The NHS Productivity Gamble: Are We Confusing Ambition with Arithmetic?

Anyone who has been paying close attention to NHS policy over the past twelve months will not have been entirely surprised by the Financial Times's recent leak of the draft NHS workforce plan. Surprised by some of the language, perhaps — but not by the broad direction. The plan has been a long time coming. Wes Streeting originally committed to publishing a refreshed long-term workforce plan by the summer of 2025 (Nursing Times, 2024). That deadline came and went. By October 2025, the Health Service Journal was reporting that publication had slipped again, to spring 2026 (HSJ, 2025). Spring 2026 has now itself essentially elapsed — and still we wait. So when a draft finds its way to the FT, those who have been watching will recognise it less as a bolt from the blue and more as confirmation of a direction the government had already openly signalled.

Because the NHS 10 Year Health Plan, published in July 2025, had already made the position fairly clear. The 2023 workforce plan's projections — staff numbers rising from 1.4 million to 2.3 million by the mid-2030s — were described, with commendable candour, as a "fiction" (Nuffield Trust, 2025). The forthcoming workforce plan, the government indicated, would not ask "how many staff do we need to maintain our current care model over the next ten years?" but rather "given our reform plan, what workforce do we need?" (Pulse Today, 2025). That reform plan is built around three structural shifts: from hospital to community, from analogue to digital, and from sickness to prevention (NHS Employers, 2025). The workforce plan is being designed to staff a fundamentally different service — not a scaled-up version of the one we have today.

All of which is perfectly logical, and in many respects the right question to be asking. But then the draft does something that lifts it out of the realm of sensible service redesign and into considerably more contested territory. Buried in the document is the proposition that technology can "completely substitute for a role." Not support. Not augment. Substitute. Roy Lilley, whose forensically sharp blog on these matters (Lilley, 2026) is essential reading alongside the FT piece, spotted that word choice immediately — and rightly so. It is doing a remarkable amount of heavy lifting.

Before we reach the AI question, though, it is worth pausing on the broader context, because the NHS's productivity challenges don't exist in isolation. They are, in many respects, a faithful reflection of something that has been vexing economists and policymakers for the better part of two decades.

Britain's Productivity Problem: A National Pastime

The UK has a productivity problem. This is not a controversial statement — it is, at this point, practically a national characteristic. Before the 2008 financial crisis, productivity (output per hour worked) was growing at around 2.2% per year. Since then, it has averaged a rather dispiriting 0.5% (Pettinger, 2025). Economists have taken to calling this the "productivity puzzle," which is the polite way of saying that nobody is entirely sure why it happened, though there is no shortage of theories.

The usual suspects include chronic underinvestment — the UK's total investment as a share of GDP is among the lowest in the developed world — inadequate infrastructure, a persistent skills gap, and, perhaps most damningly, poor management quality. Research cited by former Bank of England Chief Economist Andy Haldane found that the UK has roughly double the proportion of firms with low management scores compared to the US and Germany, and that management quality is a statistically significant determinant of productivity (Biddle et al., 2024). We are not just under-investing in kit; we are under-investing in the people who are supposed to deploy it.

The public sector has fared no better. EY analysis found that if public sector productivity growth had merely kept pace with the private sector between 2019 and 2024, UK GDP would be 3% larger today — worth around £80 billion a year in lost output, a figure that could rise to £170 billion by 2030 (EY, 2025). Over that same period, inputs into public services grew by nearly 25%, while outputs grew by just 14%. According to the ONS, overall public service productivity in 2024 remains 4.6% below pre-pandemic levels — and in healthcare specifically, the picture is identical (ONS, 2025; Productivity Lab, 2025). This is the landscape against which the new workforce plan is being drawn.

Productivity and Efficiency: Not the Same Thing (And This Really Matters)

These two words are routinely used interchangeably in policy documents and political speeches, often by people who really ought to know better. They are not the same thing — and conflating them here is more than a semantic quibble; it goes to the heart of what this plan is actually trying to do.

The Health Foundation (n.d.) offers as clear a distinction as any: efficiency is about making the best possible use of available funding — reducing cost per unit of activity. Productivity is about improving the quantity or quality of health outcomes using the same amount and type of resource. The King's Fund (2019) frames it helpfully: there is a limit to what can be achieved by squeezing efficiency out of a fixed resource base, and beyond that limit, genuine productivity means actually doing more — or doing it better — with what you have.

