Methodology

Methodology for 2026 to 2027 PGITT trainee need and drivers analysis

• Published: 23 April 2026

Background

This document provides technical details and information to accompany the ‘Calculation of 2026-27 postgraduate initial teacher training (ITT) trainee need and related data’ and ‘Calculation of drivers of 2026-27 postgraduate ITT trainee need’ Excel workbooks. This includes the scope of the calculations, key terms, the calculation processes, and the assumptions used. 

Publication description

The publication provides the teacher demand trajectory, and the number of trainees needed to start postgraduate initial teacher training (PGITT) in 2026/27, as estimated using the teacher workforce model (TWM). 

The teacher demand trajectory is an estimate of future teacher demand and accounts for projections of future pupil numbers. The trajectory assumes that the current number of teachers is sufficient to meet current teacher demand. Going forward, the trajectory assumes that increased numbers of pupils increases teacher demand, with some of the additional demand being met through schools allowing pupil to teacher ratios (and class sizes) to increase. The opposite is true if pupil numbers are projected to fall. 

The publication also covers PGITT trainee need, this being the number of trainees needed for PGITT in 2026/27 to meet the teacher demand of 2027/28, i.e. the teacher demand trajectory is used to calculate future PGITT need.

This trainee need accounts for all routes into and out of the teacher workforce. These include leavers, losses due to teachers reducing their individual working hours between years, and all forms of wider recruitment. Wider recruitment may include undergraduate higher educational institution (HEI) courses, assessment only (AO), the teacher degree apprenticeship route (a new salaried route introduced in 2025/26 academic year), returners, new to the state-funded sector entrants, and newly qualified entrants that defer entry into the profession (deferrers).

Changes since the last release

The TWM is updated annually to reflect the latest data, assumptions, and policy changes. For the 2026 publication, analytical methodology has largely remained unchanged, but several presentational changes have been made this year. These include increasing the quantity and scale of the data published and presenting it using a more user-friendly and interactive format.

To ensure that these expanded data may be published within this year’s publication, it has been collated and published within Excel workbooks this year (‘Calculation of 2026-27 postgraduate initial teacher training (PGITT) trainee need and related data’ and ‘Calculation of drivers of 2026-27 postgraduate ITT trainee need’). Such data may not have been published in previous years.

Specific changes to the publication this year include:

• The ‘teacher demand trajectory’ is given greater prominence within the publication this year rather than PGITT.[1] Doing so will help readers to understand the ultimate purpose and driver of this year’s PGITT recruitment needs, and what the school system requires in terms of teacher workforce as pupil demographics shift.  
• To further assist this process, a new interactive Teacher workforce analysis dashboard (opens in new tab) presents a range of metrics including the teacher demand trajectory, PGITT trainee need, inflow and outflow trajectories etc. The publication of this data, and the new ‘drivers analysis’ helps to explain how and why PGITT trainee need has changed in the way it has this year.
• Note - the teacher demand trajectories represent teachers needed in service based on the estimated number of pupils over time, they are not forecasted workforce size outcomes. Actual workforce levels will depend upon recruitment, retention, and movements into and out of the state‑funded sector (among other factors). The trajectory is calculated using the same high-level methodology as that used to calculate teacher demand figures that were published in previous years.

Footnotes:

[1] The teacher demand trajectory was published in previous years as part of the ‘Calculations of 2025-26 postgraduate initial teacher training (PGITT) targets’ workbook within the PGITT targets publication.

Key terms

Term	Description
AO	Assessment only Teachers who have gained qualified teacher status (QTS) by the assessment only route to QTS rather than by initial teacher training.
Entrant	A teacher recorded in the school workforce census in a given year that was not part of the school workforce (as covered by the census) in the previous year are said to have ‘entered service’. Types of entrants include newly qualified entrants, new to state-funded, and returners.
Flows	The flows into and out of the teacher workforce. For example, one inflow are those teachers which are newly qualified that enter service immediately after completing initial teacher training. An example outflow are those teachers leaving service via retirement. Both inflows and outflows shape the overall number of teachers in service.
FTE	Full-time equivalent A teacher who is employed full-time is considered as being 1.0 FTE. By contrast, part-time teachers will have an FTE value of less than 1.0. Although teacher numbers are in FTE format, initial teacher training trainee need is in the format of headcount as one trainee is equivalent to one individual person.
HEI	Higher educational institution
HPITT	High potential initial teacher training
ITT	Initial teacher training
Leaver	A teacher that was not recorded in the school workforce census in a given year but was part of the school workforce in the previous year are said to have ‘left service’. For example, a leaver in 2024/25 would have been included in the census in 2023/24 but not included in 2024/25.
NQE	Newly qualified entrant  A teacher who has entered service in a state-funded primary or secondary school in England as recorded in the school workforce census in the months immediately after they completed initial teacher training. For example, a NQE in 2024/25 would be assumed to have gained qualified teacher status in 2023/24, entering service as a qualified teacher at the start of the 2024/25 academic year.
NTSF	New to state-funded sector entrant A teacher who has entered service in a state-funded primary or secondary school in England having not been employed as a regular teacher in the sector previously (as recorded in the school workforce census). This group will include some newly qualified entrants that deferred their entry into service by 4 to 16 months.  For example, a deferrer in 2024/25 would have successfully completed their ITT course in 2022/23 but only entered service in 2024/25. They would not have been present in the 2023/24 census.
PGITT	Postgraduate initial teacher training
PSHE	Personal, Social, Health, and Economic education
PTR	Pupil teacher ratio The ratio of pupils to teachers (both in full-time equivalent terms).
QTS	Qualified teacher status
Returner	A teacher who has entered service in a state-funded primary or secondary school in England having been employed as a regular teacher in the sector previously (as recorded in the school workforce census).
SWC	School workforce census An annual census of the state-funded schools teaching workforce in England taken each November. The census records the characteristics of teachers, including whether they have qualified teacher status, and the subjects that individual secondary teachers teach. The census data that has been used in the latest version of the teacher workforce model was taken in November 2024 and is used to represent stock size in the 2024/25 academic year. To note, there may be more recently published census data at the time of reading but this was the most up to date data available at the time of updating the model.
UGITT	Undergraduate initial teacher training

1. Scope of the calculations – the state-funded schools sector

All calculations have been made on the basis of meeting the future demand for teachers in state-funded primary and secondary schools in England. This includes post-16 provision within state-funded secondary schools. All calculations are at a national level (England). 

