WIREDGOV

Key findings

1.     In June, NHS England published its much-awaited long-term workforce plan, which was welcomed across the political spectrum. The plan aims to increase the number of staff employed by the English NHS from around 1.5 million in 2021–22 to between 2.3 and 2.4 million in 2036–37. This would be equivalent to average growth in the size of the NHS workforce of between 3.1% and 3.4% per year. For context, we estimate that NHS staffing numbers grew by around 1.1% per year between 2009–10 and 2019–20. 

2.     If this increase in the NHS workforce is delivered, we estimate that almost half (49%) of public sector workers in England will work for the NHS in 2036–37, compared with 38% in 2021–22 and 29% in 2009–10. By 2036–37, we estimate that one in eleven (9%) of all workers in England will work for the NHS, compared with one in seventeen (6%) in 2021–22 and 2009–10. 

3.     While the plan included £2.4 billion of additional funding for the training of new staff, it did not consider the (much larger) medium-term implications of this large increase in staffing for the NHS paybill, nor the required increase in other inputs if the NHS is to treat substantially more patients. This will mean difficult fiscal decisions at future Spending Reviews. 

4.     Increasing the size of the workforce so rapidly will likely require NHS wages to become more generous in real terms and – potentially – match or even exceed growth in wages in the rest of the economy. It will also likely require an increase in non-staffing inputs to healthcare (most obviously things such as drugs and equipment). 

5.     Under a central set of assumptions, the workforce plan implies annual NHS budget increases of around 3.6% per year in real terms (or 70% in total by 2036–37). This would be in line with the long-run average real-terms growth rate in UK health spending (3.6% per year from 1949–50 to 2022–23), but higher than the 2.4% per year seen since 2009–10. In other words, returning to the NHS’s long-run average funding growth rate could be enough to fund the workforce plan, but those increases would need to be delivered in a period when economy-wide productivity growth is expected to be lower than what it has been for much of the NHS’s history. 

6.     In the central case, spending on the NHS in England would be around 2% of GDP higher by 2036–37, relative to 2021–22 (the starting point for the workforce plan). That is similar to the increase in UK health spending over the 15 years to 2008–09 (2.2% of GDP) and equivalent to around an extra £50 billion in today’s terms. To give a sense of scale, raising that sort of sum would require increasing the standard rate of VAT from 20% to around 27% by 2036–37 or increasing all income tax rates by around 6 percentage points. Other funding options would of course be available.  

7.     By the NHS’s own estimates, the staffing increases contained in the plan will only be enough to meet NHS demand if productivity can be increased by between 1.5% and 2% per year: an extremely ambitious target well above what the NHS is estimated to have achieved in the past. The Office for National Statistics estimates that quality-adjusted productivity in the NHS increased by an average 0.8% per year between 1995–96 (when the data series starts) and 2019–20, and 1.2% per year between 2009–10 and 2019–20. While not perfectly comparable to the measure of labour productivity used in the workforce plan, this gives a sense of the scale of the ambition – particularly as, since the onset of the pandemic, measured productivity performance in the NHS has been even weaker. 

8.1 Introduction

In January 2000, in what was subsequently dubbed ‘the most expensive breakfast in history’, then Prime Minister Tony Blair announced on Sir David Frost’s Sunday morning TV show that UK health spending would be increased to the European Union average. The then Chancellor Gordon Brown was rather displeased, and reportedly raged to the prime minister that ‘you’ve stolen my f***ing budget’ (Rawnsley, 2010). 

In June, NHS England published its much-awaited long-term workforce plan (NHS England, 2023). For the first time, the plan sets out official estimates of how many staff the NHS will need in the future and proposes a range of actions and targets to achieve this. Its publication was welcomed across the political spectrum. 

Many have discussed the strengths and weaknesses of the plan elsewhere (e.g. Black, 2023; Dixon, 2023; Murray, 2023; Palmer, 2023). But one of the less remarked-upon aspects of the plan is that it has probably stolen more than a decade’s worth of budgets, for funding the plan will require NHS spending in England to increase by around 2% of national income over 15 years, equivalent to around £50 billion in today’s terms. That will set the context and constraints in every Budget and Spending Review for at least the next decade. 

Here, we do not seek to evaluate the merits of the plan, its modelling assumptions or its achievability. Instead, we start from the assumption that the plan will be implemented in full and consider the potential implications for the size of the NHS workforce and NHS budget in England. While the plan includes funding for new training places, it does not estimate nor provide for the longer-term increases in funding that will be required to cover the salaries – and other costs – of an expanded workforce. 

