Too few schools or too few kids?

8th February 2024 by Timo Hannay [link]

Among the most politically salient topics in the UK today are (i) population growth, driven largely by net immigration, and (ii) shortfallings in the provision of public services such as healthcare and education. Often these are combined into a single narrative in which the growing number of people is a driver of scarcity in public services.

As our previous analyses have shown, it's true that some parts of England are under-served by schools, but at least for primary schools, the cause is not too many children for the available places, it's insufficient children to keep schools open – and this trend now looks set to extend to secondary schools too. Our last demographic analysis was in 2019, since when birth rates and school cohort sizes have continued to decline. This post brings us up to date.

To summarise:

• In 2016, schools in England had seen 10 years of unbroken year-on-year growth in cohort sizes. But by 2018, new cohorts had begun to decline in size and they have fallen ever since. Recent birth rates indicate that this overall trend will continue for at least another few years and possibly much longer. The result is that currently declining primary school enrolments will continue to fall.
• As these cohorts grow older, secondary school intakes will also start to reduce in size. For the time being, the total secondary school population will continue to rise as the graduating cohorts remain smaller than the incoming ones, but in the next four years or so this balance will tip and overall secondary school numbers will also start to fall.
• Crucially, the effects of these macro trends do not impact all schools equally. For example, the decline in primary school enrolments since 2019 has disproportionately affected poorer schools, smaller schools and those with lower Ofsted ratings. Some regions (eg, London and the South West) have also been affected more than others. These same schools also tended to start with lower occupancy rates and are therefore at even at greater risk of becoming unviable.
• Though secondary schools in those same groups (poor, low-rated etc.) have shown particularly strong growth while overall headcounts have been rising, the coming decline in numbers seems likely to affect them disproportionately too, resulting in rapid reversals in enrolment trends that they might not be anticipating.
• At the same time, the proportion of pupils in England attending an independent school have been in gradual decline, albeit from very different starting points in different parts of the country.
• Whilst local situations obviously vary, and this analysis looks only at numbers of school places, not their quality or suitability, it suggests that the greatest demographic challenge for schools in England is not too many children, but too few. This is particularly true when, as now, funding is tight and tied directly to pupil numbers.

 

The Children of Men

It is an often overlooked fact of our education system that the numbers of children in each year group can vary, sometimes by large proportions and often proceeding in waves, as shown in Figure 1. (These statistics come from the DfE's annual pupil census, conducted each January; see our previous analysis for further details.)

In 2016, the size of each subsequent school cohort had been rising for about a decade, but by 2018 this trend had begun to reverse. If we fast-forward to 2023, it becomes apparent that incoming cohort sizes are now in long-term decline, with the number of Reception pupils in 2023 at around the same level as the group in Year 11.

 

The Matthew effect

But how have these general trends impacted schools? In particular, are they evenly spread or do they affect some schools more than others? Figure 2 shows 2019 school occupancy rates (the number of pupils as a percentage of reported school capacity; horizontal axis) and the changes in these occupancy rates between 2019 and 2023 (vertical axis) for different types of primary school. The large grey dot in the centre represents all state primary schools, which were at 95.4% occupancy in 2019, a proportion that fell by 2.8 percentage points in the period up to 2023.

Looking by Ofsted rating, 'Outstanding' schools not only had the highest occupancy in 2019 (99.2%), they also went on to fall by the smallest amount (0.8 percentage points). In contrast, 'Inadequate' schools started well below full capacity in 2019 (91.7%) and have fallen by more than 10 percentage points since that time. (Hover over the dots in Figure 2 to see accompanying details.)

There is a similar pattern for different school sizes, with smaller schools showing both lower initial occupancy rates and larger declines, as well as for schools with higher levels of disadvantage, and those in certain regions – notably London and the South West. In other words, schools that were already struggling to fill their places are often those coming under more even pressure as these trends continue to bite: "from him who has not, even what he has will be taken away". This is consistent with what we found in our previous analysis. Not only have school populations continued to decline, but the effects are still experienced disproportionately by these same groups of schools.

It is worth mentioning that, as before, not all groups of schools show the same pattern. For example, among different types of academies or faith schools, those with lower occupancies in 2019 tended to show more modest declines.

(Explore further by clicking on the legend in Figure 2 to turn different groups on or off, or double-clicking to show one group on its own.)

Figure 3 shows the same analysis for secondary schools. These had lower overall occupancies (87.5% in 2019) and have been growing (up 5.3 percentage points from 2019 to 2023). Perhaps because of this growth, the trends tend to run in the opposite direction to primary schools. For example, schools with higher Ofsted ratings started with higher occupancy rates but achieved lower growth. Similar trends were evident by disadvantage and prior attainment levels. Note also the comparable patterns for single-sex schools and grammar schools. Presumably these trends are due at least in part to capacity limits: when total pupil numbers are growing, not much of the increased demand can be absorbed by schools that are already close to full.

