Performance Tracker 2023: Methodology

0. Cross-service analysis 

Public services spending, including estimates of the total spend on public services 

To estimate the real cost of public spending, we deflate government spending figures using the GDP deflators published by the Office for Budget Responsibility in the Economic and Fiscal Outlook from March 2023. ⁴² Office for Budget Responsibility, Economic and fiscal outlook – March 2023, 15 March 2023, https://obr.uk/efo/economic-and-fiscal-outlook-march-2023 To better reflect the underlying inflation conditions present in 2020/21, we estimate our own figures by generating a mid-point that averages across values from 2019/20 and 2021/22. We deflate spending figures in our financial analysis across all Performance Tracker chapters to 2023/24 prices.

In cases where we calculate real-terms changes in figures that relate to individuals – for example, wages or the adult social care means test – we use the consumer price index rather than the GDP deflator. The CPI that we use also comes from the OBR’s Economic and Fiscal Outlook from March 2023.

Change in demand for public services

General practice

To project likely growth in demand for general practice, we use analysis from The Health Foundation. Its main published analysis for ongoing demand, published in October 2021, ⁴³ Rocks S, Boccarini G, Charlesworth A and others, Health and social care funding projections 2021, The Health Foundation, 2021, https://www.health.org.uk/publications/health-and-social-care-funding-projections-2021 includes an estimate for how much primary care activity will need to increase to maintain standards, factoring in growing case complexity due to comorbidities. General practice excludes some services that are included in The Health Foundation’s measure of primary care but includes others (such as drugs dispensed in general practice) that it excludes. But we nonetheless assume that demand for genera practice services changes in the same way as demand for primary care.

To ensure comparability with the demand projections shown for other services (for which we do not include service-specific cost pressures or possible productivity gains), we only factor in increases in activity, rather than additional assumptions around changes to pay and productivity.

Hospitals

To project likely growth in demand in hospitals, we again draw on analysis from The Health Foundation. Its analysis provides an estimate of the rate of growth in activity, adjusted for morbidity, needed to meet growing demand for acute care, while maintaining its scope and quality. We assume that demand for acute and specialist trusts (our focus in this chapter) changes in the same way as The Health Foundation’s projection of demand for acute care.

To ensure comparability with the demand projections we show for other services – where we do not include service-specific cost pressures or possible productivity gains – we only factor in increases in activity.

The Health Foundation kindly provided us with a breakdown of its model to allow us to derive an overall estimate of acute care, based on a weighted average of elective, nonelective, A&E and outpatient activities.

Adult social care

For adult social care, we take the projected increase in demand from The Health Foundation’s REAL Centre, published in October 2021. This model incorporates several factors, including increases in pay and projected changes in productivity. We take only the increase in activity projected in the model, as the outlook for pay has changed since it was published.

In other services, we only factor in increases in demand, without incorporating above inflation cost pressures. This is because, for most services, we expect wages to broadly increase in line with economy-wide inflation (the GDP deflator, which will increase less quickly than consumer price inflation), which our spending projections already account for. But this is not the case for adult social care, where a substantial proportion of the workforce is paid the national living wage (NLW), which is set to increase much more quickly than economy-wide inflation. The impact of the NLW is explored in more detail in the adult social care section of this Methodology, below.

Children’s social care

To project demand for children’s social care, we break down children’s social care spending into three service categories based on the data returns that local authorities make to the DfE under Section 251 of the Apprenticeships, Skills, Children and Learning Act 2009. ⁴⁴ Education and Skills Funding Agency, ‘Section 251: 2021 to 2022’, 8 March 2021, last updated 6 July 2022,
https://www.gov.uk/government/publications/section-251-2021-to-2022 For each category, we make the assumptions about rates of growth set out in Table 10.1.

Table 10.1 Projected growth rates for children’s social care 

As 2020/21 was an unusual year due to the pandemic, we project demand from 2019/20 onwards. In common with the assumptions made for other service areas, to project demand growth we assume that the service remains as efficient as it was in 2019/20 and that the cost of providing each service grows in line with economy-wide inflation. If there are cost pressures beyond the projected increases in demand described above, then spending would have to rise faster. 

Schools 

To project how much schools would have to spend to meet increased demand, we separate primary and secondary schools because: 

• on average, the government spends slightly more on each secondary school pupil than on each primary school pupil 
• the DfE projects that the number of primary school pupils will fall over the period 2019/20–2024/25 while the number of secondary school pupils will increase. 

As 2020/21 was an unusual year, we base our projections on spending in 2019/20 (see Table 10.3). We multiply the 2019/20 level of spending per pupil in primary and secondary schools by expected growth in pupil numbers between 2019/20 and 2024/25 and add together the implied figures for spending on primary and secondary schools. We assume that the costs of the inputs used in providing school services rise in line with economy-wide inflation. 

Table 10.3 Projected growth rates for schools 

Criminal courts 

We project demand for the crown and magistrates’ courts separately. 

For the crown court, we calculate demand as the number of cases received each year, weighted by the average hearing time for cases completed in each year. We do this separately for cases that are ‘for trial’ and other cases (such as appeals and sentencing). We assume that: (i) longer hearing times are a result of cases being more complex, rather than the court operating inefficiently; and (ii) the cases received would have had similar hearing times to the ones disposed of, within case type (cases for trial and others), in the year in question. 