Whilst special schools, pupil referral units, independent schools, supply teachers, Welsh schools, and Scottish schools are considered as being out of scope, these calculations account for the needs of such institutions indirectly.

For example, these calculations assume that some teachers will leave service in state-funded primary or secondary schools in England to teach in independent schools (or other school types that are ‘out of scope’). Similarly, they assume that some newly qualified teachers that completed training in England will first enter service within such schools.

However, these calculations do not ignore teachers that move in the opposite direction. For example, some teachers may enter service in English state-funded primary and secondary schools from Wales, Scotland, and other English school sectors.

The teacher flows data used to calculate postgraduate initial teacher training (PGITT) trainee need includes the net impacts of all these movements and so ensures that PGITT need indirectly accounts for the teacher workforce needs of sectors that are considered as out of scope.

As the calculations cover state-funded primary and secondary schools in England, some data within these calculations may differ from Official Statistics, such as national pupil projections and school workforce census (SWC) data, due to coverage differences. The data within these calculations is not intended to supplement other publications, it has been provided for separate purposes relating to improving transparency.

2. How are secondary teachers assigned to individual subjects?

The school workforce census (SWC) collects information on the curriculum taught by individual teachers to pupils in years 7 to 13 respectively. The curriculum data is only collected from secondary schools that use electronic timetabling software that can produce data in the format required. This means that the number of schools that provide data can change from year to year. 

As curriculum information is sourced from a subset of schools in England, the data are weighted and grossed so that the resulting totals provide a consistently representative, national picture over time.

The census does not state that an individual teacher is a ‘Mathematics teacher’ or an ‘English teacher’ etc, it merely records which subjects a teacher teaches and the extent to which they do so in a typical week (usually census week).

To avoid double counting of individual teachers and to reflect that teachers may teach multiple subjects, individual teachers are assigned to the subjects they teach pro rata. 

For example, take a full-time teacher who teaches: 12 hours of Mathematics, 4 hours of Physics, 2 hours of General Studies, and 1 hour of PSHE (Personal, Social, Health, and Economic education) in a typical week. For the purpose of calculating teacher demand and postgraduate initial teacher training (PGITT) trainee need, both General Studies and PSHE are out of scope, and so are excluded from the calculation. In this example, this teacher is counted as 0.75 full-time equivalent (FTE) of a Mathematics teacher (12 hours out of 16) and 0.25 FTE of a Physics teacher (4 hours out of 16). 

Teaching of ‘Science’ and ‘General Science’ is assigned pro rata to Biology, Chemistry, and Physics respectively to reflect that trainee need for PGITT are not set for ‘General Science’. Similarly, teaching of ‘Humanities’ is assigned pro rata to Geography and History.

This methodology is applied to the assigning of secondary teachers to individual subjects for historical stock, entrants, and leavers data.

Other publications (e.g. the ‘school workforce in England’ publication) may count individual teachers multiple times, counting them against each subject that they teach.

A list of the subject classifications used, e.g. those subjects classed as being part of the group called ‘Others’ may be found on the ‘Overview’ tab of the “Calculation of 2026-27 postgraduate initial teacher training (PGITT) trainee need and related data” Excel workbook.

3. Estimating future demand at phase level (Teacher demand trajectories)

The starting assumption for these calculations is that the number of teachers as reported in the most recent school workforce census (SWC), and the associated pupil to teacher ratio (PTR) were sufficient to meet teacher demand at that time. This assumption uses the best available, most comprehensive, and most recent evidence on schools' deployment of the teaching workforce. This is supported by the well-established evidence on the relationship between pupil numbers and PTRs.

Firstly, there is limited evidence around an optimum PTR that the sector could adopt. Secondly, there has been a long-established relationship between pupil numbers and PTRs over many decades; and both pupil and teacher numbers change on a year-on-year basis. As a consequence, the Department for Education (DfE) updates the demand trajectory on an annual basis to reflect this. 

Historical and future teacher and pupil numbers are included in the ‘Input data’ worksheet within the ‘Calculation of 2026-27 postgraduate initial teacher training (PGITT) trainee need and related data’ file. Historically the teacher workforce model (TWM) has used national pupil projections (NPP), but these differ slightly from those published in the NPP due to slight coverage differences, to ensure that they are consistent with the methodology of the TWM. In particular, NPP numbers exclude post-16 pupils within English state-funded secondary schools.

Additionally, whilst calculating 2026/27 postgraduate initial teacher training (PGITT) trainee need, these have been adjusted slightly to account for the actual number of pupils captured in October 2025 via school returns in the dedicated schools grant (DSG).

Historically, PTRs have increased as pupil numbers have grown and fallen when pupil numbers decreased. In other words, schools meet increased demand for teachers by both increasing teacher numbers, and by allowing PTRs and class sizes to grow. DfE assumes that as future pupil numbers grow, future PTRs will grow in line with historical increases. If pupil numbers are to fall in future, DfE assumes that future PTRs will fall in line with historical falls.

Using official pupil projections data, DfE estimates the rate and direction that PTRs will change in future as pupil numbers change for primary and secondary respectively. Using these estimated future PTRs and official pupil projections, DfE estimates the number of teachers needed to meet these future PTRs in each future year. This provides the estimated future demand for teachers in state-funded primary and secondary schools respectively in England. 

Equation 1: Teacher workforce demand calculation 

The estimate of overall teacher demand for primary includes both qualified and unqualified (without qualified teacher status (QTS)) teachers. By assuming that the percentage of primary teachers that are unqualified will remain constant, and in steady state at the current level, the qualified teacher demand for primary is calculated. As with all assumptions, this assumption on future unqualified teacher rates is reviewed on an annual basis. 

Next, the overall secondary demand needs to be broken down by subject, see “Section 4. Estimating future secondary demand at subject level”.

