth emphasizing that these reforms applied to England, since education is a devolved responsibility.

In the case of further and higher education, the Coalition’s principal reforms were:

  • Abolition of Labour’s Education Maintenance Allowance and replacement by a 16-19 Bursary Fund (with a much lower level of funding)
  • Implementation of Labour’s policy of Raising the Participation Age (RPA) to 17 (from 2013) and 18 (from 2015)
  • Extension and reform of apprenticeships
  • Increasing the fee cap in higher education to £9000 for full-time students

A detailed analysis of both inputs and outcomes for schools has been carried out by Ruth Lupton and Stephanie Thomson (2015), and for further and higher education by Ruth Lupton, Lorna Urwin and Stephanie Thomson (2015). We summarize some of their key findings below.

  1. Inputs

The Coalition protected school spending in real terms. Current spending on schools in England increased from £39.5bn in 2009/10 to £43.8bn in 2013/14, although capital spending decreased from £6.5bn to £2.8bn. The pupil premium directed more money to schools with ‘poor’ intakes (judged by Free School Meal eligibility). Among primary schools the least deprived schools saw a small (real) increase in grant funding of around 1.1%, while the most deprived saw a larger increase of around 7%. Among secondary schools, the least deprived saw a loss of around 2.5%, while more deprived schools had increases of about 4.3%.

There was also a major expansion of the Academies programme. In January 2010, just over half of state-funded secondary schools were community schools overseen by local authorities with only 6% being academies independent of LAs. By January 2014 the proportion of secondary schools overseen by LAs had fallen to around one fifth, and over half were now academies (in which Free Schools are technically included). Among primary schools, changes were less dramatic. The percentage of Academies increased from 0% to 11%, almost wholly at the expense of community schools rather than of voluntary-aided or voluntary-controlled primary schools.

Despite the protection of school spending in real terms, and the use of the pupil premium to help primary schools, class sizes in primary schools began to increase (largely due to demographic pressures).   After declining considerably under the Labour governments between 1997 and 2009, class sizes increased from an average of 26.2 pupils per class in 2009 to 26.9 in 2014 (Neve, 2015). The average hides differences between classes for different age groups; the average class size for 5-7 year-olds grew by 7% from 2007 to 2014 (from 25.6 to 27.4), while average class sizes for 7-11 year-olds remained constant throughout the period. Meanwhile, average secondary school class sizes fell steadily from 21.7 in 1998 to 20.6 in 2009 and to 20.1 in 2014.

  1. Outcomes

Changes in assessment procedures make it difficult to evaluate changes over time in school attainment. However Lupton and Thomson (2015) report that KS2 reading and mathematics attainment both improved. At KS4 (GCSE) the DfE released two sets of headline data in 2014, the official results using the new qualification rules and an adjusted set following the old qualification rules (intended to enable comparison on a like for like basis over time). The trends over time also differ according to whether one uses the criterion of 5 or more GCSEs at grades A* to C, or the more challenging criterion which requires passes at both English and Maths among the 5 subjects (5 A*-CEM).   The most positive story from the Coalition’s point of view was that the percentage obtaining 5 A*-CEM was slightly higher in 2014 than it had been in 2010.

There remain considerable doubts, however, whatever rules or criterion one takes, whether percentages obtaining GCSEs (or expected standards at KS2) actually give valid measures of changes over time in the skills and competencies of school children. Independent measures conducted as part of cross-national programmes, such as that of OECD, suggest that changes over time have been much smaller than the changes shown by government data on attainment at KS2 or KS4. Between 2009 and 2012 (the latest year available), scores on the OECD’s PISA tests (taken around age 15), for example, showed only small improvements for reading, small declines for maths, and stability in science (Wiertz 2015).

In addition it is not clear yet whether Academies actually do improve educational standards. The House of Commons Education Committee concluded in January 2015 that “it is too early to judge whether academies raise standards overall or for disadvantaged children”.   There have also been concerns expressed about the governance and financial control of Academies. In October 2014 the National Audit Office concluded that the DfE and other oversight bodies had not demonstrated the effectiveness of their interventions. “The NAO finds that the DfE and others, such as the Education Funding Agency and local authorities, have not tackled underperformance consistently. The spending watchdog, therefore, cannot conclude that the oversight system for maintained schools and academies is achieving value for money.”

