Are rainy day savings by schools affecting outcomes?

Is there any relationship between the reserves a school holds and the outcomes for its pupils? One thesis might be that schools that spend more of their revenue budgets, and don’t add to their reserves each year, do better for their pupils than those schools more concerned with ‘saving for a rainy day’ or adding to their reserves for some other reason. Of course, this thesis only really works with a funding model that funds schools appropriately. Nevertheless, it is question worth asking.

This post looks at the evidence around the relationship between the quoted reserves of the primary schools across three multi-academy trusts (MATs) of differing sizes. The schools are in a range of different locations, and have a range of pupil numbers, but are all located within the boundaries of a single local authority.

KS2 achievement Percentage %No of KS2 Pupils 24/25 reserves per KS2 pupil
4230 £        24,333
6916 £        16,313
6422 £        16,273
3625 £        14,480
2711 £        12,364
9111 £        12,249
4715 £        10,600
3429 £        10,172
6913 £        10,077
5016 £          9,375
5016 £          8,688
3327 £          7,667
5645 £          7,267
5648 £          6,896
6916 £          6,813
6939 £          6,692
9315 £          5,933
5831 £          5,484
8318 £          5,287
8211 £          5,091
7218 £          5,000
6531 £          4,258
5022 £          4,182
5531 £          4,129
4821 £          4,098
7359 £          3,847
5028 £          3,536
7315 £          3,467
7528 £          3,214
4827 £          3,111
7061 £          2,738
6758 £          2,535
9123 £          2,531
5657 £          2,509
8323 £          2,087
5944 £          1,864
7131 £          1,677
5770 £          1,614
9212 £          1,362
5260 £              950
7629 £              862
7260 £              700
5618 £              611
4529-£         1,655
8614-£         1,714
7919-£         2,474
5831-£         3,084
867-£         3,295
8938-£         4,756
5315-£         8,667
4825-£       11,040

The range of reserves, as taken from the reserves in the accounts filed at Companies House by the MAT, vary per KS2 pupil from a high of £24,333 per pupil, to £611 for schools with those schools with reserves. Eight schools are shown as having deficits

Leaving aside, at this point for schools with deficits, it is worth comparing the other schools in the table with the average outcome for all pupils in the local authority of 62%.

The schools in the table, are of different sizes. In the schools in the table the range of KS2 pupils extends from schools with just 11 KS2 pupils, to one with 70 KS2 pupils. Where a school has a stable intake, the KS2 reserves per pupil is easy to spread across the schools, although some schools still start pupils, at age 5, many at age 4 and a few even younger and with a nursery class, and this may make a difference to reserves.

In this era of falling rolls, allied to parental choice, some school’s KS2 numbers would overestimate the school population if grossed up across all year groups. For that reason, I have just used the KS2 pupil number for this exercise.

Six of the top 10 schools with the highest reserves per pupil had achievements below the local authority level of 62%. This compares with four of the ten schools with the lowest reserves (not with a deficit).

It is interesting to note that one MAT has 12 schools in the top 15 schools with the largest reserves per KS2 pupil, and 15 in the top 20 schools. Of the other two MATS, one has three schools in the top 20 schools by reserves per KS2 pupil, and the other just two schools.

I suspect that both these MATs more actively work with their schools than the other MAT in terms of making use of their cashflow. All the schools in this latter MAT in the top 20 by reserves are also well above average for performance levels. The position is more mixed for the other smaller MAT, but I suspect its policy is evolving as it has acquired more schools.

How much does parental choice, and falling rolls play a part in determining reserves? Certainly, primary schools that have no spare places across all year groups will do better with a funding model that is weighted towards pupil numbers than schools unable to make full classes even by mixing year groups together. In another post, I will look at school size and reserves.

It would be interesting to conduct this exercise across all schools in a local authority. However, pooling of reserves by some MATs, and a lack of visibility for the reserves of maintained and voluntary schools makes that task effectively impossible.

No doubt the DfE could ask the question about why there is so much difference in reserves per pupil between schools, and could ask if some schools could achieve more for their pupils if they kept less in their reserves? School governors, MAT trustees and the teacher associations, as well as local politicians, might also like to ask this question.