The expected growth of secondary teacher demand between 2023/24 and 2027/28 represents the minimum contribution we expect from secondary schools to meet the 6,500 additional teachers pledge (alongside the contribution we expect from special schools and the further education sector which are not in scope of this publication).

4. Estimating future secondary demand at subject level

The percentage of teaching hours currently dedicated to each subject at a national level is calculated using school workforce census (SWC) data. It is currently assumed that these percentages will be maintained in future years as the curriculum is broadly stable over time. The teaching of subjects like General Studies and PSHE (Personal, Social, Health, and Economic education) are excluded, as to do otherwise would assume that a substantial number of General Studies and PSHE trainees would be required each year. 

It is assumed that the total number of future secondary teaching hours will grow or shrink, from the current SWC level, in line with future pupil numbers. This reflects recent trends as recorded in the SWC. This provides an estimate of the total number of secondary teaching hours for future years.

This number is broken down into the number of secondary teaching hours dedicated to each individual secondary subject by applying the current proportion of teaching hours as calculated from the latest SWC. This provides a future estimate of teaching hours dedicated to each subject.

If there are policies that might impact upon the future demand for specific subjects, an adjustment can be applied to these percentages. Relevant policies might include a new, expanded GCSE qualification for example. Policies are only included once they have been fully developed and ready for implementation and will be reviewed on an annual basis.

This year there are no policies that have been included within DfE’s demand estimates.

Using these estimates, the year-on-year change rate in the number of hours forecast for each subject is calculated. It is assumed that the total (qualified and unqualified) teacher demand for each secondary subject will change year-on-year (from the current stock size as recorded within the SWC) at that rate. This provides the total teacher demand for each secondary subject in each future year.

Lastly, it is assumed that the percentage of secondary teachers that will be unqualified (teachers without qualified teacher status (QTS)) will be in steady state in future years. The current percentage of teachers that are unqualified is calculated for each secondary subject from the SWC. This reflects that some subjects have a greater percentage of teachers that are unqualified than others. By applying these percentages to the total teacher demand for each subject that was estimated previously, the unqualified and qualified teacher demand is estimated for each secondary subject for future years. 

5. Estimating future supply, and adjustments relating to under-recruitment impacts

When calculating the 2026/27 postgraduate initial teacher training (PGITT) trainee need, the calculations use ‘demand met’ and ‘estimated supply’ scenarios respectively to make an adjustment to reflect any under-supply from the previous two initial teacher training (ITT) cycles (2024/25 and 2025/26). 

Firstly, 2026/27 PGITT trainee need are estimated using 2027/28 teacher demand figures for primary and each secondary subject. This calculation is designed to estimate a sufficient number of trainees to meet teacher demand in 2027/28. The calculations assume that the stock size in 2024/25 (as recorded in the school workforce census (SWC)) is sufficient to meet demand and that the future demand in 2025/26 and 2026/27 will be met precisely. In other words, when undertaking calculations based on demand, any under- or over-recruitment impacts from the previous two ITT cycles (2024/25 and 2025/26) are ignored. This is the ‘demand met’ scenario.

Secondly, these calculations are concurrently undertaken using a short-term estimate of future teacher supply for the years 2025/26 and 2026/27. Therefore, these supply-based calculations do account for any impacts of forecasted under- or over-recruitment from the previous two ITT cycles. This is the ‘estimated supply’ scenario.

For all the calculations, data on the current teacher workforce is derived from the November 2024 SWC. This census includes those teachers that completed ITT during 2023/24 and joined the teaching workforce in September 2024 as newly qualified entrants (NQEs).

However, the trainees of 2024/25 were already in training when the analysis to estimate 2026/27 PGITT trainee need was undertaken. Furthermore, the recruitment of trainees for 2025/26 was already complete. Neither of these ITT training cohorts featured in the 2024/25 SWC as qualified teachers, but the size of those two training cohorts were known. Therefore, the recruitment impacts upon future SWCs from ITT in these two years can be estimated. 

DfE estimates future supply in 2025/26 and 2026/27 for primary, and each secondary subject by:

• Taking the current, 2024/25 stock size, as recorded by the SWC and:
  • Adding the number of entrants expected in 2025/26 – see more details on estimating entrant numbers in “Section 6. Estimating future returner and new to state-funded sector (NTSF) entrant numbers” and “Section 7. Estimating future newly qualified entrant (NQE) numbers”.
  • Subtracting the number of leavers expected in 2025/26 – see more details on estimating leaver numbers in “Section 8. Estimating future leaver numbers, and the flows-stock adjustment”.
  • The resultant figure is an estimate of the qualified stock size in 2025/26. 
• The cycle is repeated for 2026/27 using the estimated 2025/26 stock size and: 
  • The estimated number of entrants expected in 2026/27 is added.
  • The estimated number of leavers expected in 2026/27 is subtracted.
  • The resultant figure is the qualified stock size in 2026/27 (supply estimate). 

Having calculated the 2026/27 PGITT trainee need (see the next sections) based upon both the ‘demand met’ and ‘estimated supply’ scenarios respectively, a MAX statement is used to choose the larger of the two values. This value is the mainstream 2026/27 PGITT trainee need, to which the 2026/27 high potential initial teacher training (HPITT) trainee need is added to obtain the overall 2026/27 PGITT trainee need. 

In those cases where the PGITT trainee need based upon the ‘demand met’ scenario has been chosen; the calculations have estimated that supply in 2026/27 will exceed demand. If the trainee need calculated under the ‘estimated supply’ scenario was chosen instead, this would result in the trainee need being deflated because the estimated supply exceeds demand. In other words, the trainee need would be reduced to account for historic over-recruitment. To avoid this issue, the PGITT trainee need is calculated based on teacher demand being met and not exceeded each year. 

By contrast, for those subjects where the PGITT trainee need based upon the ‘estimated supply’ scenario is chosen, the calculations have estimated that supply in 2026/27 will not meet demand. Taking this approach results in the 2026/27 PGITT trainee need being inflated to counter the under-supply impact from the two most recent PGITT recruitment rounds (i.e. forecasted under-supply between the most recent SWC data point of 2024/25 and our forecasted position for 2026/27). This is referred to within these calculations as the ‘under-supply adjustment’.