Lupton and Thomson also examine trends in socio-economic inequalities, using so-called Eligibility for Free School Meals as the proxy for socio-economic background (a criterion which also appears to be the main one currently used by government). (Eligibility is in fact a misnomer, since families have to apply in order for their eligibility to be approved by the local authority. Possibly many ‘eligible’ families do not apply in the first place.) Lupton and Thomson found that socio-economic inequalities at GCSE declined up until 2013 but then widened sharply in 2014 when the new qualification rules were introduced.

Academic research (Mills 2015) using the standard ONS socio-economic classification (NS-SEC) has also shown a slight narrowing of the inequalities up until 2013 at GCSE (based on data from the Labour Force Survey) but data for 2014 are not yet available. The narrowing of class inequalities, however, had begun before 2010 and may well not be due to reforms undertaken by the Coalition (which might in any event take a longer time to bear fruit).

In the case of 16-19 education there were modest increases in proportions of 16, 17 and 18 year olds in full-time education or training, although it is clear that not all 16 year olds are as yet participating despite the RPA. There has also been a decline in the proportion not in education, employment or training (NEET) which had declined to its lowest level since 1997 by the end of 2013. Lupton et al (2015), however, show that the increase in participation is a continuation of longer-term trends and is unlikely to be explained solely by the Coalition’s reforms.

With respect to higher education, applications fell sharply in 2012 with the introduction of the new fee regime. This was almost certainly because many students had decided to apply in the previous year, rather than taking a gap year, in order to avoid the increase in fees.  Applications recovered by 2014.   The application rate for students domiciled in England rose from 31.3% in 2010 to 33.2% in 2014, again a continuation of longer-term trends.

However, there was a dramatic drop in mature students’ participation in higher education. The number of mature undergraduate entrants to universities fell by 40% between 2007-8 and 2012-13. Universities UK have attributed this to increased fees and the switch to loans in 2012, although as Lupton et al (2015) point out, the economic downturn and reductions in employer funding may also have been contributory factors.

  1. Conclusions

The coalition clearly did implement the two central planks of the coalition agreement on the pupil premium and new providers. However, it is not as yet clear whether these reforms have had any effect either on educational standards or on reducing class inequalities. Class inequalities at GCSE (although not perhaps at higher levels of education) have been slowly declining, but it is far from clear that this has been due to the coalition’s reforms.

The absence of independent and up-to-date evidence on students’ skills continues to be a major obstacle to monitoring the effectiveness of government policy in raising standards, just as we reported in our previous article on education under New Labour (Heath et al 2013). The whole area is bedevilled by the paucity of authoritative, independent assessment and is largely reliant on administrative statistics, with all the pitfalls to which they have long been known to be prey.

At present, the most that can be said is that the jury is still out on the question of whether the Coalition’s reforms actually made any difference to either standards or socio-economic inequalities.



The Coalition’s record on poverty and inequality

In this fourth article on Coalition Economics, Robert Joyce and Luke Sibieta (Institute of Fiscal Studies) consider the effects of the Coalition’s policies on inequality and poverty. They discuss the impact of the recession, which tended to reduce inequality, and then of government tax and benefits policies, which have tended to increase it, and the rising incidence of child poverty. Sibieta and Joyce’s earlier assessment of income inequallity under the Labour government, 1997-2010, is included in the special journal edition which is free to download until the election.

The coalition government came to power just after the Great Recession and during the associated fall in real earnings. One cannot properly understand the coalition’s record on inequality and poverty without considering that context. The Great Recession led to large falls in workers’ real pay between 2009 and 2011, which led to bigger falls in incomes further up the income distribution. However, the coalition government also had to implement a large fiscal consolidation in response to a hole in the public finances that had opened up or revealed itself due to the recession.