For magistrates’ courts, where the data we have is less detailed, we measure demand simply as the number of cases received each year. 

We weight magistrates’ and crown court demand to come to an overall measure of court demand. We do this using two components. First, we use the number of sitting days in the crown court and magistrates’ courts in 2018. ⁴⁵ Maynard P, Response to parliamentary question, UK Parliament, 10 June 2019, retrieved 27 October 2023, https://questions-statements.parliament.uk/written-questions/detail/2019-06-03/259164 
  Second, we use the average costs per sitting day in the crown court and magistrates’ courts, which the National Audit Office reported in 2016, as these are the latest available figures. ⁴⁶ Comptroller and Auditor General, Efficiency in the Criminal Justice System, Session 2015–16, HC 852, National Audit Office, 2016, p. 10, https://www.nao.org.uk/wp-content/uploads/2016/03/Efficiency-in-the-criminal-justice-system.pdf  This implies that 61% of court demand comes from the crown court and around 39% comes from the magistrates’ courts. We then project demand forward separately for the crown and magistrates’ courts. 

The main driver of our projection of court demand is the increase in police officers; the government met its commitment to increase officer numbers by 20,000 on top of 2018/19 figures by April 2023. ⁴⁷ Hannah G, ‘Police recruitment target hit, now to secure the benefits’, National Audit Office, 26 April 2023, https://www.nao.org.uk/insights/police-recruitment-target-hit-now-to-secure-the-benefits We assume that an increase in the number of officers means the police can charge more cases, because as it stands the number of charges is only a small fraction of total crimes reported. The number of charges per police officer has fallen steadily for several years, in part due to underfunding elsewhere in the criminal justice system. 

We assume that once officers are embedded the number of charges per officer will return to and stay at 2019/20 levels. However, we assume that there is a lag of three years between recruitment and a return to 2019/20 levels of charges, as this is the time it has taken between the start of the uplift programme and an increase in that indicator. We therefore project that charges per officer will return to 2019/20 levels in 2026/27, increasing uniformly between 2022/23 and 2026/27. 

In the magistrates’ courts, we assume that the least serious ‘summary’ cases are unaffected by the number of police officer charges as some of these are brought by non-police organisations and they are simple, routine offences. With all other cases, in both the crown and magistrates’ courts, increases occur in line with the lag described above. 

Prisons 

To project demand for prisons, we use the Ministry of Justice’s (MoJ) central estimate for prisoner numbers over the five years from 2022 to 2027, which was published in February 2023. ⁴⁸ Ministry of Justice, ‘Criminal court statistics quarterly: April to June 2023’, 28 September 2023, retrieved 27 October 2023, https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-april-to-june-2023

The MoJ’s central estimate is that the prisoner population will rise by 13.5% between March 2020 and March 2025 (and by 20.9% between March 2021 and March 2025). This projection incorporates the recruitment of the additional 20,000 police officers and the estimated impact of other policies, including: the provisions for increasing the release point for violent and sexual offenders sentenced to a standard determinate sentence of four to seven years; the statutory instrument to increase custodial sentences for serious offenders with a custodial sentence of seven years or more; the Serious Crime Act 2015; the Offensive Weapons Act 2019; and the Domestic Abuse Bill 2020. 

Average criminal justice system timeliness in days (Figure 0.1) 

Figures reported in this chart are median figures and refer to calendar years. This is also true of the total ‘offence to completion’ measure. As such, the individual ‘offence to charge’, ‘charge to first listing’ and ‘first listing to completion’ may not sum to the total ‘offence to completion’ figures. 

‘Offence to charge’, ‘Charge to first listing’ and ‘First listing to completion dates’ waiting times (in days) are taken from the MoJ’s quarterly criminal courts dataset. ⁵³ Ministry of Justice, ‘Criminal court statistics quarterly: April to June 2023’, 28 September 2023, retrieved 27 October 2023, https://www.gov.uk/government/statistics/criminal-court-statistics-quarterly-april-to-june-2023 
 

Capital spending index, public service departments, 2004/05–2022/23 (Figure 0.3) 

The primary problem in the creation of this chart is change in departmental structures and responsibilities. To circumvent this issue, we create a retrospective time series, using the most recent outturn as the baseline for the years 2018/19 to 2024/25 (the years for which there is data in the PESA tables). ⁵⁴ HM Treasury, ‘Public Expenditure Statistical Analyses 2023’, 19 July 2023, retrieved 27 October 2023, https://www.gov.uk/government/statistics/public-expenditure-statistical-analyses-2023 We then input the PESA results for previous years. We transform the last outturn year (2018/19) by the change in spending for the previous PESA outturn. We then replicate this method for all departments and for all years going back to 2004/05. 

Average annual real-terms change in planned capital spending at successive multi-year spending reviews (Figure 0.4) 

We deflate the capital department expenditure limit (CDEL) at each multi-year spending review by the most recent GDP deflator at the point when the spending review was published, for each department. We then calculate the average annual increase in real-terms CDEL. To account for changes in departmental responsibilities, we combined police and justice CDEL to make spending reviews comparable between 2002 and 2021. 