As a consequence, when calculating whether an adjustment relating to under-supply impacts should be made, DfE has made a broad assessment of short-term teacher supply, considering both future retention, and recruitment from routes other than PGITT. DfE does not assume that a historical trainee need was missed by ‘x’ and then add ‘x’ on to future trainee need; a broader, and more holistic approach to supply and recent recruitment is taken instead. 

For example, when setting the 2022/23 Mathematics trainee need four years ago, the PGITT trainee need for 2020/21 and 2021/22 had been missed. This might suggest that an adjustment to counter this under-supply would have been necessary. However, the impacts of under-supply against PGITT trainee need were fully offset by more favourable recruitment via other routes and improvements in short-term retention, making such an adjustment unnecessary.

The scale of the adjustment to counter recent under-supply used for the 2026/27 PGITT trainee need may be estimated by subtracting the PGITT trainees required calculated under the ‘demand met’ scenario from that under the ‘estimated supply’ scenario. This calculation would generate negative figures for those subjects where such an adjustment was not used, i.e. the actual adjustment used was equal to zero.

6. Estimating future returner and new to state-funded sector (NTSF) entrant numbers

DfE estimates future returner and new to state-funded (NTSF) entrant numbers for primary, and each secondary subject. NTSF entrants include those teachers who are new to the sector, but who are not newly qualified entrants (NQEs) (according to the school workforce census (SWC)), and those NQEs who deferred their entry into the workforce after initial teacher training (ITT) by between 4 and 16 months. 

These estimates consider recent trends as recorded in the SWC, the demographics of those teachers, the relationship with leaver rates (demand for returners falls when teacher retention improves), ITT recruitment, and broader economic factors. The calculations use full-time equivalents (FTEs) to account for those teachers who are not employed in full-time roles. 

For example, deferred NQE numbers grew during the pandemic, as they did during the recession of 2008. This suggests that during an economic downturn, permanent employment opportunities for NQEs in the months immediately following ITT are likely to become more limited as the retention rate of existing teachers improves. Additionally, if ITT recruitment improves, we would expect the number of deferred NQEs to increase, all things being equal. Trend analysis, ITT recruitment data, and consideration of broader economic factors was used by DfE this year to estimate how deferred NQE numbers might change in 2025/26, 2026/27, and 2027/28.

7. Estimating future newly qualified entrant (NQE) numbers

The number of newly qualified entrants (NQEs) entering service in the next 2-3 years must be estimated for two reasons. Firstly, it is used to estimate future supply in 2026/27 (see “Section 5. Estimating future supply, and adjustments relating to under-supply impacts” for more details). Secondly, the number of NQEs who will enter the state-funded sector via initial teacher training (ITT) routes other than postgraduate initial teacher training (PGITT) in 2027/28 is required to estimate the number of NQEs required from PGITT in 2027/28 (see “Section 10. Estimating NQE need and PGITT NQE need”). 

There are a number of different training routes for NQEs in England including PGITT, undergraduate initial teacher training (UGITT), and high potential initial teacher training (HPITT). The forecast for the number of NQEs trained through such routes who will enter service in the state-funded sector each year are estimated for both primary, and each secondary subject by:

• Taking ITT census recruitment data, or internal ITT recruitment data where census data is unavailable, to estimate the number of trainees on training courses who will finish training in a specific year. 
• Estimating the number of those trainees who will complete their training (‘completers’). This is achieved by applying subject specific ITT completion rates, derived from data published as part of the ITT performance profiles. 
• Estimating the number of ‘completers’ that will enter service in state-funded mainstream schools in England as an NQE in the first part of the academic year that follows their teacher training. This is accomplished by applying subject specific post-ITT employment rates derived from data published as part of the ITT performance profiles. 
• Applying a full-time equivalent (FTE) rate for NQEs to reflect that not all NQEs enter service in a full-time role. This rate is derived from the latest school workforce census (SWC) data.

Note – the ITT completion and post-ITT employment rates data used within these calculations differ to that published in the ITT performance profiles publication. All ITT rates data used are assumed rates rather than the latest published ones because of the following reasons: 

• Firstly, for those subjects where there were small cohorts of trainees (or the rates were disproportionately high/low compared to the secondary average, typically more than 10 percentage points difference), a proxy rate may have been used.[2] For example, because the cohort of secondary undergraduate initial teacher training (UGITT) trainees was small, the calculations instead used the completion/employment rates for the total secondary UGITT trainee cohort instead, rather than subject specific rates.
  • Note - employment rates for Classics are extremely low compared to other secondary subjects, e.g. below 40%. 
  • Given this low value and the small data sample size, we have historically used an assumed rate that was no more than 10 percentage points below the secondary average. 
  • However, recent internal analysis suggests that Classics employment rates are extremely low because Latin teachers disproportionately enter employment in independent schools. 
  • To reflect this demand, since setting the 2025/26 trainee need, the employment rates calculations for Classics ignore the 10-percentage points difference limit. 
  • This methodological change has been a key driver of the sharp increase in the PGITT trainee need for Classics from 2025/26 onwards - narrowing the gap between the trainee need and likely ITT demand.
• Secondly, the employment rates reflect the state-funded school sector coverage as defined by the trainee need calculations (e.g. state-funded primary and secondary schools in England). Some newly qualified teachers may gain employment in other sectors (e.g. Scotland and the independent sector), and some may become deferrers and are subsequently picked-up as new to state-funded sector entrants within these calculations. 
• Lastly, because of impacts from the covid-19 pandemic, it was judged that some of the most recent historical rates will likely not be representative of what will happen in future years. Therefore, the assumed rates used in these calculations have been derived using weighted averages from pre-pandemic data and/or post-pandemic data.

In addition to those NQEs sourced via ITT in England, some NQEs gain qualified teacher status (QTS) via the ‘assessment only (AO) route to QTS’ each year. Future estimates of NQEs that will be sourced from this route are taken from internal estimates of future AO recruitment that are based on both recent trends and consideration of future provision.