Decisions on how to implement this consolidation have had further important implications for inequality, with cuts to working-age benefits leading to reductions in income primarily towards the bottom of the income distribution and some net tax rises primarily hitting a smaller group towards the top.

Hence, falling real pay meant that incomes initially fell more towards the top of the distribution – partly at the end of Labour’s reign and partly at the start of the coalition’s - but the fiscal consolidation probably means that incomes are falling at the bottom of the distribution towards the end of the current parliament.

  1. Changes in household income during the Great Recession

The coalition came to power right in the middle of a period of large falls in workers’ pay. Between 2009 and 2011 – a period which neatly sandwiched the May 2010 election – median weekly earnings fell by 7% in real terms ( Because earnings are a larger source of household income, on average, further up the income distribution, this acted to reduce income inequality. This came on the back of a reduction in inequality during the recession itself, as the real value of benefits and tax credits had been boosted by falling inflation and some discretionary increases, meaning that the bottom had been catching up with the middle. These patterns are shown in Figure 1.

The precise timing of this fall in real earnings was not entirely independent of policy choices. The Labour government’s temporary VAT reduction between 1 December 2008 and 31 December 2009, as a fiscal stimulus measure, had kept inflation very low and hence real earnings did not fall during this period. But ultimately the fall in earnings was an inevitable, if somewhat delayed, impact of the severe economic contraction.

Had the election happened a year earlier this would all have happened under the coalition; had it taken place a year later it would have been largely on Labour’s watch. In truth it would be economically arbitrary to attribute trends in the income distribution since May 2010 to the coalition and anything before that to the previous government. The large fall in income inequality between 2009 and 2011 was driven by the large scale of the recession and the historically unusual nature of the labour market shock that came with it – namely widespread falls in workers’ pay, rather than sharp rises in unemployment.

Figure 1. Real changes in real household income from 2007–08 to 2012­–13, by percentile point sibjoyfig1 Notes: Percentiles 1-4 and 96 to 99 are excluded due to high levels of statistical and modelling uncertainty. Incomes are deflated using the RPIJ index.

Source: Cribb et al (2015);

  1. Role of tax and benefit policy

It is more instructive to think about the coalition’s record on inequality and poverty by looking at what it has done since coming to power in response to the hole in the public finances. One obvious thing to look at, which affects inequality via direct policy levers, is tax and benefit policy. Considering all tax and benefit changes between January 2010 and May 2015 together, this group of measures have had the largest impact on a small group at the very top of the income distribution. This is mainly due to the implementation of tax measures pre-announced by the previous Labour government (such as the introduction of a 50% income tax rate, subsequently reduced by the coalition to 45%), which had their biggest effects just before or early in the coalition’s period in government.

The coalition government has chosen to implement a package of cuts to benefits and tax credits totalling £16.7 billion per year by 2015–16, and a net tax rise of £16.3 billion (all in 2015–16 prices). Figure 2 below (taken from Browne and Elming, 2015) shows the net impact of these changes on the incomes of different groups. Households are divided into ten equally sized deciles depending on their household income, but also different demographic groups within each decile (pensioners, working-age adults without children and working-age adults with children).

These measures (such as the increase in VAT and cuts to benefits and tax credits) have had the largest impact on approximately the bottom third of the income distribution but have been focused on those of working age. Low-income pensioners seem to have been largely protected from these cuts. Individuals in the middle and upper-middle parts of the income distribution have actually gained on average from tax and benefit changes under the coalition (such as increases in the personal allowance), though they had been more affected by cuts in earnings during and immediately after the Great Recession. In sum this probably means that inequality has risen in the latter half of the parliament, at least across much of the distribution.