Change in median gross earnings of selected public sector professionals since 2009/10 (Figure 0.7) 

The ONS has changed how it classified professions twice since 2009/10. To ensure consistency we cross-referenced codes against volumes of employees to check similar numbers of staff were being assessed. The most affected data series was nursing professionals, for which we used data for health associate professionals for 2009/10, nursing and midwifery professionals for 2010/11–2019/20 and nursing professionals for 2020/21-2021/22. Additionally, figures for primary education teaching professionals include nursery staff up to 2020/21. 

Unlike other data series in the report, which were deflated using a smoothed GDP deflator, figures were deflated using the consumer price index as published by the OBR. 

Working days lost to strike action in the public sector over the previous 12 months (Figure 0.8) 

To calculate the line in this chart, we sum the total number of days lost to industrial action in the public sector in the previous 12 months of any given day. 

Average annual real-terms change in spending between 2021/22 and 2024/25 relative to demand under different inflation scenarios (Figure 0.10) 

For the nine services we cover in this report, we project how much money the public sector would have to spend to meet demand. To estimate the cost of doing this, we project growth in underlying demand for each service as described above. 

For each service we also project how spending is likely to evolve over the course of the spending review (up to 2024/25). The 2021 spending review did not provide budgets for particular public services, only government departments (with the exception of schools and the NHS, which have their own budget lines). For each service, we take the most relevant department’s settlement, implicitly assuming that all spending within those budgets will increase at the same rate. 

This means that we assume that spending on hospitals will increase in line with NHS spending, spending on courts and prisons will increase in line with MoJ spending, spending on the police will increase in line with Home Office spending, and schools’ spending will increase in line with the specific school funding line in the spending review. For the three local government services (adult social care, children’s social care and neighbourhood services), we take the government’s projections for local authority spending power, which incorporate changes to grants and assumed increases in local taxes (council tax and business rates). For GPs, we assume that spending will increase in line with the amounts laid out in the 2019/20–2023/24 GP contract; for 2024/25, we assume that spending on GPs increase in line with the average change in the GP contract between 2021/22 and 2023/24. We also adjust these spending amounts for recent government announcements. For example, we include the extra £515m that the Home Office announced it would provide to police in 2024/25 to pay for the recently agreed pay deal. ⁵⁵ Braverman S, Statement to parliament, UK Parliament, 13 July 2023, retrieved 27 October 2023, https://questions-statements.parliament.uk/written-statements/detail/2023-07-13/hcws945  

Average annual real-terms change in spending planned at successive multi-year spending reviews (Figure 0.11) 

We replicated the methodology that we used for CDEL in Figure 0.4 for resource departmental expenditure limit (RDEL). 

Average annual real-terms change in spending between 2024/25 and 2027/28 under current government plans relative to demand (Figure 0.12). We first increased total RDEL across government by 1% in real terms for each year between 2024/25 and 2027/28. From that we subtracted protected spending. 

Protected spending is spending on health, defence and foreign aid. To calculate health spending, we increased spending for the NHS (in this case, general practice and hospitals) by 3.6% in real terms for each year of this future spending review as this is the amount that the Institute for Fiscal Studies (IFS) estimates that it would require to meet the commitments laid out in the NHS Long Term Workforce Plan. ⁵⁶ Warner M and Zaranko B, ‘Implications of the NHS workforce plan’, Green Budget 2023, Ch. 8, Institute for Fiscal Studies, 30 August 2023, retrieved 27 October 2023, https://ifs.org.uk/publications/implications-nhs-workforce-plan For defence and foreign aid, we increased the budgets of the Ministry of Defence (MoD) and the Foreign, Commonwealth and Development Office (FCDO) by the real-terms GDP forecast from annex A of the OBR’s March 2023 Economic and Fiscal Outlook. We subtracted these totals from the total RDEL increases previously calculated to work out how unprotected RDEL would change. 

We assumed that all unprotected departments would change in line with the change in unprotected RDEL. From there, we offset the spending increases by demand for each service as has already been described in the methodology of Figure 0.10. 

1. General practice 

Change in GP pay (Figure 1.2) 

Forecast increases in salaried GPs pay in 2022/23 and 2023/24 is based on pay settlements for those years. It is not possible to make the same calculation for GP partners. Their pay is variable from year to year because they take pay as drawings from GP funding after other expenses have been incurred. 

Size of job groups within the primary care workforce (FTE) (Figure 1.5) The number of GPs used in this chart is the ‘All qualified permanent GPs (excludes GPs in training grades and locums)’ line from the GP Workforce bulletin tables. 

GPs leaving the NHS, by age group (Figure 1.8) 

The ‘GP workforce – joiners and leavers’ dataset gives leaver rates and total leaver numbers for GPs under 30 and then in five-year age bands until 70, at which point all GPs older than that are grouped together. To calculate wider age bands, we calculated the implicit number of GPs in each age band (total leavers/leaver rate), then summed the leavers in those bands and divided that by the sum of the implied number of GPs in that band. 

Appointments in general practice (Figure 1.14) 

The NHS changed how it collects information on the number and type of appointments in primary care in October 2018. There is an overall time series going back to November 2017, but granular daily counts of appointments are only available from December 2018. 

Appointments delivered per FTE (Figure 1.15) 

This uses data from the ‘Appointments in general practice’ dataset and the GP Workforce bulletin. It takes the number of appointments delivered by each staff group for a particular month and divides them by the number of staff members in that staff group in that month. To come to the rolling average, we take the average number of appointments conducted per month in the previous 12 months and the average number of staff in each staff group in that time. For this chart we use the number of fully qualified permanent GPs as the denominator for GP efficiency. ‘Other practice staff’ is the sum of ‘All DPC staff’ and ‘All nurses’. All calculations are done on an FTE basis. 