Finally, there are a small number of NQEs that are recorded within the SWC who cannot be attributed to either ITT in England or the AO route. For example, a teacher that recently gained QTS in Scotland but has entered service in England instead. The numbers of such NQEs have been estimated using ITT and SWC data and a weighted average of their numbers is applied within these calculations to estimate the future number. This is referred to within these calculations as the ‘ITT-NQE adjustment’.

Footnotes:

[2] For example, the secondary average rate might be estimated to be 65%, and the rate for subject ‘x’ is calculated to be 76%. As 76% is 11% higher than the secondary average (65%), the difference would be limited to 10%, and the rate used for subject ‘x’ would be held at 75%.

8. Estimating future leaver numbers, and the flows-stock adjustment

The future number of leavers must be estimated to calculate both future short-term supply (see “Section 5. Estimating future supply, and adjustments relating to under-supply impacts”) and the future demand for postgraduate initial teacher training (PGITT) trainees.

The calculation to estimate the number of leavers in the following year uses two components: an estimation of the leaver rate, and the stock of teachers at the end of the previous academic year. Applying the estimated leaver rate to the previous year's stock of teachers produces an estimate of the number of leavers, and indirectly, the number of teachers who remain in service at the start of the following academic year.

An estimated leaver rate is separately produced for teachers aged over and under 55 years old which recognises that the likelihood of leaving the profession is driven by different factors for older and younger teachers respectively. Separate leaver rates are produced for primary, and each individual secondary subject, reflecting differences in teacher retention levels between the teacher cohorts of different subjects. Leavers rates within the calculations file have been summarised across all ages, and separately for both over and under 55-year-old teachers.

In 2020/21, there was an unprecedented decrease in the number of teachers leaving service, with leaver rates falling to the lowest level observed since the school workforce census (SWC) started being collected in 2010. This fall was most likely a result of covid-19. 

In 2021/22, both primary and secondary leaver rates increased slightly but remained below their pre-pandemic levels. This was likely to be a result of continuing effects of the covid-19 pandemic. 

As expected, leaver rates increased back up to nearer pre-pandemic levels in 2022/23 - as the economic effects of the pandemic subsided. In 2024/25, we have seen a decrease in most leaver rates compared to 2022/23 and 2023/24. Primary has seen a larger year-on-year decrease in leaver rates compared to most secondary subjects.

Historical leaver rates were retrospectively revised downward in the SWC last year. This related to a data issue within the Teacher Pension Scheme extracts that inflated leavers and returners for <1,000 teachers per year that has since been resolved by the publication team. This revision has been a key driver in there being a lower leaver rate trajectory in this year’s calculations of 2026/27 PGITT trainee need compared to those for 2025/26.

To develop estimates for future leaver rates, the latest available economic forecasts from the OBR (as part of their Autumn Budget in November 2025) were accounted for, and any other known factors that might impact on future teacher numbers not yet reflected in the baseline data. For example, an assessment was made of the relative strength of teachers pay compared to pay in the broader economy and the impact this may have on future retention.

To note, the Classics teacher leaver rate in 2024/25 was 14.3% which is uncharacteristically high in comparison to 2023/24 (9.2%). This is likely a consequence of a small cohort size of Classics teachers and because of this, the Classics leaver trajectory was fixed as an average leaver rate across the latest 7 years of data for Classics.

Aside from teachers leaving the workforce, existing teachers that stay in service may choose to reduce their working hours between years, leading to a reduction in capacity in terms of full-time equivalent (FTE) teachers that need to be replaced. To consider this, the calculations use a ‘flows-stock adjustment’ that has been calculated for primary and secondary separately using data from historical SWC data.

This adjustment was calculated by examining flows and stocks figures. For example, if the stock size (FTE) as recorded in the SWC in a given year (e.g. 2019/20) is taken, and the number of entrants in the following year is added to this figure, and the number of leavers subtracted, the resultant stock size is not equivalent to the stock size in 2020/21. This is because some of the existing stock in 2019/20 reduced their working hours between 2019/20 and 2020/21. Typically, newly qualified entrants (NQEs) tend to enter service full-time, and teachers will, on average, reduce their working hours over the course of their careers.

9. Estimating future entrant need

These calculations estimate future entrant need based upon both the ‘demand met’ and ‘estimated supply’ scenarios for primary and each secondary subject for three academic years: 2025/26, 2026/27, and 2027/28. Performing these calculations based upon both scenarios enables entrant need to be estimated that reflects any short-term under- or over-recruitment impacts (see “Section 5. Estimating future supply, and adjustments relating to under-supply impacts”).

There are two elements to the entrant need: replacing leavers and accounting for year-on-year changes in demand. These two elements are added together, to estimate the overall entrant need. The leaver numbers element is estimated as outlined in “Section 8. Estimating future leaver numbers, and the flows-stock adjustment”.

The second element, accounting for year-on-year changes in demand is estimated as follows, the calculations for 2027/28 teacher entrant need under the ‘demand met’ scenario are used as an example: 

• Firstly, the calculations take the 2027/28 teacher demand and multiply this by the ‘flows-stock adjustment’ to account for teachers whose hours will reduce year-on-year.
• From this, the 2026/27 teacher demand (from the previous academic year) is subtracted.

For primary, the teacher demand figure reflects that the stock size can fall in size year-on-year because of falling pupil numbers. 

Whilst secondary pupil numbers are forecasted to fall in the later 2020s, DfE doesn't expect teacher demand to fall rapidly as the pupil bulge moves into post-16 where class sizes are smaller and the curriculum is broader.

Therefore, these calculations assume that entrant need can fall if teacher demand falls, or because the workforce does not need to grow as rapidly year-on-year. Secondly, the calculations assume that any leavers require replacement.

10. Estimating NQE need and PGITT NQE need

The entrant need that was previously calculated (see “Section 9. Estimating future entrant need”) covers entrants from all forms of recruitment, not just initial teacher training (ITT). Therefore, this must be converted into the number of newly qualified entrants (NQEs) needed to meet demand, and then, the number required that would need to be trained via postgraduate initial teacher training (PGITT).