That distinction matters enormously here, because it reframes what the new workforce plan is actually proposing. The plan seeks to reduce annual staffing growth from the 2.6–2.9% per year envisaged under the 2023 plan to 1.1–2%, resulting in up to 380,000 fewer people in the NHS by the mid-2030s than previously forecast (Financial Times, 2026). AI will, apparently, close the gap. But here is the question the plan seems to sidestep rather elegantly: if the goal is to deploy technology to replace people — to do the same amount with fewer staff at lower cost — that is an efficiency play. A cost-reduction exercise dressed in the language of transformation. Genuine productivity would be using that same technology to enable existing staff to do more — see more patients, reduce waiting times, clear backlogs, address conditions that currently go undetected because people cannot get a GP appointment. These are fundamentally different ambitions, with fundamentally different outcomes for patients.

The Theory of Change: What the Three Shifts Are Actually Supposed to Do

The three shifts are not simply a restructuring exercise. They represent a theory of change about how the NHS's demand profile, operating model, and staffing requirements will evolve over the next decade. It is worth examining each in turn, because the logic underpinning the workforce plan depends heavily on how plausible that theory actually is.

The shift from sickness to prevention is, in many ways, the most ambitious of the three. The proposition is that by encouraging the population to take greater responsibility for its own health — through better access to preventive services, earlier diagnosis, lifestyle interventions, and a genuine reorientation of NHS infrastructure towards community settings — the inexorable rise in demand for primary, secondary, and tertiary care can be meaningfully moderated. The evidence that prevention is cost-effective in certain areas is genuine and growing (Institute for Government, 2024), and there is good reason to support this direction of travel. But there is an important and frequently overlooked caveat: prevention delivers its financial returns over long time horizons. Research published in the Oxford Review of Economic Policy found that even where primary prevention measurably improves population health, its effects will not reduce spending pressures in the short or even medium term — because the biggest impacts are felt when younger cohorts begin to enter old age, which is where morbidity and healthcare consumption begin to rise sharply (Stoye et al., 2025). Prevention is a long-term investment in a healthier population, not a near-term lever on workforce numbers. It would be wishful thinking to plan around it as if it were the latter.

The shift from hospital to community raises a rather different set of questions — and, it has to be said, a rather important one for those doing the workforce arithmetic. Moving care closer to home is genuinely desirable: it is often better for patients, it can relieve pressure on overcrowded hospitals, and virtual wards, remote monitoring, and integrated community teams are showing real promise in specific pathways. NHS England's own operational framework acknowledges growing evidence that virtual wards, when implemented well, can provide a better patient experience and help narrow the gap between demand and capacity for hospital beds (NHS England, n.d.). But here is the point that tends to get somewhat glossed over in the enthusiasm for this shift: moving care out of hospitals does not make the care disappear. It still needs to be provided — by nurses, therapists, paramedics, community pharmacists, GPs, and a whole range of support staff, working in different places and often in more complex, less supervised contexts. The British Geriatrics Society was admirably direct when commenting on virtual ward expansion: the NHS faces record workforce shortages, they noted, and "simply moving staff around from one part of the NHS to another can only work as a short-term fix" (NHS Support Federation, 2025). The Neighbourhood Health Service model envisaged by the 10-year plan — multidisciplinary community teams, Neighbourhood Health Centres open 12 hours a day, six days a week — is not a staffing reduction strategy. It is, more accurately, a staffing redeployment and retraining strategy, albeit one that will, over time, require different skills, different career structures, and potentially more staff in community settings than currently exist. The staff don't vanish. They move.

Which brings us back to the digital and AI shift, which we have already explored at some length in the context of the efficiency/productivity distinction. Technology can undoubtedly change how work gets done, what can be automated, and the scope of what individual clinicians can achieve in a given working day. Remote monitoring extends a clinician's reach; AI-assisted diagnostics can accelerate triage and reduce error; virtual consultations remove unnecessary travel and expand access for patients who struggle to attend in person. All of this is real, and most of it is genuinely welcome. But the question the workforce plan needs to answer more honestly is this: will these technologies enable existing staff to deliver better healthcare — to tackle backlogs, improve outcomes, extend access to underserved populations, and genuinely transform patient experience — or will they simply mean that staffing does not need to grow at quite the same pace as the population expansion and demographic pressures that are coming regardless?