Specific and acute referrals as a proportion of GP appointments (Figure 1.17) 

We calculate this by dividing the number of specific and acute GP referrals in a given month by the number of attended GP appointments carried out in that month. Before July 2021, the ‘Appointments in general practice’ dataset provided information on the number of attended GP appointments. This stopped from July 2021 onwards. Instead, to calculate the number of attended GP appointments, we take the total number of attended appointments across all of general practice in a given month (as outlined in the ‘Appointments in general practice’ dataset) and multiply that by the percentage of appointments that GPs carried out (using the SDS Role Group categorisation, rather than HCP categorisation). This step requires us to assume that the attendance rate of GP appointments is the same as the attendance rate of all primary care appointments, an assumption that is unlikely to be met in any month but which will be close enough to make this analysis meaningful. For the monthly number of referrals, we use the ‘GP referrals made (specific acute)’ data from the ‘Monthly referral return’ dataset.

Change in GP numbers and registered patients (Figure 1.19) 

As with Figure 1.5, the change in the number of GPs refers to the percentage change in the number of ‘All qualified permanent GPs (excludes GPs in training grades and locums)’. The change in the number of patients comes from the ‘Total number of patients’ line in Table 5 of the GP Workforce bulletin tables. The starting date for both of these time series is September 2015 because this is when the time series starts in that dataset. 

Patient to GP ratio by decile of deprivation (Figure 1.20) 

For this we use a snapshot of the number of patients and the number of fully qualified, permanent GPs in March 2016 and March 2023, to account for any seasonality in either number. To find how many GPs and patients there are in each decile of deprivation, we use the practice postcode to place a practice into a lower layer super output area (LSOA). We then use the English indices of multiple deprivation at a LSOA level to assign a decile of deprivation to each practice. From there it is possible to sum both the patients and the number of fully qualified permanent GPs and then divide the former by the latter to come to a patient–GP ratio. 

2. Hospitals 

Seasonal trend lines (multiple charts) 

For multiple charts (Figures: 2.15, 2.16, 2.23, 2.25, and 2.27) in this chapter we use seasonal trend lines to show the likely path of activity or performance for a given indicator in the absence of Covid. These are calculated in multiple stages. First, we calculate the monthly total over the previous 12 months for every month of data before February 2020 (the last month of data that is mostly unaffected by Covid). Second, we calculate the average monthly change in activity between the first 12-month average and February 2020. Third, we then divide the monthly actual by the 12-month average for that month to show the performance against the average. Fourth, we then average these monthly performances to create an average multiplier vs the 12-month average for every month of the year. Fifth, we then strip out the trend growth by normalising these monthly multiplies, by dividing each one by the average multiple across all 12 months. 

We then start the trend line in March 2020. We calculate the trend by multiplying the 12-month average activity to February 2020 by the average monthly change for that activity type, risen to the power of the number of months past February 2020 (for example, for May 2020, the power would be 3). We then give that trend change a seasonality by multiplying it by the normalised multiplier for the relevant month. 

Hospitals beds occupied by Covid patients (Figure 2.3) 

This shows a rolling seven-day average. For the value shown on a given day, it is the average number of beds occupied by Covid patients in the previous seven days. 

Change in doctor and nurse numbers (Figure 2.4) 

Nursing numbers include adult and children’s nurses who work in hospitals, but do not include community nurses from the ‘NHS workforce statistics’ dataset. Doctors numbers include all those in the HCHS dataset. 

Net nurse & health visitors and doctors joiners and leavers (Figures 2.5 and 2.6) 

These charts show headcount, rather than FTE. The ‘EEA’ category is the sum of the European Union and the European Economic Area totals. The net number is calculated by subtracting the number of leavers from the number of joiners. A net negative number therefore shows there were more leavers than joiners in the previous 12 months. 

Rolling average of NHS vacancy rates, by type of role (Figure 2.7) 

The vacancies dataset gives the number of vacancies and the proportion of vacancies at the end of each quarter. To smooth out seasonal variation, we calculated a rolling 12-month average. We did this by first calculating the implied headcount for a given quarter, which is the number of vacancies divided by the proportion of vacancies. For the full 12-month average, we then average the number of vacancies for the previous four quarters and average the implied headcount for the previous four quarters, then divide the former by the latter. 

Rolling average of the hospital and community workforce resigning in the previous 12 months, by reason (Figure 2.8) 

For this chart, we group some of the voluntary resignation categories from the ‘NHS workforce, reasons for leaving’ dataset to make the chart easier to read. The groupings are as follows: 

• Work–life balance: this includes just the ‘work–life balance’ category from the original dataset, as this is the category that we want to highlight. 
• Working conditions: this includes the ‘better reward package’ and ‘incompatible working relationships’ categories. 
• Career development: this includes the ‘lack of opportunities’, ‘promotion’ and ‘to undertake further education and training’ categories. 
• Family/external: this includes the ‘adult dependants’, ‘child dependant’, ‘health’ and ‘relocation’ categories. 
• Other/not known: this includes only the ‘other/not known’ category. 