Firstly, to estimate the total number of NQEs required, the forecasted numbers of returners and new to state-funded (NTSF) sector entrants (see “Section 6. Estimating future returner and new to state-funded sector (NTSF) entrant numbers”) are subtracted from the entrant need. This provides the overall NQE need for future years, including 2027/28.

To estimate the number of NQEs who will need to be trained via mainstream PGITT: the number of NQEs expected from alternative training routes in the relevant year is subtracted from the overall NQE need. These NQEs include those trained via high potential initial teacher training (HPITT), undergraduate initial teacher training (UGITT), and assessment only (AO) in the relevant year (see “Section 7. Estimating future newly qualified entrant (NQE) numbers”). 

The residual figure is the estimate of the number of NQEs required each year to be trained via mainstream PGITT. As with the previous step to calculate entrant need, figures are calculated based upon both ‘demand met’ and ‘estimated supply’ scenarios respectively.

11. Converting 2027/28 PGITT trainee NQE need into 2026/27 PGITT trainees required

Finally, the number of newly qualified entrants (NQEs) required via postgraduate initial teacher training (PGITT) in 2027/28 needs to be upscaled into the number of mainstream PGITT trainees required in 2026/27. This step accounts for those trainees who will not complete initial teacher training (ITT), and those who will not enter service in state-funded mainstream schools in England as NQEs immediately after ITT. Additionally, the process accounts for those NQEs who enter service as part-time, as opposed to full-time, teachers. 

To do this, the number of NQEs required from PGITT (see “Section 10. Estimating NQE need and PGITT NQE need”) is divided by the full-time equivalent (FTE) rate of NQEs, the assumed completion rate of ITT trainees, and the assumed post-ITT employment rate of ITT trainees. 

Equation 2: Number of PGITT trainees required calculation

Lastly, a MAX statement is applied, as outlined in “Section 5. Estimating future supply, and adjustments relating to under-supply impacts”, to select the larger of the PGITT trainee need calculated under the ‘demand met’ and ‘estimated supply’ scenarios respectively. This allows an adjustment to be applied to counteract under-supply impacts from the previous two ITT cycles. 

Equation 3: Number of PGITT trainees required calculation

As a very final step of the calculations, upon the ‘Mainstream PGITT & HPITT trainee need’ worksheet, mainstream PGITT trainee need is collated from all the previous tabs that relate to primary and individual secondary subjects. This mainstream trainee need is added to the relevant 2026/27 high potential initial teacher training (HPITT) trainee need to calculate the overall 2026/27 trainee need. This is then rounded to the nearest 5 in line with the publication outputs.

12. Teachers per 1,000 pupils

To better understand the ratios of teachers to pupils, in this year’s publication we have added an extra table which summarises the number of qualified and unqualified teachers (full-time equivalent (FTE)) in service compared to every 1,000 pupils in state-funded schools.

For primary, the number of primary teachers is divided by the number of primary pupils each year. For secondary subjects, the number of secondary teacher FTE in each subject (i.e. if a teacher spends 20% of their timetable teaching Mathematics and 80% teaching Physics, they are classes as 0.2 FTE of a Mathematics teachers and 0.8 FTE of a Physics teacher) is divided by the number of total secondary pupils. Therefore, the ratio of teachers per 1,000 pupils in secondary subjects is lower than for primary given it is split across the subjects rather than overall.

Equation 4: Number of qualified teachers per 1,000 pupils calculation

The number of teachers for every 1,000 pupils has fallen in recent years for almost every subject compared to a decade earlier, this reflects that overall secondary pupil teacher ratios have grown over this time. 

The values have changed to different extents in different subjects as a reflection of changes in teaching time in different subjects. However, care should be taken when comparing different subjects to each other as some subjects naturally have greater teaching demands. For example, there are disproportionately more teachers of Mathematics than of Classics as relatively few pupils are taught Classics. By contrast, Mathematics is compulsory to most secondary school pupils with considerably greater teaching demands. 

13. Drivers of 2026 to 2027 PGITT trainee need

What are drivers?

Drivers are factors that may act to increase or decrease the number of postgraduate initial teacher training (PGITT) trainees required between years. 

For example, for 2026/27 PGITT the trajectory for returner recruitment may have acted to increase or decrease PGITT need for primary by ‘x’ trainees compared to the 2025/26 trajectory, in isolation of other factors. 

Not all drivers necessarily act in the same direction. If PGITT need were to increase for a specific subject this year, many drivers would likely be acting to increase PGITT need. However, some may be acting in the opposite direction, in isolation of other drivers, to decrease PGITT trainee need this year.

The PGITT trainee need figures upon which this analysis is undertaken include high potential initial teacher training (HPITT). 

Key terms

The table below provides definitions for terms used within the drivers analysis.