These are not the same ambition. The first is transformational. The second is arithmetical. And the risk embedded in the current framing is that we reach for the grand language of transformation while doing the arithmetic — and then find, a decade from now, that we have built an NHS that is leaner but not better; one that has held costs steady by managing demand down, shifting care sideways, and automating at the margins, but that has not fundamentally improved what it can do for the people who depend on it.

The Demand Sponge

Roy Lilley (2026) makes another point that deserves amplification here: the NHS is, as he puts it, a "demand sponge." Technology rarely reduces demand in healthcare — it reveals more of it. If AI enables faster consultations, cheaper diagnoses, or earlier detection of disease, more patients will be seen, more conditions identified, more follow-up generated, and more costs incurred. This is not a criticism of AI; it is simply the nature of a universal, free-at-the-point-of-use health service operating against a backdrop of an ageing population, rising multimorbidity, and expectations that have been ratcheting upwards for decades.

The Institute for Fiscal Studies (2025) estimated that the previous workforce plan would have seen the NHS employing 9% of all workers in England by 2036–37, up from 6% today, adding around £50 billion in costs. The draft plan is a response to the uncomfortable reality that such growth is simply unaffordable — a reality the plan itself describes, with rather dramatic flourish, as "a path to financial ruin." The IFS's Paul Johnson acknowledged the 2023 plan "would have meant a big increase in spending," and nobody is seriously arguing for that.

So the honest account of what the new workforce plan proposes is this: rather than building an NHS that can genuinely do more for a growing population, we are building one that can accommodate growing demand within a constrained cost envelope, using technology to prevent the staffing bill from becoming ruinous. That is, let us be clear, a legitimate and arguably necessary goal. But let us not dress it up as something it isn't. At its core, this is a plan to build a more affordable NHS — not necessarily a more productive one.

Can the NHS Actually Do This?

Here it is worth drawing heavily on Roy Lilley's analysis (2026), which cuts through to some uncomfortable truths about this aspect of the plan with characteristic precision.

The assumption embedded in the strategy is that AI can be deployed at scale across the NHS to "completely substitute" for certain roles — reliably, safely, and in ways that maintain or improve patient outcomes. As Lilley observes, the NHS has never struggled to buy technology. It has always struggled to redesign work around it. Digital maturity across the service is, to put it diplomatically, wildly uneven. Some trusts are digitally sophisticated; others wrestle with systems that belong in a museum. Data quality, interoperability, and governance frameworks vary enormously. And this matters because, as Lilley rightly notes, AI is not like buying a new MRI scanner and plugging it in — it changes workflow, staffing models, decision-making, and accountability. It requires sustained organisational competence, distributed consistently across what is not a single machine with one operating system, but a federation of semi-autonomous organisations with variable capability. The risk Lilley identifies is obvious: the organisations most in need of productivity improvements may be the least capable of implementing them safely, producing a two-speed NHS that compounds existing inequalities rather than resolving them.

Paul Johnson's observation in the FT piece captures this neatly: hopes of transforming the NHS through AI had to be met with "a degree of scepticism about the capacity of an organisation that struggles to use 2005 technology, let alone anything more recent" (Financial Times, 2026).

There are also safety questions that deserve more than a footnote. A 2025 Nuffield Trust survey found that almost nine in ten non-users of AI clinical tools cited medico-legal risk as a key concern — but so did eight in ten users (Skills for Health, 2026). Experience of the technology does not dissolve the anxiety; it recontextualises it. Safety standards have not kept pace with deployment rates: clinical risk standards were designed for fixed-logic software, and modern AI is neither fixed nor entirely predictable (Digital Health, 2025). The MHRA launched a new National Commission into the Regulation of AI in Healthcare in December 2025 — welcome, but also a regulatory framework being assembled while the technology is already accelerating into the system (MHRA, 2025). Meanwhile, public support for AI in direct patient care stands at only 54%, according to the Health Foundation (2026) — a majority, but not a comfortable foundation for a plan that appears to envisage AI playing a central, substitutive role in clinical care.