Do means matter?

The DfE has published some performance data for academies and multi academy trusts Multi-academy trust performance measures (key stages 2, 4 and 5) – GOV.UK (www.gov.uk) The outcomes are quite rightly heavily hedged about with qualifications about how schools have become academies, and also that schools differ in size, character and parental choice. Indeed, I wonder whether the original reason for why a school became an academy, perhaps more than a decade ago is still relevant?

What struck me at first glance was that as the secondary sector becomes dominated by academies there is a reduction to the mean (average). Various Secretaries of State have wanted all schools to be above average, as this exchange with Michael Gove when in front of the Education Select Committee revealed. Michael Gove’s Kafkaesque logic – Left Foot Forward: Leading the UK’s progressive debate However, I don’t think he was the only Secretary of State to fall foul of this aspiration. Academies in the group forced to change their status because of under-performance not surprisingly do less well than those that chose to become an academy.

The key question for the current Secretary of State must be what do you take from this data in terms of the ‘levelling up’ agenda? I don’t think the present incumbent of the post of Secretary of State for Education has been asked the question about averages, but that shouldn’t stop her asking the question about what policy changes are needed based upon these outcomes?

As an additional discussion point, the secretary of State might like to ask her officials two further questions. What is the relationship between the schools in these tables and the percentages of NEETs produced by different types of schools, and how can schooling work to help ensure as many as possible of our young people eventually enter the labour market at the end of their initial education and training journey? After all, we should all be life-long learners.

After last year’s aborted attempt to make all schools academies, and the mauling of the Bill in the House of Lords there is still a need to ensure the middle tier works to the best advantage for all children. Whether there is a role for local democracy in schooling is still a live issue, but not one that will feature highly at the next election.

But, regardless of who runs schools, there is still work to be done to achieve excellence for all and that no child is left behind, to quote just two aspirational messages from past attempts at improving the outcomes for our schooling system.

Of course, without sufficient teachers, the risk is of deterioration not improvement in outcomes; not what the Chancellor wants to see if the economy is to continue to grow.

Some still do better than others

The DfE has published an interesting report on outcomes by ethnicity and Free School Meals. It might have been even more useful with a section on gender added and also some regional breakdown to see if the additional funding in the London area makes any difference to outcomes. Outcomes by ethnicity in schools in England – GOV.UK (www.gov.uk)

For many, the inclusion of Traveller children and their performance will come as a shock as these groups don’t regularly appear in most tables. Funding for the education of traveller children hasn’t been high on anyone’s agenda for many years. Perhaps now it the time to reassess how we offer education to the children in the travelling community.

The Report makes clear the poor outcomes for certain sections of some communities, especially those pupils on Free school Meals.  

According to the DfE report there were over 8 million pupils recorded in the school census in 2021. Ethnicity responses in the report are grouped into 17 options or “Any other” as a catch-all category. I guess some South American children might fall into this group.

The DfE points out that it is important to consider variation within groups, especially when aggregated into larger groups such as “White” or “Asian”.  When aggregated 72% of pupils (5.9 million) described their ethnicity as White, 11% (900,000) Asian excluding Chinese, and around 6% each as Mixed (520,000) or Black (460,000). 2% of pupils described themselves as belonging to ethnic groups not captured in the census (170,000), and 0.5% of pupils identify as Chinese (37,000).

An important finding is that the proportion of White British pupils meeting the expected standard falls at each stage in their education.

Other groups also see fluctuations across stages, but the effect is most pronounced in the White British Group. As seen in figure 2, White British pupils fall from 5th of 18 groups in younger groups to 10th later in school. This is reflected in the average Progress 8 score of a White British GCSE entrant being negative (-0.14) where 0 represents average progress through secondary school. The DfE comment that since Progress 8 is a relative metric, we cannot say whether this trajectory represents “catch-up” of some non-White British groups or a “falling behind” effect.

However, some other groups also fare badly according to the report

‘Black Caribbean, Mixed White and Black Caribbean and Other Black pupils are all less likely to meet the expected standard at all stages than White British pupils, and the size of this gap doubles between 4- to 5-year-olds and GCSE pupils.

