Cutting Agency Spending is Not Workforce Planning

A constructive critique of the REC’s intervention on NHS temporary staffing

The recent press release from the Recruitment & Employment Confederation, authored by Neil Carberry, raises an important and legitimate challenge: whether the Department of Health and Social Care’s instruction to reduce agency use by 30 per cent in the short term represents a blunt policy instrument that risks unintended consequences for patient care.

That question deserves serious consideration. It is also right that agencies, as part of the NHS workforce ecosystem, expect decisions of this scale to be grounded in evidence and transparency. This piece is therefore offered in the spirit of helping sharpen the argument, not undermining it, and of contributing to a more productive dialogue between policymakers, Trust leaders and workforce suppliers.

Where the debate should start

Any discussion about temporary staffing must begin with system-wide trends, not individual Trust anecdotes.

Chart 1: Total NHS temporary staffing expenditure by quarter (agency, bank and substantive), England, Q3 2019/2020 to Q3 2025/26 (£m)

Source: NHS Financial Performance Reports

This chart, drawn from NHS England quarterly financial returns, shows that after a peak in spending in the immediate post-pandemic recovery period both bank and agency spending have been falling since 2023/24 Q4. This matters because the underlying policy intent is often misunderstood. The objective is not to replace agency with bank (which we saw with pre-pandemic policies to curb agency spending), it is to reduce overall reliance on temporary staffing as a proportion of the workforce.

The REC press release draws attention to real historical growth in bank expenditure. What it does not fully acknowledge is that the most recent data shows both bank and agency spending falling in parallel. Any critique of policy effectiveness needs to anchor itself in that most recent trajectory, not earlier peaks[1]. Admittedly, the impact of reduced workforce spending has been greatest on agency (with simple logistics dictating that providers at the end of the chain of supply will be hit first and hit hardest). But the picture isn't one of substitution, it is one of reduction.

Why “impact assessments” are a fragile foundation for the argument

A core claim in the press release is that Trusts failed to undertake formal impact assessments on patient outcomes before cutting agency staffing. While understandable as a framing device, this assumption risks weakening the argument because it does not reflect how workforce planning decisions are actually made in the NHS.

Workforce decisions are rarely taken in isolation or through discrete assessments of individual policy changes. They are the product of continual trade-offs between affordability, minimum safe staffing requirements, recruitment feasibility and operational risk.

This point has been made repeatedly by NHS finance and workforce leaders in public forums. The point is that boards are not choosing between “safe” and “unsafe” staffing, but between “different levels of managed risk within a fixed financial envelope”. [2] They are also making tough decisions about what services they can continue.. and what they can't.

That is critical to understanding how staffing decisions are currently being governed. Under today’s financial rules, Trusts are operating with far less tolerance for overspend and far less expectation of retrospective bailout than in previous years. Temporary staffing, particularly agency staffing, is one of the few areas of discretionary expenditure that can be adjusted at pace. It is therefore unsurprising that it has become a focal point for financial recovery activity.

Importantly, however, this does not mean that safe staffing considerations are being set aside. Rather, many Trusts are increasingly approaching the issue upstream, through explicit work to redesign services, pathways and roles in ways that reduce establishment requirements safely. This shift is now routinely referenced in Trust Board papers, where organisations describe efforts to reshape demand, streamline workflows and change skill mix so that staffing requirements themselves are reduced, rather than workforce simply being cut.

In that context, the absence of a discrete document labelled as an “impact assessment” should not be interpreted as an absence of consideration. It reflects the fact that workforce risk is being addressed through existing governance structures and planning mechanisms. These include board risk registers, service-level quality impact assessments, formal service redesign programmes and financial recovery plans that explicitly link affordability, workforce configuration and patient safety.

It is also the case that the universally accepted tool used to determine nursing numbers in acute settings (the safer nursing care tool), is under scrutiny, as finance leaders are concerned that the tool fails to sufficiently adjust for innovation, digitisation, different skills mix or changes in environment. The output of more granular assessments of how work can be done safely - but 'differently' and based more on tasks than roles - will inevitably lead to further demand reductions in the immediate future. [3]

The balancing act between patient need and affordability conspires to create an enviornment in which a persistent backlog in demand, corridor care, staff burnout and recruitment freezes (and, in some areas, redundancies), are all features of the system at the same time.

What the policy is trying to achieve, explicitly

Chart 2: Substantive FTE numbers, indexed, England, 2022/23 to 2025/26 £m

Source: NHS Digital Workforce Data

While temporary staffing spend has fallen since mid-2024 this chart shows that substantive workforce numbers have grown - particularly among professionally qualified clinical staff. That is not incidental. It reflects a clear and longstanding policy preference to increase the proportion of care delivered by staff employed directly by the NHS under substantive or flexible contracts. This preference has been articulated consistently in national policy. NHS England’s Long Term Workforce Plan is explicit that the service has “relied too much on temporary staffing” and that agency staffing in particular “offers poor value for money for the taxpayer”, setting out a strategic intent to reduce agency use materially by expanding substantive supply and retention. That direction has since been reinforced through formal instructions to providers to deliver substantial reductions in agency spend, alongside expectations to curb bank usage.