The plan suggests the number of staff employed by the English NHS will increase from around 1.5 million in 2021–22 to between 2.3 and 2.4 million in 2036–37. This would be equivalent to average growth in the size of the NHS workforce of between 3.1% and 3.4% per year. For context, we estimate that NHS staffing numbers grew by around 1.1% per year between 2009–10 and 2019–20. Meeting the plan would almost certainly see the NHS account for a growing share of not just public sector employment but also the overall workforce. By 2036–37, we estimate that almost half (49%) of public sector workers in England will work for the NHS, compared with 38% in 2021–22 and 29% in 2009–10. Moreover, we estimate that one in eleven (9%) of all workers in England will work for the NHS, compared with one in seventeen (6%) in 2021–22 and 2009–10.

Such an expansion in the workforce has implications for the size of the NHS wage bill. First and most obviously, employing hundreds of thousands of additional people will increase staffing costs. But in addition, to attract and retain this many more workers, it seems likely that – at a minimum – NHS pay will need to keep pace with earnings in the wider economy. Taken together, we estimate that the workforce plan implies real-terms increases in the NHS wage bill of around 4.4% per year. Even if all other NHS spending (on non-staff items) were frozen in real terms for the next 15 years, this implies that the NHS budget would need to increase by around 2.3% per year, over and above inflation, to cover these higher staffing costs alone. 

But more staff will surely mean increased spending elsewhere, whether on new hospitals, equipment or medicines. We estimate that under a central set of assumptions, the workforce plan implies annual NHS budget increases of around 3.6% per year in real terms. This would be in line with the long-run average real-terms growth rate in UK health spending (3.6% per year from 1949–50 to 2022–23), but above the 2.4% per year seen since 2009–10. In other words, returning to the NHS’s long-run average funding growth rate could be enough to fund the workforce plan. But with prospects for productivity growth over the next decade weaker than what was realised over the first 60 years of the NHS’s life, continuing to deliver the same real-terms growth rate would mean harder trade-offs elsewhere. It would mean that funding for the NHS in England would increase by around 2.0% of GDP by 2036–37: similar to the increase in UK health spending over the 15 years to 2008–09 (2.2% of GDP) and equivalent to an extra £52 billion in today’s terms. By the NHS’s own estimates, though, this will only be enough to meet demand if productivity can be increased by between 1.5% and 2% per year: an extremely ambitious target. 

Stepping back, the publication of the NHS workforce plan is probably as close as any government is ever going to get to explicitly acknowledging the virtual inevitability of big increases in health spending over the medium term. That acknowledgement should be front and centre of the Chancellor’s fiscal event this autumn. 

8.2 What is in the plan?

The workforce plan is organised around three themes: training new staff, retaining existing staff and reforms to improve productivity. To train new staff members, the plan proposes large increases in medicine and nursing university places, as well as expansions to apprenticeships and alternative routes into clinical roles. For example, the plan aims to double the number of medical school places, increase the number of GP training places by 50% and almost to double the number of adult nursing training places (an increase of 92%), all by 2031–32. The plan includes additional funding of £2.4 billion over the next six years for this expansion in training. As the domestic supply of staff expands, the plan forecasts that the share of NHS staff recruited from overseas will fall from nearly 25% to around 10% by 2036–37 (specifically between 9% and 10.5%). 

To retain existing staff, the plan outlines a range of proposals, including increasing flexible working, improving the health of the workforce, and improving local and national culture and leadership. The objective is to reduce the annual NHS-wide leaver rate from 9.1% to between 7.4% and 8.2%, a target which is dubbed ‘a stretching but realistic trajectory’. 

Finally, the plan focuses on increasing productivity. One key part of this is the expansion of new roles – such as physician associates and nursing associates – that aim to free up time for more trained, traditional roles such as doctors and registered nurses. Another part of the plan to increase productivity is through the use of new technology, such as artificial intelligence and robotic assisted surgery. Overall, the plan is based on an ‘ambitious’ assumption that labour productivity will increase by between 1.5% and 2% per year. For context, the Office for National Statistics estimates that (quality-adjusted) productivity in the NHS increased by an average 0.8% per year between 1995–96 (when the data series starts) and 2019–20, and 1.2% per year between 2009–10 and 2019–201 .Since the COVID-19 pandemic began, measured productivity has fallen substantially (Warner and Zaranko, 2022). While not perfectly comparable to the labour productivity measure used in the workforce plan, this provides an indication of the degree of ambition of the plan’s assumptions. 