Term	Description
2025/26 PGITT need	The number of postgraduate initial teacher trainees required to start training in 2025/26 (as published in spring 2025). This includes high potential initial teacher training (HPITT) trainees.
2026/27 PGITT need	The number of postgraduate initial teacher trainees required to start training in 2026/27 (as published in 2026). This includes high potential initial teacher training (HPITT) trainees.
Demand growth YOY	The difference between the number of teachers required in service in the relevant year, and the year before (the year-on-year difference). I.e. how fast is the workforce needing to grow because of pupil number growth, or how fast can it shrink as pupil rolls fall. In the case of the 2026/27 PGITT trainee need this would be the difference in teacher demand between 2026/27 and 2027/28. The latter being the academic year that 2026/27 trainees would first enter the workforce as newly qualified entrants. For 2025/26 PGITT trainee need this would be the difference in teacher demand between 2025/26 and 2026/27.
ITT-NQE adjustment	This adjustment is used to inflate the number of newly qualified entrants (NQEs) expected when estimating PGITT trainee need. It accounts for NQEs that are picked up within the school workforce census each year having not been accounted for within previous year's ITT data. These individuals could have been omitted from ITT data for a variety of reasons including: Entrants who have gained qualified teacher status (QTS) through mutual recognition rather than ITT Entrants that are based in missing schools Entrants that gained QTS in Scotland and Wales Entrants that were simply missing or had incomplete records in ITT data Phase level movements between primary and secondary of NQEs
ITT-NQE conversion rate	If 100 trainees begin PGITT in 2026/27, this would not translate to 100 full-time equivalent (FTE) NQEs entering service in 2027/28. Some NQEs enter part-time roles, some trainees do not complete training, and some newly qualified teachers do not immediately enter service in state-funded schools.  This conversion rate is used in the TWM to reflect these losses. It is based on FTE rates of NQEs from the school workforce census, ITT completion rates, and post-ITT employment rates.
Leavers	The number of teachers that are expected to have 'left service' in a state-funded primary or secondary school in England. This will include retirements and deaths in service.
NQEs	Newly qualified entrants. Teachers that enter service within 4 months of gaining QTS (qualified teacher status).
NQEs from other sources	Teachers who enter service in a state-funded primary or secondary school in England immediately after gaining qualified teacher status (QTS) that have entered through routes other than postgraduate ITT (including high potential initial teacher training (HPITT). For example, those trained via undergraduate ITT, or those gaining qualified teacher status via assessment only (AO).
NTSF	New to state-funded sector (NTSF) entrants. Teachers who enter service in a state-funded primary or secondary school in England having not been employed as a regular teacher in the sector previously (as recorded in the school workforce census). This group will include some newly qualified entrants that deferred their entry into service by 4 to 16 months (deferrers).
Relevant year	Relevant year or ‘RY’ is a shorthand way of referring to the academic year in which trainees in the relevant PGITT trainee need year enter service as newly qualified entrants. For 2026/27 PGITT trainee need, the 'relevant year' is 2027/28. E.g. for those training in 2026/27, the number of returners expected in the relevant year, 2027/28, will impact upon the need to recruit newly qualified entrants. I.e. if more returners are recruited in 2027/28, this reduces the need for PGITT trainees in 2026/27 all else being equal. When estimating the drivers of 2026/27 PGITT need vs 2025/26 PGITT need we must not compare the number of returners expected in the same academic year. We must compare like-with-like and compare those in the 'relevant year' for each PGITT cohort. So, for 2025/26 PGITT, the number of returners in the relevant year which influence PGITT need is the number expected in 2026/27. For 2026/27 PGITT, it is the number a year later, in 2027/28.
Returners	The number of teachers who enter service in a state-funded primary or secondary school in England having been employed as a regular teacher in the sector previously (but with a break in service).
PGITT	Postgraduate initial teacher training.
PG NQE need in the relevant year	The number of newly qualified entrants (NQEs) needed to meet demand in the relevant year. These are NQEs from mainstream postgraduate initial teacher training (PGITT) and high potential initial teacher training (HPITT) only.
Under-supply adjustment	An adjustment accounting for potential supply shortfalls resulting from the two ITT cycles prior to the year for which we are setting PGITT need (2024/25 and 2025/26). These are ITT cycles that have already occurred but are yet to be reflected in the school workforce census as of 2024/25. If a shortfall is estimated, the model assumes additional teachers will need to be recruited via PGITT to correct it. The model accounts for ITT recruitment, teacher retention, and other recruitment routes (e.g., returners). This holistic assessment means the impact of missing historical PGITT trainee need may be offset by wider recruitment or retention being better than expected. Negative driver numbers reflect that the adjustment is smaller than last year resulting in reduced PGITT need. If it is a positive driver the adjustment is larger. If the value is zero, there was no adjustment, either this year or last, making this driver irrelevant. More information may be found within “Section 5. Estimating future supply, and adjustments relating to under-supply impacts”.
Working hour losses	Losses from the workforce in the form of individual teachers reducing their working hours between years. This is the net loss, accounting for teachers that might increase working hours between years.

How are drivers calculated?

The process of estimating the scale of different PGITT trainee need drivers is complex. 

Ideally it would be possible to quantify the scale and direction of different drivers by keeping all modelling the same but changing the data for one single driver to be equal to the value used in the previous year. The difference in PGITT trainee need calculated by changing that one factor could then be assumed to be the estimated impact of that driver.

Unfortunately, that methodology would not work because the total impact of adding all the different resultant drivers together would not equal the year-on-year change in PGITT need. This is because all drivers are intertwined with one another to an extent. Additionally, the assumed conversion rate of ITT trainees to newly qualified entrants (NQEs) is different in different versions of the teacher workforce model. Finally, some drivers are in the form of PGITT trainees (such as the undersupply adjustment), some are in the form of rates (e.g. completion rates), and others are in the form of full-time equivalent teachers such as the number of returners.

This methodology narrative outlines the process step-by-step that is taken to avoid these issues.

What drivers are tested?

The impact of the following isolated drivers is estimated: 

• The under-supply adjustment.
• The number of entrants expected in the relevant year via returners, new to state-funded sector (NTSF) entrants, the ITT-NQE adjustment, and newly qualified entrants (NQEs) from other sources respectively. 
• The number of leavers expected in the relevant year via standard leaver routes (including retirement) and working hour losses respectively. 
• The ITT-NQE conversion rate. 
• Teacher demand growth in the relevant year. 

Whilst there are some factors that have not been accounted for, e.g. changes in the assumed full-time equivalent rate of teachers in service, the impact of such minor factors will be extremely small and is largely absorbed within the impact of those drivers listed above.

Because this analysis is estimating the drivers of changes in PGITT trainee need in two different academic years, differences must be calculated for factors which relate to two different academic years. 

• For example, the PGITT trainee need in 2025/26 is influenced by the number of returners in 2026/27. By contrast, the PGITT trainee need in 2026/27 is influenced by the returner numbers in 2027/28. 
• To get around this issue, for simplicity, the analysis refers to numbers from different versions of the teacher workforce model (TWM) being for the ‘relevant year’ rather than a specific academic year.
• The relevant year is the year after which the respective PGITT trainees complete training and so enter the workforce as NQEs. 