(Explore further by clicking on the legend in Figure 3 to turn different groups on or off, or double-clicking to show one group on its own.)

 

Pupil projections

Given these recent falls in primary school populations, increases in secondary school populations, and their consequent effects on different types of schools, what can we expect to see over the coming years? In short, more of the same for primary schools, but a change of direction for secondary schools, driven by continuing declines in birth rates.

Figure 4 shows a comparison of birth rates with corresponding pupil cohort sizes (bearing in mind that birth rates use calendar years while pupil headcounts use academic years, so the alignment is not perfect). Red columns show numbers of live births in England and Wales during each calendar year; blue columns show 2023 cohort sizes by year group, as shown in Figure 1. There is an obvious – and unsurprising – correspondence between birth rates and pupil cohort sizes. (The former exceed the latter for two main reasons: (i) the numbers of births are for England and Wales, while the school cohort sizes are for England only; and (ii) school cohort sizes capture only pupils in the state sector, not those attending independent schools.) The main message here is that, apart from the small mid-COVID bounce – what else was there to do during lockdown? – birth rates have continued to decline. We therefore expect pupil cohort sizes to continue to shrink by similar proportions.

Consistent with this, the Department for Education forecasts that the nursery and primary in England will continue to fall – by a further 9% between 2023 and 2028. They also expect the secondary school population in 2028 to be very similar to that in 2023, but on a declining trajectory having risen slightly in the interim.

Where does this leave schools? We anticipate that primary school populations will continue to fall over the next few years, and that these declines will disproportionately affect schools that are small, have high proportions for children from poor families and/or have low Ofsted ratings. Over the same period, secondary school populations will stabilise and then start to decline. As they do, we expect the reductions in headcount to fall disproportionately on schools of similar kinds to those that are bearing the brunt of the decline in primary headcounts – ie, those that are smaller, poorer or lower-rated. Since these are also the kinds of secondary schools that have grown the most during the recent boom years, it is not clear whether they will be anticipating such a shift in fortunes.

Of course, fluctuating pupil rolls represent a challenge to schools not only because of the practical consequences of providing for different numbers of children each year, but also because funding is determined by intake. Ultimately, whether or not a school is viable is largely a financial judgement. As we have reported previously, school budgets have been increasingly squeezed, presenting a potentially dangerous double whammy. The financial angle is something we intend to return to soon on this blog.

 

Independence movement

The above analysis has focused on state schools, partly because that's where most of the pupils are, and partly because they provide much more data. But it's also worth considering the independent sector. We can't do the same level of analysis because independent schools don't provide pupil numbers by year group, or even phase, only total pupil headcounts. Nevertheless, we can use this to look at the proportions of pupils who attend independent schools and how, if at all, these have changed over time.

Figure 5 shows the proportions of school pupils attending independent schools across the nine English regions. England as a whole had a proportion of just under 6% in 2023, down from just under 7% in 2016. The region with the lowest proportion is the North East followed by Yorkshire and the Humber, the East Midlands, the North West, the South West, the East of England, the South East and London. To see all lines again, click here.

There are at least two messages to take from this. First, England is an unbelievably stratified country, socioeconomically and geographically. (In the unlikely event that anyone tries to suggest otherwise then, after you've finished laughing, you could do worse that point them to Figure 5.) The second is that all regions have shown gradual relative decline in independent school attendance, though the trend has been flat in the last two or three years.

In our current low-growth, high-inflation economic environment, and with the distinct possibility that VAT will soon be levied on school fees, might pupils moving from the independent sector make up some of the anticipated decline in state schools rolls? Only time – and data – will tell.

(To share your views or questions, please write to us at [email protected]. And to keep track of more analyses like this one, sign up for our free monthly-ish newsletter. SchoolDash Insights subscribers can explore the school capacity and enrolment data further in the Pupils section. Non-subscribers are welcome to request a trial account or a demo. For help finding and choosing a school, see our service for families, The Schools Guide.)

Are independent schools losing their heads too?

7th February 2024 by Timo Hannay [link]

Last October, we took a look at headteacher turnover in English state schools. As in so many other areas, the pandemic had left its mark: turnover was lower than usual in 2020-21 and 2021-22 before bouncing back to historically high levels in 2022-23. Early signs were that the current school year is also on track to show unusually high turnover among heads.

Our method made use of the fact that the Department for Education (DfE) publishes the name of the headteacher (or equivalent) at every registered school in England. We have tracked this to assemble a history of the headship at each school over the last 8 years or so (filtering out trivial variations such as spelling corrections and new surnames that do not appear to indicate an actual change in school leadership).