Importantly, as well as outlining a range of actions and objectives, the plan includes detailed workforce modelling. For each professional staff group within the NHS, the plan forecasts the full-time-equivalent (FTE) number of staff needed by 2036–37. It then forecasts how supply would change without the workforce plan (counterfactual supply) and how each aspect of the plan will increase or decrease the workforce over time. This leads to forecasts for the total workforce in each staff group in 2026–27, 2031–32 and 2036–37.

Here and throughout, our primary focus is on the number of workers employed by the NHS (defined as ‘substantive’ staff in post, plus those employed on a bank-only contract), as these are the groups relevant for the NHS paybill and the groups the workforce plan aims to increase. Agency staffing is implicitly assumed to fall to almost zero by 2036–37; indeed, reducing dependency on agency staffing is an explicit objective of the plan and one that, if achieved, could reduce budgetary pressures. We therefore exclude agency workers from our analysis of the workforce, but allow for an elimination of agency spending by 2036–37 in our funding projections. 

Figure 8.1 shows the plan’s target growth for each staff group between 2021–22 and 2036–37. The starting point includes substantive staff and an estimate for the number of ‘bank-only’ staff in each group. The plan includes a lower bound (the green bars) and an upper bound (the yellow bars) to help highlight the uncertainty over the impacts of actions taken in the plan. The first panel of the figure includes all professional staff groups. It is clear that the two emerging roles – advanced practitioners and nursing associates – are planned to increase dramatically, by far more than other, more traditional roles within the NHS. The number of advanced practitioners is forecast to increase from 8,300 in 2021–22 to 39,000 in 2036–37, an increase of 370%, while the number of nursing associates is forecast to increase from 5,300 to 64,000, an increase of 1,100%.

Related content

Taken together, the plan has two key implications for the NHS workforce. One is a substantial expansion in the numbers in all staff groups. Even the smallest increases are still large: a 29–37% increase in the number of midwives, for instance. The second implication is a change in the composition of the workforce – increasing the number of associate roles, and expanding the mental health sector relative to the acute and community sectors.

8.3 What does the plan mean for the size of the NHS workforce?

We now consider the implications of the plan for the total size of the NHS workforce. The plan includes all of those working in the hospital and community healthcare sectors, as well as GPs working in primary care. We do not include pharmacists or dentists in our analysis, as these were modelled separately in the plan and are not typically employed directly by the NHS. These are relatively small groups of staff and their exclusion will not materially alter our headline conclusions. 

Overall, the plan aims to increase the permanent NHS workforce from 1.4 million in 2021–22 to between 2.2 and 2.3 million in 2036–37, an increase of between 57% and 64%. 

This measure excludes temporary staff: those employed via an agency, and those employed by the NHS on ‘bank-only’ contracts. As described above, our preferred measure is the sum of ‘substantive’ staff in post and bank-only staff, all of whom are employed by the NHS, but excluding agency staff. On this definition, NHS staffing is planned to grow from about 1.5 million to between 2.3 and 2.4 million over the 15-year period: between 58% and 64%, or between 3.1% and 3.4% per year. 

Figure 8.2 shows how the size of the NHS workforce in England has changed since 2009–10, and how it will continue to grow if the workforce plan is met. Crucially, the forecast annual growth rate – between 3.1% and 3.4% per year – is far higher than the pre-pandemic growth rate of 1.1% per year between 2009–10 and 2019–20. Even after including the COVID-19 pandemic, the NHS workforce grew by 1.5% per year between 2009–10 and 2021–22, far lower than the growth rate implied by the workforce plan. The grey dashed line in Figure 8.2 illustrates what would happen to the NHS workforce if it continued to grow at this rate. For context, UK-wide NHS employment grew by 3.0% per year between 1999 and 2009, slightly slower than under the workforce plan (Office for National Statistics, 2023).

This large increase in staffing means that the NHS workforce will almost certainly make up a bigger share of the public sector and economy-wide workforce. In Figure 8.3, we estimate the future path for NHS England employment as a fraction of total public sector employment in England. To do so, we assume that the non-NHS public sector workforce grows in line with the Office for Budget Responsibility (OBR)’s forecasts for general government employment, and that NHS staffing grows in line with the workforce plan.

Figure 8.4 repeats this analysis for the economy-wide workforce in England. We assume that the non-NHS workforce grows in line with OBR projections for overall employment until 2027–28 (the end of the forecast period), and that it then grows in line with ONS projections for the working-age population. We estimate that the NHS workforce rose from 5.7% of the economy-wide workforce in England in 2009–10 to 6.0% in 2021–22. This is a much smaller increase than the increase in the NHS as a share of the public sector workforce because, whereas the public sector workforce was reduced in size over the 2010s, the economy-wide workforce increased in size. 