Rolling average of NHS staff staying in post over the previous 12 months (Figure 2.9) 

In this chart, the ‘Nurses and health visitors’ and the ‘Total’ are taken straight from the ‘HCHS staff in HCHS trusts – turnover tables’. For HCHS doctors we exclude foundation year 1 and 2 doctors because they frequently leave roles as part of their rotations. 

Proportion of NHS staff days lost to mental ill health (Figure 2.11)

This sums the total number of days lost to the reasons classified as ‘S10 Anxiety/ stress/depression/other psychiatric illnesses’ in a given calendar year and divides them by the total number of available staff days in that financial year. 

NHS staff absence, by reason (Figure 2.12) 

‘Mental health’ refers to the reason classified as ‘S10 Anxiety/stress/depression/other psychiatric illnesses’. ‘Respiratory and other infectious diseases’ is the sum of: ‘S13 Cold Cough Flu – Influenza’, ‘S15 Chest & respiratory problems’ and ‘S27 Infectious diseases’. ‘Other’ is the sum of all the remaining reasons for absence (with the exception of unknown, which is captured in its own category). 

Real-terms change in NHS staff earnings, by staff group (Figure 2.13) ‘Consultants’, ‘Nurses & health visitors’ and ‘All staff’ have been taken directly from the dataset. ‘Ambulance staff’ is a combination of the staff groups ‘Ambulance staff’ and ‘Support to ambulance staff’ from the dataset, weighted by the size of the workforce in each month. We combined these staff groups because some staff were reallocated from the former to the latter in April 2019, which made it appear as though ambulance staff pay rose dramatically. For ‘Junior doctors’, we combined the ‘Foundation Doctor Year 1’, ‘Foundation Doctor Year 2’, ‘Core training’ and ‘Specialty registrar’ staff groups, weighted by size of the staff groups in a given month. We chose these four because these are the staff groups that the BMA uses as a proxy for junior doctors. We deflated staff earnings by CPI, which is a more realistic index of the costs that an individual faces than the GDP deflator. 

Ambulance arrivals resulting in a handover delay of 30+ minutes during winter peaks (Figure 2.24) 

This is a rolling seven-day average of ambulance handovers resulting in a delay of 30- 60 minutes, >60 minutes, and ambulance arrivals. The percentage is the sum of the first two divided by the third. 

Hospital activity in the previous 12 months compared to the 12 months to February 2020 (Figure 2.28) 

For each of these types of activity we calculated the total activity in the previous 12 months for every month to August 2023 and then divided that by the activity in the 12 months to February 2020. 

NHS beds (Figure 2.29) 

This shows the average number of each category of bed across the four quarters worth of data for a given year. For 2023/24, there is only one quarter’s worth of data so this is the total for that financial year. The number of ‘general and acute (G&A) overnight’ and ‘day only’ beds are given in the dataset, as is the total number of beds. ‘Other’ is the difference between the total number of beds and the sum of G&A overnight and day only beds. 

Attendances and admissions at major A&E departments in the previous 12 months (Figure 2.32) 

This shows the average number of type-1 A&E attendances and admissions in the previous 12 months for a given month, divided by the average number of the respective metric in the 12 months to February 2020. 

Change in the number of managers per FTE NHS staff member (Figure 2.33) 

We calculated the number of senior managers and managers per NHS staff on an FTE basis for every month of this time series. We then calculated how this changed over the course of the time series. 

Gross capital formation in health care as a percentage of GDP, by OECD country (Figure 2.34) 

The weighted OECD average is calculated by first calculating the total spent on gross capital formation in every year by multiplying the percentage given in the dataset by the GDP number for the relevant year. This is then summed and divided by the sum of GDP for every country that has a data point in that year. Any country that does not have a data point in a given year is excluded from the analysis for that year. 

Hospital diagnostic equipment per million population, by OECD country (Table 1) 

The OECD average in this table is calculated as an average of the OECD countries in this table, weighted by the size of their population. 

Diagnostic tests conducted, March to July 2023 (Figure 2.36)

Diagnostic tests conducted in ‘non-community diagnostic centre locations’ is calculated by subtracting the number of tests conducted in community diagnostic centres from the total number of tests conducted in a given month in hospital and community settings. 

Procedures conducted by independent providers, by specialty (Figure 2.38) 

This is the sum of all admitted and non-admitted pathways by independent service providers in 2022/23 divided by the sum of all completed pathways in that year. 

3. Adult social care 

Spending on adult social care (Figure 3.1) 

The data point for 2022/23 is the actual spend on adult social care, as outlined in Table 5 of Appendix B of the ‘Adult social care activity and finance report, England – 2022-23’ (ASCAFR) dataset. We use this time series rather than the spend on adult social care as outlined in the local authority revenue outturn dataset because it captures a wider range of spending on the service than just local authority related spending. To calculate the forecast spend in 2023/24, we average out two methods. For both methods, we use the ‘Revenue Account Budget’ (RA) – which is the amount that local authorities forecast that they will spend in a given financial year. While this means that the forecast amount spent in 2023/24 will not be completely accurate, we feel that it is a fair assumption as the amount that local authorities spent on adult social care accounts for approximately 90% of the amount spent on adult social care in the ASCAFR dataset. 