The same data are available for independent schools, but we omitted them last time for two main reasons: (i) the patterns are different, so including them would have complicated the analysis, and (ii) the data for independent schools appears to be messier, with higher proportions of spurious entries (eg, supposed headteacher appointments that last only a week or two), so we wanted to take some time to check and tidy up things a bit first. This post looks into what the resulting data say about mainstream independent schools in England.

To summarise:

• Headteacher turnover at independent schools is less seasonal than at state schools. Historically, about 40% of new headteacher appointments at state schools are made in September. At independent schools it's more like 20-30% (as well as being more statistically volatile, probably just because of the smaller number of schools).
• While state schools saw a substantial decline in new headteacher appointments in 2020-21 (ie, the first full year of the COVID-19 pandemic), independent schools saw a large increase in the same year, taking them to around 40% above normal pre-pandemic levels. They then declined in 2021-22 before jumping up again in 2022-23. So far, 2023-24 looks like being another hyperactive year.
• The result is that, unlike state schools, independent schools have so far seen no years during or after the pandemic during which headteacher turnover has been lower than during a normal pre-pandemic year.
• Similar patterns are evident when the analysis is limited to schools that are members of the Independent Schools Council (ISC), which generally represents larger and more mainstream independent schools. So the effects described above do not appear to be due to a subset of unrepresentative 'outlier' schools.
• There has also been a big increase in new headteacher appointments made in quick succession. Historically, around 10-15% of appointments at independent schools occur within 2-12 months of a previous change. Since the pandemic, this has risen to well over 20%.

 

Getting a head

Figure 1 shows the number of new headteacher appointments each month over the last five academic years (yellow, green, blue, red and purple lines) and so far in 2023-24 (black line). For now, we will look at all schools together, and use cumulative data, which make aggregate annual differences easier to see. In 2018-19 and 2019-20 were relatively consistent, with around 270-310 new appointments a year. This shot up in 2020-21, particularly during the period from September 2020 to January 2021, which roughly corresponds to the period between the two main national lockdowns in England. Things then returned to more or less normal in 2021-22. But in 2022-23, roughly 18 months after the end of the last national lockdown, turnover increased again. Something similar – though if anything, more extreme – seems to be happening again during the current 2023-24 school year. Click here to see all years again.

(Use the menus below to switch between monthly or cumulative data, and between all schools, primary schools or secondary schools. Click on the figure legend to turn individual years on or off; double-click to see one year on its own.)

What's going on here? Perhaps these results are skewed by a small subset of unusual schools. After all, independent schools come in a wide variety, include some very specialist, even idiosyncratic institutions. To explore the possible effects of this on the results, Figure 2 shows the same analysis as above, but limited to members of the Independent Schools Council (ISC). They generally represent larger and more mainstream independent schools (claiming to account for roughly 50% of independent schools, and about 80% of independent school pupils, in England).

In short, the results are more or less the same: looking at cumulative data for all schools, 2018-19 and 2019-20 provide a reasonably stable pre-pandemic baseline, 2020-21 was much higher, 2021-22 was relatively calm again, 2022-23 rebounded upwards and 2023-24 is so far running hot too. Click here to see all years again.

(Use the menus below to switch between monthly or cumulative data, and between all schools, primary schools or secondary schools. Click on the figure legend to turn individual years on or off; double-click to see one year on its own.)

 

Head-scratcher

At the risk of getting technical, these trends are what data analysts sometimes refer to as "well weird". In other words, we have to be honest and declare that we don't fully understand what's going on, and any purported explanations would be little more than speculation.

But that's never stopped us before, so here goes. Perhaps that the period during the 2020 and 2021 lockdowns was a one-off event that provoked lots of independent school heads to move on or retire. Following a year or so of relative normality, the more recent increases in turnover could either be a very delayed effect of the pandemic or something new, perhaps brought about by current economic trends. On top of this, maybe head turnover at independent schools is just intrinsically more volatile. Probably it's a combination of such factors.

We also need to bear in mind that new headteacher appointments can reflect a wide variety of situations, including long-term maternity cover and management reorganisations as well, as conventional one-for-one changes of personnel. All that said, it's still surprising to see how different these results are to the same analysis for state schools.

One thing that is rather similar between the sectors is the recent rise in the proportions of headteacher changes that occur within 2-12 months of a previous change. Until the pandemic, these were typically in the range of 10-15% for independent school, but have since risen to well over 20%. The corresponding figures for state schools when we last looked were around 20% and 25%, respectively. So the recent rise in turnover has been accompanied by a disproportionate increase in short tenure. We don't yet know why this is, but plan to try and find out.

(We welcome your thoughts and feedback: please write to [email protected]. To hear about future analyses like this one, sign up for our free monthly-ish newsletter. SchoolDash Insights subscribers can explore the headteacher appointment data further in the Headteachers section. Non-subscribers can request a trial account or a demo.)