8.4 What does the plan imply for NHS funding?

We now consider what the workforce plan could mean for NHS funding. There are two direct costs of the increase in staffing: the costs associated with training new staff, and the costs of actually employing new staff (and retained existing staff). The plan includes funding for the former (staff training). We focus here on the additional and substantially larger cost of employing a much larger workforce, funding for which will (presumably) need to be provided at future Spending Reviews. We then consider the indirect implications of the workforce plan: if the NHS has between 58% and 64% more staff by 2036–37, it will almost certainly require more spending on other areas. To take one simple example, if the health service employs far more doctors, we might expect an increase in the amount of medicine being prescribed.

In this section, we focus on the impacts of the plan for NHS day-to-day (resource) spending, but the plan will also likely have implications for capital spending: such a large increase in staffing will also likely (at some point) require new equipment and new hospitals. As the plan itself states, to achieve the required productivity assumptions will require ‘a sustained increase in capital investment in the ageing NHS estate … [and] investment in digital infrastructure throughout the NHS’ (NHS England, 2023, page 109). 

Staff costs

We first consider the total staffing costs for NHS providers, including NHS hospital and community trusts. In 2021–22, NHS England’s resource budget was £146.5 billion (equivalent to £156.1 billion in 2022–23 prices) and it spent £66.2 billion on staff working for NHS providers, equivalent to 45.2% of its total budget (Department of Health and Social Care, 2023). This does not include staff employed by non-NHS providers, such as those working in primary care, dentistry, or community pharmacy. In our calculations of the wage bill, we therefore exclude the costs of employing the additional GPs (who have been included in our analysis up to this point) included in the plan, as well as dentistry and community pharmacy staff (who have not). Larger staff numbers in these other sectors of the NHS will implicitly be part of the increase in ‘other’ NHS costs that we subsequently consider.

According to the workforce plan, the total number of staff employed by NHS providers will grow by between 3.1% and 3.4% per year from 2021–22 to 2036–37. A very simple forecast, therefore, would be that the NHS wage bill will also grow by between 3.1% and 3.4% per year in real terms. Implicitly, this assumes that both real wages and the mix of professional groups stay constant between 2021–22 and 2036–37.

Neither of these assumptions seems plausible. As Figure 8.1 shows, the largest staffing increases in the plan are for new associate roles, which typically have lower wages than the traditional roles they are designed to be partial substitutes for. This change in professional mix will therefore reduce the real growth rate in the wage bill relative to the growth rate in staff numbers, though only very marginally: we estimate that annual growth in the wage bill might be reduced by around 0.05 percentage points as a result. 

More significantly, if the NHS is going to expand its workforce by more than half, it will need a strategy for attracting workers into the sector (and subsequently retaining them). That is almost certain to require real-terms wage increases, and highly likely to require pay increases that match – or perhaps even exceed – economy-wide earnings growth. 

The OBR forecasts that between 2024–25 and 2036–37, economy-wide nominal wages will grow on average by 3.3% per year (for the years before 2024–25, we use estimates of NHS pay deals). After accounting for expected inflation (as measured by the GDP deflator – the relevant measure in this case since we are focused on public service spending rather than household spending power), this would equate to average real-terms pay growth of 1.2% per year. We use this as our central scenario for future NHS pay growth. Combined with planned growth in the size of the workforce (taking the midpoint of the NHS’s two scenarios), this would lead to average growth in the real-terms wage bill of around 4.4% per year. 

Table 8.1 presents this central scenario, along with two alternatives. (The table also shows scenarios for non-pay spending, which are discussed in the next subsection.) In our ‘low’ scenario, we assume that the size of the NHS workforce grows according to the workforce plan’s lower bound and that wages grow 0.2 percentage points (ppt) slower than forecast economy-wide wages (1.0% per year, rather than 1.2%). This is to reflect the possibility that other aspects of NHS employment – such as job security or a valuable public sector pension – mean that the NHS can continue to recruit staff even as NHS wages fall relative to the average economy wage. In this case, we estimate that the NHS wage bill would grow on average by around 4.0% per year in real terms between 2021–22 and 2036–37. 

Table 1.1. Scenarios for estimated NHS funding growth if the workforce plan is achieved

Source: Authors’ calculations using NHS England’s long-term workforce plan (NHS England, 2023), NHS England’s NHS staff earnings estimates (various) (https://digital.nhs.uk/data-and-information/publications/statistical/nhs-staff-earnings-estimates), NHS Digital’s GP earnings and expenses estimates 2020–21 (https://digital.nhs.uk/data-and-