For the first method, we use 2021/22 ASCAFR spending as the baseline, and we uplift that amount by the percentage change between that year’s ‘Revenue outturn’ (RO) and the 2022/23 and 2023/24 RAs. 

There has subsequently been another ASCAFR release for 2022/23, but no RO for 2022/23. The ASCAFR spending came in slightly higher than the first method forecast for 2022/23. We assumed that not all of that spending will be exceptional and will likely imply higher than expected spending in 2023/24, meaning that the first method would likely underestimate 2023/24 spending. 

We then used the second method, which was to apply the percentage uplift between the 2022/23 RA and the 2023/24 RA to the 2022/23 ASCAFR outturn. But we then assumed that this would overstate the amount spent on the service in 2023/24. We therefore decided to average out the 2023/24 forecasts for both methods to try and balance the respective under and over forecast. 

Cost pressures on 2023/24 uplift in spending on adult social care (Figure 3.3) 

The total spending increase for 2023/24 comes from the calculation described in the methodology for Figure 3.1. To calculate the impact of the national living wage (NLW), we assume that 70% of the total spent on the sector in 2022/23 goes on wages – a figure that comes from interviews with reliable stakeholders. We then assume that 65% of that total is spent on staff who receive the NLW, an assumption that comes from the Care Policy Evaluation Centre’s work. We then increase this amount by the NLW increase for 2023/24 (9.7%). The difference between the 2022/23 and the 2023/24 NLW is the impact of the increase of the NLW. 

For the impact of ‘other wages’, we take the remainder of the wage bill that is not related to the NLW and increase it by 6%, which is the Bank of England’s assumption for economy-wide wage growth for this financial year. 

To calculate non-wage inflation, we take the remaining spending on adult social care after wages have been taken out and increase it by the GDP deflator forecast for 2023/24 from the OBR’s Economic and Fiscal Outlook from March 2023. 

To calculate the demographic demand pressure, we take the total spending for 2022/23 and increase it by the forecast increase in the size of the over-65 population in England for the year, then subtract the former from the latter. This likely underestimates the true impact of demographic demand on the service as demand from working-age adults is rising more quickly and is more costly than increases in demand from the over-65 population. It is, however, difficult to estimate the demand for working-age adults so we have decided to provide a more cautious estimate. 

Adult population providing 20+ hours per week of unpaid care by local authority deprivation (Figure 3.10) 

We calculated the proportion of the population of each upper- and single-tier local authority that provides 20–50 hours and 50+ hours of unpaid care per week and then summed them to come to a total proportion of the population that provides 20+ hours of care per week. We replicated the calculation for 2011 and 2021, with different populations depending on the year. We then matched these percentages to the level of deprivation in the 2019 English indices of multiple deprivation. 

Change in proportion of the population accessing long-term support since 2014/15, by age group (Figure 3.13) 

We took the total number of people receiving long-term care at the end of the financial year and divided this by the relevant population group for each year to come to a ‘per-capita number of people in long-term care’. We then calculated the percentage change for working-age adults and adults over 65 to show the change compared to 2014/15. 

Change in the means test for publicly-funded care, compared to inflation (Figure 3.15) 

To calculate the level the means test would have been if it had risen in line with inflation, we multiplied the base year (2009/10) by the consumer price index for a given year to show how it would have progressed. 

4. Children's social care 

Change in local authority spending on children’s social care in England since 2009/10 (real terms) (Figure 4.1) 

The data point for 2021/22 is the actual spend on children’s social care, as outlined in the DfE’s ‘Local authority and school expenditure 2021 to 2022’ dataset. We use this time series rather than the spending on children’s social care as outlined in the local authority revenue outturn dataset because it captures a wider range of spending 

on the service than just local authority related spending. To calculate the forecast spend in 2022/23, we calculate the change in spending between the amount spent on children’s social care in ‘Revenue outturn social care and public health services (RO3) 2021 to 2022’ and the amount that local authorities were forecast to spend on children’s social care in 2022/23 and 2023/24, as outlined in the ‘Revenue account budget 2022 to 2023’ document. We then take this percentage change and apply it to the amount spent in 2021/22 from the DfE figures. While this means that the forecast amount spent in 2022/23 and 2023/24 will not be completely accurate, we feel that it is a fair assumption as the amount that local authorities spent on children’s social care in 2021/22 (£10.4bn) accounts for approximately 99% of the £10.5bn spent on children’s social care in the DfE dataset. 

Children entering care who are unaccompanied asylum-seeking children, 2009/10–2021/22 (Figure 4.7) 

For a given year-end data point, we have taken the number of unaccompanied asylum- seeking (UAS) children that become ‘looked after children’, released via an FOI request, as a percentage of the total number of new looked after children from the DfE’s dataset, ‘Children looked after in England including adoptions: reporting year 2022’. 

Number and proportion of children in care who are unaccompanied asylum seekers by local authority, 31 March 2022 (Figure 8) 

The proportion of children in care who are unaccompanied asylum-seekers has been calculated by taking the total number of asylum-seeking children who are looked after children as a proportion of the total number of looked after children for a given local authority. This was then plotted against the total number of UAS children for each local authority that had complete data. Fourteen local authorities lacked data on the number of UAS children and an additional local authority did not report a figure for the total number of children in care. Each of these local authorities was excluded from this analysis. 