Comparing between a class of 30 White British pupils and a class of 30 pupils from these 3 groups, on average 1 or 2 more pupils in the Black and Mixed class would be below the expected standard at 4 to 5 years-old, whereas 3 to 5 fewer pupils would receive a strong pass in English and Maths GCSE. 11. The average progress 8 scores of Black Caribbean (-0.30) and Mixed White/Black Caribbean (-0.37) pupils are more negative than the White British group. Pupils selecting Black Other (+0.08) have slightly positive average progress 8 score.

Gypsy/Roma, and Irish Traveller pupils have the consistently lowest levels of attainment of any ethnic group, and the most negative progress 8 scores.’

The report notes that five groups made below average progress throughout secondary school. These groups are – White British, Black Caribbean, Mixed White/Black Caribbean, and Gypsy/Roma, Irish Traveller pupils. These groups start with lower attainment scores following primary school, so low progress scores represent a confounding effect where these groups are falling further behind their peers. Controlling for FSM, only White and Black Caribbean groups have below average progress among non-FSM recipients. White FSM recipients have the lowest progress among all aggregated groups, and Mixed and Black groups have below average progress.

This report is powerful evidence for the levelling up agenda discussion and also for the discussion on the hard National Funding Formula currently being discussed as part of the Schools Bill before parliament. Once again, it raises the question over the degree of hypothecation required in funding schools and how the money is both used and evaluated. Interestingly, there is nothing in this report about the use of Pupil Premium monies as a hypothecated grant.

Education counts, but so does the family

The report on social mobility issued today https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/917278/The_long_shadow_of_deprivation_-_differences_in_opportunities_across_England.pdf raises a number of interesting questions. Most are not new, but they are none the worse for restating.

Life changes, at least as far as incomes are concerned, seem to be a combination of education, local labour markets, soft skills and parental ability to offer support for life chances.

Education effects are relatively similar, although areas where there are either selective schools or strong local private school clusters seemingly do have an effect on outcomes.

The comments in the report on Oxford, my home area and where I represent a Division on the county council are interesting

Oxford is an example of a place where social mobility is high, with the smallest pay gaps – it is in the top 10 places with the smallest pay gaps outside London, as shown in Table 1.2. Swindon represents a place with average pay gaps and mobility, while on the right, Bolton represents a place with large pay gaps and less mobility. In all three authorities, the proportion of the pay gap explained by education (the orange part of the coin stack) is the same (around 14 percentiles). But because total pay gaps are smaller in Oxford, education explains a larger proportion of the pay gap there compared with Bolton. The main difference between the most and least mobile areas is the black part of the coin stack – the role of family background that persists beyond education into the labour market.

This finding challenges the notion that educational investment alone will remove differences between areas. Education gaps account for most of the pay gaps in all areas, and reducing these is important in and of itself. However, we also need to look at equalising labour market opportunities available for young people with the same education level as those from richer backgrounds if we are to ‘level up’ between places. Beginning to tackle this gap requires us to understand what drives it – only then can we design effective interventions that address the specific roots of intergenerational disadvantage. Pages 40 and 41

Personally, I don’t think we should give up on education investment in order to ensure those currently not benefiting from our education system are able to improve their outcomes. In 2011, Oxford Key Stage 1 results were the worst for any local authority. This was despite the success levels of schools in the North and West of the City. They have improved since then. In my view, the current National Funding Formula does not provide enough incentives to help improve outcomes.

However, I accept that creating new employment opportunities is critical to social mobility. In the 1960s, the Intermediate Areas Report recognised this issue. More recently, government have seemed to accept a policy of building up successful areas such as London and the South East and ignoring other ‘smokestack’ areas. This report recognises the importance of employment opportunities at all levels need to be provided across the country. Without these opportunities, vacancies for those remaining in many areas will be limited and often lower paid jobs.

Of course, the greatest sadness about this study is that it only deals with males due to data issues. Are women more socially mobile and better able to make use of educational outcomes or are they even more fettered by circumstances/