The impact of recruitment and retention in the two years between the latest school workforce census and the relevant year is accounted for within the drivers analysis via the undersupply adjustment. This includes the estimated impacts of ITT recruitment reported in the latest ITT census. 

• For each subject, the TWM estimates whether there is a supply surplus or deficit in the year before the relevant year. 
• For subjects where there is a deficit, the model makes an adjustment to inflate PGITT need to counteract the undersupply by estimating PGITT trainee need under the supply scenario. Therefore, improvements in retention and/or recruitment in the intervening two years are accounted for via a larger/smaller under-supply adjustment. 
• By contrast, for subjects where there is an estimated supply surplus, the model does not make an adjustment, and the PGITT trainee need is calculated under the demand scenario. Under those circumstances, improvements in retention and/or recruitment in the intervening two years has no impact upon PGITT trainee need as no adjustment is made to either increase or decrease PGITT trainee need. The supply surplus is ignored with regards to PGITT trainee need. 

Step 1 – Calculating the conversion rate between ITT trainees and NQEs (the ITT-NQE conversion rate)

These rates are used to reflect that some trainees do not complete training, some ITT completers do not gain employment in the state-funded sector immediately after ITT, and some newly qualified entrants (NQEs) do not work full-time. I.e. 100 FTE teacher trainees in year 1 do not provide 100 FTE NQEs in year 2.

Whilst employment and completion rates are used in the workbook called 'Calculation of 2026-27 postgraduate initial teacher training (PGITT) trainee need and related data' those rates reflect mainstream PGITT only and exclude HPITT.

As this drivers analysis is performed on PGITT trainee need figures that include HPITT, suitable conversion rates are calculated separately that reflect the conversion including both mainstream PGITT and HPITT trainees.

As the impact of the under-supply adjustment is a distinct driver in itself, and because the PG NQE need does not include NQEs needed as part of the adjustment, the adjustment is subtracted from PGITT need before calculating these rates.

Note - the Classics conversion rate is intentionally low and reflects that a disproportionately high number of trainees seek employment in the independent schools sector.

Step 2 - Calculating the differences in postgraduate newly qualified entrant need because of different teacher flow factors

The impact of all inflows and outflows upon the number of newly qualified entrants (NQEs) required having been trained via PGITT is estimated. 

This is calculated as the difference in the assumed flow from the calculation of the 2025/26 PGITT trainee need and the 2026/27 PGITT trainee need and is in the form of NQEs.

These values are calculated for returners, new to state funded, the ITT-NQE adjustment, NQEs from other sources, leavers, and losses from reductions in working hours respectively.

For all entrant routes (e.g. returners), any positive figures are converted into negatives and vice versa. This reflects that if there is a higher number of entrants expected this would have a negative impact upon PGITT need. For example, if there was a higher number of returners expected in the relevant year that would reduce PGITT trainee need in the preceding year as fewer NQEs would be required to meet demand. 

By contrast, a reduction in leavers would lead to a reduction in PGITT trainee need, and so no such conversion is needed for outflows.

Step 3 - Estimating the impact on PGITT trainee need from the assumed conversion rate between PG NQEs and PGITT trainees

These conversion rates reflect that not all teacher trainees will complete training and enter the teaching workforce as full-time teachers in the year after training. 

Every year, different assumed ITT completion rates, post-ITT employment rates, and newly qualified entrant (NQE) full-time equivalent rates are used within the teacher workforce model. This is because assumed rates are derived from the latest data in the ITT performance profiles and school workforce census publications, and new data is published every year.

To estimate the impact upon PGITT trainee need of these rates changing year-on-year, the number of NQEs needed following PGITT is converted into PGITT trainees required to deliver them. This calculation is undertaken using both the conversion rates and PG NQEs needed from both this, and last year’s teacher workforce model. 

The resulting differences in PGITT need are then used to calculate an average impact, and this is assumed to be the impact upon PGITT need this year of there being a different conversion rate.

If the assumed conversion rate is higher this year than last, this acts to reduce PGITT trainee need. For example, if the conversion rate was 62% when calculating PGITT last year and it is assumed to be 64% this year the teacher workforce model would calculate that a greater number of NQEs would be achieved for every 100 ITT trainees that start training. Therefore, fewer PGITT trainees would need to be recruited to meet demand, all things being equal. 

Step 4 - Calculating a ratio to convert estimated impacts in the format of PG NQEs into impacts in the form of PGITT trainees

All teacher flow drivers have been calculated so far in the form of postgraduate newly qualified entrants (NQEs), i.e. a full-time equivalent number of NQEs.

For the purposes of PGITT trainee need drivers analysis, they must be converted into PGITT trainees.

However, it is extremely challenging to calculate the isolated impact of individual drivers as all factors are intertwined with one another and the conversion rate between NQEs and ITT trainees has changed between different versions of the teacher workforce model (TWM). 

To get around this, the difference in PGITT trainee need from last year is taken, and from this, the two drivers that are already in the form of PGITT trainees are subtracted. These include the conversion rate impact (calculated above in step 1) and the under-supply adjustment.

The remaining figure is the difference in PGITT trainee need that cannot be explained by the conversion rate and under-supply adjustment impacts. This is then divided by the net impact in PG NQE terms (previously calculated in step 2) that can be explained by changes in flows factors.

The resulting ratio is then used to convert factors in PG NQE terms into impacts in the form of PGITT trainee need.

The net total of all the converted drivers in the format of PGITT trainees should be equal to the difference in PGITT trainee need between this year and last year. 

A waterfall chart is used to show the direction and scale of the different drivers upon PGITT trainee need.

Calculating overall secondary drivers

These are calculated by totalling up all the values for all individual secondary subjects.

Help and support

Contact us

If you have a specific enquiry about Teacher demand and postgraduate trainee need statistics and data:

Teacher Workforce Modelling team

Email: ittstatistics.publications@education.gov.uk
Contact name: Melissa Cook

Press office

If you have a media enquiry:

Telephone: 020 7783 8300

Public enquiries

If you have a general enquiry about the Department for Education (DfE) or education:

Phone: 0370 000 2288

Opening times:
Monday to Friday from 9.30am to 5.30pm (closed on bank holidays)