5. Neighbourhood services 

Local authority spending, 2009/10–2021/22 (2023/24 prices) (Figure 5.1) 

Figures have been deflated using the smoothed deflator described at the start of the Methodology chapter. 

Local authority spending on neighbourhood services 2009/10–2021/22 (2023/24 prices) (Figure 5.2) 

The data for 2009/10 is inflated due to Revenue expenditure funded from capital by statute (RECS) which across all local authority expenditure for the year amounted to £1.8bn. It is not possible to remove this from individual service lines. For further details see Table 2 ‘Revenue expenditure and financing 2008-09 and 2009-10’ in Local Authority Revenue Expenditure and Financing England 2009-10 Final Outturn (revised) 27 January 2011 available at: https://assets.publishing.service.gov.uk/media/5a79a57940f0b63d72fc7676/1826743.pdf 

Median un-ringfenced reserves as a percentage of service expenditure, by local authority type, 2017/18–2021/22 (Figure 5.3) 

Reserves analysis has been carried out on a new dataset released by DLUHC: ‘Local authority general fund earmarked and unallocated reserve levels, 2017–18 to 2021–22’. 

These figures differ from DLUHC revenue expenditure and financing revenue outturn data, which represents the unadjusted reserves position. The department has explained that “unadjusted reserves figures are higher than normal for many authorities in 31 March 2021 and 31 March 2022”. ⁶⁰ Department for Levelling Up, Housing and Communities: Local authority general fund earmarked and unallocated reserve levels, 2017-18 to 2021-22, Note 2, 18 May 2023, retrieved 19 October 2023, available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1173145/Local_Authority_Reserves_England_-_2018-2022… This was due to emergency government support used to compensate for lower business rates income due to the Covid-19 business rates reliefs (expanded retail relief and Covid additional retail relief). ⁶¹ Department for Levelling Up, Housing and Communities: Local authority general fund earmarked and unallocated reserve levels, 2017-18 to 2021-22, Note 2, 18 May 2023, retrieved 19 October 2023, available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1173145/Local_Authority_Reserves_England_-_2018-2022… As the department explains: “Accountancy regulations require that this grant income is included in local authorities’ revenue accounts. These are shown in ‘other earmarked reserves’ until the following year when they compensate for what would otherwise be lower retained business rates income.” ⁶² Department for Levelling Up, Housing and Communities: Local authority general fund earmarked and unallocated reserve levels, 2017-18 to 2021-22, Note 2, 18 May 2023, retrieved 19 October 2023, available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1173145/Local_Authority_Reserves_England_-_2018-2022…

The figures reported in this new data series have been adjusted by the department to remove the distortionary effects on reserves data linked to the timing of emergency Covid grants including business rate reliefs. For further details see https://www.gov.uk/government/publications/local-authority-general-fund-earmarked-and-unallocated-reserve-levels-2017-18-to-2021-22

6. Schools 

Per-pupil funding, 2010/11–2023/24 (2023/24 prices) (Figure 6.2)

Figures have been deflated using the smoothed deflator described at the start of the Methodology chapter. 

Median teacher pay by role, 2010–2022 (2022 prices) (Figure 6.7)

Figures have been deflated using CPI figures from the spring budget, as a better estimate of the cost pressures teachers face than the GDP deflator. 

Attainment at the end of primary and secondary school in state-funded schools, 2010–2023 (Figure 6.11) 

All figures include those not in mainstream education. 

Primary: KS2 assessments were reformed between 2015 and 2016, and did not take place in 2020 and 2021. 

Secondary: between 2013 and 2014 a number of changes were made to secondary qualifications. In 2017 reformed English and maths GCSEs were awarded for the first time; in 2020 and 2021 GCSE results were awarded on the basis of centre-/teacher- assessment rather than external assessment; in 2022 GCSE results were set between pre-pandemic, 2019 levels and 2021 levels; in 2023 grades were allowed to return to pre-pandemic levels, with some grading protections to ensure they were not below 2019 levels. 

7. Police 

Spending on police, 2009/10–2022/23 (2023/24 prices) (Figure 7.1) 

To project spending for each year between 2023/24 and 2024/25, the previous year’s spending total is multiplied by the planned percentage increase in Home Office spending between the two years. These figures can be found in Table 2.1 of the autumn statement. ⁶⁴ HM Treasury, Autumn Statement 2022, CP 751, The Stationery Office, November 2022, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1118429/CCS1022065440-001_SECURE_HMT_Autumn_Statement…; This step is repeated for each year in the forecast. 

Figures are deflated using a smoothed GDP deflator, which arrives at a figure for 2020/21 by averaging figures from 2019/20 and 2021/22. 

Victim-reported crime methodology change (Figures 7.2 and 7.3) 

The Crime Survey in England and Wales (CSEW) suspended face-to-face interviews on 17 March 2020 due to the pandemic. From the year ending June 2020, data tables include data from the Telephone-operated Crime Survey in England and Wales (TCSEW). These do not include crimes experienced by children aged 10 to 15 years. Further details are outlined in ONS, ‘User guide to crime statistics for England and Wales: measuring crime during the coronavirus (COVID-19) pandemic’, 21 July 2022, retrieved 7 October 2022, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/methodologies/userguidetocrimestatisticsforenglandandwalesmeasuringcrimeduringthe…;

The CSEW figures used in these charts are taken from ‘Worksheet A1’ of ONS, ‘Crime in England and Wales: Appendix tables’. The figures for 2022/23 are comparable with the entirety of the time series included within this worksheet (that is, comparable with figures before the methodology change). It would not, however, be appropriate to draw comparisons between this worksheet and the corresponding worksheet in previous editions of the dataset published prior to the methodology change. 