Taken together, these policy signals make clear that the objective is not simply to rebalance demand between agency and bank staffing, but to reduce reliance on temporary labour overall by rebuilding the NHS’s employed workforce. This position is widely understood across national and regional leadership and underpins the financial and workforce decisions Trusts are now being asked to make. [4] More and more we are seeing Trust workforce leads migrating agency (and bank) workers into substantive roles.

There are also political considerations that are rarely stated explicitly but remain influential. The perception that significant volumes of taxpayer funding flow into private staffing companies, some backed by venture capital, sits uncomfortably with both ministers and senior officials. Equally, there is a strong belief that staff trained at public expense should, where possible, be employed directly by the NHS rather than supplied back into the system for a fee by third parties.

There is also a growing evidence base linking high levels of temporary staffing to poorer patient outcomes, including higher mortality. The mechanism is not the competence of individual temporary staff, but the absence of continuity, team integration and organisational familiarity that comes with sustained reliance on non-substantive labour. Staff employed directly by the NHS, working as part of stable teams, are generally believed to deliver care more efficiently and safely, especially in high-acuity settings.

Agree or disagree, these views explain why blunt controls have been applied. Ignoring them risks talking past the real drivers of policy.

Why the bank versus agency cost debate stalls progress

The REC press release spends considerable time demonstrating that bank staffing can be as expensive as, or more expensive than, agency staffing. That argument is not without merit in specific cases, but it is unlikely to influence policymakers because it does not engage with the underlying objective. The system is not choosing between bank and agency as equivalent substitutes. It is attempting to reduce both.

Internal planning guidance for 2024/25 and 2025/26 has consistently treated a 10 per cent reduction in bank expenditure as broadly equivalent, in system effect, to a 30 per cent reduction in agency spend. Medium-term planning guidance (covering 2026/27 and 2027/28) maintain these targets. As previously stated, the intent is contraction, not substitution.

Framing the debate as a contest between bank and agency risks locking all parties into a sterile argument about unit costs, rather than addressing the workforce planning challenges of managing inefficiency when temporary labour is used to cover persistent vacancies but realising efficiencies when it is used to create agility and minimise risks.

Where the REC argument is strongest

Where the press release is most compelling is in its warning against scapegoating agencies and in its call for a more honest conversation about flexibility. Even a fully staffed NHS requires access to temporary labour to manage short-term shocks, seasonal pressures, sickness absence and unexpected surges in demand.

Several NHS leaders have acknowledged this explicitly. Chris Hopson has repeatedly argued that temporary staffing is not a failure of leadership but a structural feature of a complex system operating under constant pressure, provided it is used intentionally and proportionately. [5]

The question then becomes 'what channels of supply does the NHS actually need?'. We know that well-regulated agencies provide speed, skill mix and resilience that internal systems alone cannot always deliver. This is where they excel and where they have traditionally provided a 'critical backstop' which has allowed wards to stay open. This is the space where the debate should move.

A more helpful place to land

If the objective is to influence future policy, the most productive starting point may be simpler than the current argument suggests. What is the NHS trying to achieve? A substantively staffed workforce wherever possible. Internal flexibility that absorbs routine variation. And a controlled, transparent 'agency' market that meets genuine short-term needs without becoming a substitute for workforce planning.

If that framing sounds familiar, it is because it closely reflects how the system was originally intended to function when NHS staff banks were formalised in the late 1990s and early 2000s, as vacancy rates began to rise and workforce flexibility became a structural requirement. Banks were designed as the primary mechanism for managing predictable workforce gaps and routine variation within a Trust’s own establishment, keeping staff under direct NHS employment, governance and clinical leadership. Agencies, by contrast, were intended to act as a secondary safety valve, deployed selectively to manage genuinely short-term shocks, sudden spikes in demand or hard-to-source skills. Evidence from subsequent workforce reviews and policy papers suggests that this layered model, substantive workforce first, bank as the default flexibility, agency as a targeted backstop, delivers better continuity, stronger organisational control and clearer accountability than systems that allow agency staffing to substitute for unresolved vacancies. The argument, therefore, is not for a new workforce architecture, but for a recommitment to one that was explicitly designed to balance safety, affordability and resilience, and which may function more effectively than the blurred and reactive arrangements that have emerged over time.

That framing creates space for agencies as part of the solution rather than as an ideological foil. It also aligns more closely with the direction of travel already visible in the data and apparent on 'the front line' in terms of Trust behaviour.

The REC is right to press for evidence, transparency and engagement. The next step may be to anchor that challenge more firmly in current financial trends, clearer policy intent and a shared understanding of how workforce decisions are being made in practice.

That is a conversation worth having, and one that the sector would benefit from having together.

References

[1] NHS England. Quarterly financial returns to NHS England, 2023/24 to 2025/26. Published quarterly.

[2] The Kings Fund.Tight budgets and tough choices: the reality of an NHS living within its financial means , May 2025.

[3] Health Service Journal. Hospitals 'needlessly increasing nurse numbers', December 2025

[4] NHS England. Fit for the Future, 10 Year Health Plan for England, 2025.

[5] UK Parliament, Health and Social Care Committee. Workforce burnout and resilience in the NHS and social care (June 2021), section summarising Chris Hopson’s evidence.