Figure 2. Impact of tax and benefit reforms introduced between May 2010 and May 2015 by income decile and household type

sibjoyfig2Source: Browne and Elming (2015),

  1. Overall changes in inequality during recession and recovery

Figure 3 brings these stories together by showing expected changes in income across the income distribution between 2007-08 and 2014-15 (using simulation techniques beyond 2012–13 – the latest household income data – accounting for changes to taxes and benefits and trends in employment and earnings), covering the whole period of recession and coalition government. Perhaps contrary to popular perception, inequality will in fact still be lower in 2014–15 than in 2007–08. This is due in no small part to the nature of the labour market shock, which caused widespread falls in real pay rather than the kinds of rises in unemployment – concentrated on the low skilled – seen in previous recessions. Cuts to social security as part of the post-recession fiscal tightening have so far unwound only some of that sharp prior fall in inequality. This highlights the important point that, whilst the fiscal consolidation measures – which increase inequality across much of the income distribution – are the result of direct choices made by the coalition, these choices were of course made in a context where inequality had just fallen substantially and earnings levels had fallen relative to benefits.

There is a twist to this story though. Inflation has been hitting the poor harder over this period (primarily between 2007–08 and 2009–10), because they are less likely to have benefitted from plummeting mortgage interest rates and more likely to spend large shares of their budgets on food and energy, which have risen in relative price. The black line on Figure 3 shows that accounting for this does make an important difference. Nevertheless, even having done so there is no rise in inequality between 2007–08 and 2014–15.

Figure 3. Change in real household income from 2007–08 to 2014­–15, by percentile point sibjoyfig3Notes: Percentiles 1-4 and 96 to 99 are excluded due to high levels of statistical and modelling uncertainty. Incomes are deflated using the RPIJ index.

Source: Cribb et al (2015);

  1. Poverty

When it comes to poverty, trends and policy have been somewhat more confusing. Measures of relative poverty – such as the proportion of individuals with incomes below 60% of the contemporary median – have fallen significantly since the start of the Great Recession and over the period of coalition government (from 18% in 2007-08 to 16% in 2010-11 and 15% in 2012-13). However, this is mostly because falls in median income have reduced the poverty line, rather than because of rises in the living standards of low-income households. Absolute poverty (defined as household incomes less than the 2010­-11 median) has risen since household incomes began to fall (rising from 15% in 2009-10 to 17% in 2012-13) and is likely to have increased further as a result of cuts to benefits and tax credits taking effect from 2013-14 onwards.

However, there has been a distinct lack of clarity over what the coalition government goals really are with respect to poverty. There is currently a legally-binding target to ‘abolish’ child poverty by 2020, defined as reducing relative child poverty to below 10% as well as targets across a number of other domains. The coalition government initially announced rises in the Child Tax Credit alongside other benefit cuts in order to ensure no net impact on child poverty; but swiftly abandoned this practice (and cancelled the second planned CTC rise). Child poverty stood at 17% in 2012-13 and is expected to increase to 21% by 2020 ( ), clearly way off the legally-binding target. The coalition government has expressed some dissatisfaction with the current measure of child poverty and has consulted on a new measure of child poverty (, though this has led to nothing as of yet. We are therefore left in an odd situation whereby there is a legally-binding target to reduce child poverty by 2020 and no credible plan to meet that target.

The coalition government has outlined a number of strategies to reduce child poverty in the long-run ( For instance, it has emphasised improving work incentives, which have indeed been strengthened for most groups over the past five years ( Improving work incentives and reducing complexity have been important motivations for the introduction of Universal Credit. However, the implementation of this has been fraught with delays and is unlikely to be fully implemented until at least 2020 on current plans (

The coalition has also emphasised improvements in education outcomes, which are likely to be crucial determinants of later-life earnings capacity. The introduction of the disadvantaged pupil premium was a major plank of policy to improve the educational performance of disadvantaged children. However, there were already very significant amounts of funding targeted at poorer pupils before the coalition came to power ( and there is no evidence to suggest that this led to a larger reduction in the attainment gap. Moreover, any such improvements in educational outcomes are likely to feed through into the labour market in the long-run.

Looking to the future, an important challenge for the next government will be to set out a clear set of objectives in the area of child poverty alongside measures to meet those objectives, whatever they may be. For instance, are they still committed to the 2020 target and, if so, how are they going to meet it?