Charges/summonses recorded by police forces, 2009/10–2022/23 (Figure 7.4) 

Charge/summons figures are taken from two sources. Figures for 2009/10 to 2017/18 are taken from Home Office, ‘Crime outcomes in England and Wales’ (Table B2), 2022/23. Figures covering 2018/19 to 2022/23 are taken from the Home Office’s ‘Police recorded crime and outcomes data tables’. Table B2 in the former source stopped publishing figures for all forces from 2018/19, and instead excluded Greater Manchester and Devon & Cornwall police forces. This was due to the desire to compare the numbers between like-for-like forces from this date onwards, since Greater Manchester and Devon & Cornwall experienced issues submitting figures in 2019/20 and 2022/23 respectively. Using the latter source allowed us to include what figures were submitted by these forces in our chart. We have highlighted these issues in the chart notes. 

This poses a potential problem when we claim that the increase in the number of charges in 2022/23 is “the first increase… since 2013/14”. While the data does not definitively support this claim (given the missing Greater Manchester data in 2019/20) we feel it is a fair one to make, since if this claim were not true (that is, if Greater Manchester recorded enough charges in 2019/20 to make overall levels at least equal to 2018/19 figures), it would have had to record double the number of charges it made in 2018/19. This would be extraordinary in itself, but also contrary to the trend in Greater Manchester’s charges between 2012/13 and 2020/21. 

Public perception that local police are doing a good or excellent job, 2009/10–2021/22 (Figure 7.7) 

Supplementary tables were not published as part of the CSEW in 2020/21 or 2021/22, as the new telephone survey limited time and questionnaire length. As a result, not all of the usual questions were asked to all participants. Data for a similar question, on rating the local police, is available for each quarter, however. The 2022 figure represents the average of responses from each quarter. 

8. Criminal courts 

Spending on HMCTS, 2010/11–2022/23 (2023/24 prices) (Figure 8.1) 

To project spending for each year between 2023/34 and 2024/25, the previous year’s spending total is multiplied by the planned percentage increase in justice spending between the two years. These figures can be found in Table 2.1 of the autumn statement. ⁶⁶ HM Treasury, Autumn Statement 2022, CP 751, The Stationery Office, November 2022, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1118429/CCS1022065440-001_SECURE_HMT_Autumn_Statement… This step is repeated for each year in the forecast. 

Figures are deflated using a smoothed GDP deflator, as discussed at the start of this chapter. 

Case backlog in the crown court, Q1 2010 to Q2 2023 (Figure 8.9) 

The latest official statistics for the backlog in the criminal courts are taken from the Quarterly Criminal Court Statistics up to March 2022. We calculate a backlog adjusted for complexity in three stages: 

• We calculate the number of jury and non-jury disposals that are missing by assuming that the share of cases coming into the crown court since March 2020 that end up as jury trials is the same as pre-Covid. The ‘missing’ cases are then the gap between those assumed to be entering the courts system and those that are completed each quarter. 
• We treat jury trials and other cases separately. We multiply the ‘missing’ number of both by [share of total hearing time]/[share of total cases] to get a complexity- weighted increase in the backlog. 
• An ‘ordinary’ backlog is more complex than the average of cases processed (specifically, more cases that will end up as a jury trial), so to adjust this number to be consistent with the pre-Covid backlog we multiply it by [average hearing time of backlog case mix]/[average hearing time of all cases]. 

9.Prisons 

Spending on prisons, 2009/10–2021/22 (2023/24 prices) (Figure 9.1) 

To project spending for each year between 2022/23 and 2024/25, the previous year’s spending total is multiplied by the planned percentage increase in justice spending between the two years. These figures can be found in Table 2.1 of the autumn statement. ⁶⁸ HM Treasury, Autumn Statement 2022, CP 751, The Stationery Office, November 2022, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1118429/CCS1022065440-001_SECURE_HMT_Autumn_Statement… This step is repeated for each year in the forecast. 

Figures are deflated using a smoothed GDP deflator, as discussed at the start of this chapter. 

Prison population and capacity, 2011–2023 (Figure 9.3) 

The monthly data releases cited give a number of figures, including: 

• In Use CNA: the sum total of certified normal accommodation in all establishments minus a) (normally) cells in punishment or segregation units and health care cells or rooms in training prisons, and b) those places not available for immediate use 
• Operational capacity: the total number of prisoners that an establishment can hold taking into account control, security and the proper operation of the planned regime. 
• Useable operational capacity: operational capacity minus an ‘operating margin’ that reflects the constraints imposed by the need to appropriately accommodate different classes of prisoner (for example, by sex, age, conviction status, single cell risk assessment, geographic distribution and security category). 

In this graphic, the ‘maximum capacity’ ribbon represents the space between the latter two figures. In any given monthly data release, either the total operational capacity or useable capacity is given, along with the useable capacity margin in the notes. In cases where the total operational capacity is published, the useable figure is calculated by subtracting the margin from the total. In cases where the useable capacity is published, the total figure is calculated by adding this figure to the published margin.