The Right Gamble?

For those balancing the books — the Treasury, the new Health Secretary James Murray (himself ex-Treasury, a detail the FT notes with quiet amusement), and the officials trying to make the arithmetic work — the answer is almost certainly yes. The alternatives are either considerably more expensive or considerably more politically unpalatable, and the previous government's 2023 workforce plan was, as subsequent analysis showed, a document that described what the NHS needed without fully confronting whether the country could afford to provide it. That gap had to close eventually.

But for those whose ambition is something grander — a genuinely world-class health system delivering the kind of quality, timeliness, and personalisation that modern technology ought to make possible — the plan as reported feels less like a bold transformation and more like a managed retreat. Replacing staff with technology to hold costs steady is cost containment, not productivity growth. Both things can be valuable, but they require honest labels.

So Why Can't the Plan Do Both?

This is, of course, the obvious challenge to the argument so far — and it deserves a serious answer rather than a rhetorical sidestep. Why should affordability and genuine productivity improvement be mutually exclusive? Can a well-designed workforce plan not achieve a more cost-sustainable NHS and a better one?

The short answer is: yes, in principle, it can. And there is real evidence that technology and smarter ways of working can deliver both simultaneously when the conditions are right. Lord Darzi, in his independent review of the NHS, was admirably direct on this point: "Regrettably, productivity in the NHS has all-too-often become associated with simply spending less or working harder. Neither is correct" (quoted in Public Policy Projects, 2026). The newly established NHS Productivity Commission, for which the Health Foundation is providing the evidence base, goes further still, arguing that system productivity should be defined as the health care value created for every pound invested — and explicitly rejects the narrow view of productivity as activity throughput (Health Foundation, 2026b). When the Commission put out its call for evidence, respondents were pointed in their criticism of current measurement approaches: measuring productivity primarily as volume of appointments or procedures, they argued, creates a dangerous incentive to prioritise activity over outcomes and can actively drive the wrong behaviours (Health Foundation, 2026b).

Genuine examples of technology achieving both cost savings and improved patient outcomes do exist within the NHS. When University Hospitals Sussex deployed the HealthRota e-rostering system, the result was not only a significant reduction in locum costs but also a 68% reduction in emergency calls at weekends — better staffing led directly to better clinical safety (NHS Providers, n.d.). Population health analytics in Lancashire enabled community teams to identify unmet needs among high-intensity service users and provide more personalised support, improving outcomes while reducing unnecessary acute admissions. These are not trivial examples, and they point to a genuinely promising direction of travel.

The problem is not that it cannot be done. The problem is the conditions under which it can be done — and whether the workforce plan, as currently framed, creates those conditions or merely assumes them.

Achieving both genuine productivity improvement and greater affordability requires, at minimum, four things that the current plan's framing does not yet make sufficiently explicit. First, it requires sustained upfront investment — not just in technology but in the organisational change, training, and change management capacity needed to redesign work around that technology. NHS Providers' analysis of the barriers to realising productivity gains from digital transformation is unambiguous on this point: investment in new ways of working is required upfront, in both capital and revenue spend (NHS Providers, n.d.). The government's £2 billion technology commitment from the 2024 Autumn Budget is a meaningful step, and NHS productivity growth is currently running ahead of the 2% target set for 2025/26 — reaching 2.6% in the acute sector (NHS England, 2026). These are genuinely encouraging signals. But technology investment without the accompanying organisational development tends to produce digitised versions of old inefficiencies rather than genuinely transformed services.

Second, it requires measuring the right things. If success is defined as headcount ratios and cost per episode, that is what will be optimised — regardless of what happens to patient outcomes, waiting times, or the quality of care received. The Health Foundation's Productivity Commission call for evidence found widespread concern that current productivity metrics drive volume at the expense of value (Health Foundation, 2026b). Carnall Farrar's analysis suggests there are potentially £15–27 billion in productivity opportunities in the acute sector alone — but capturing them requires outcome-based commissioning, value-based incentives, and longitudinal patient data, none of which are straightforward to implement at speed (Carnall Farrar, 2025).

Third, it requires time — more of it than the workforce planning cycle typically allows. NHS England's own productivity plan explicitly acknowledges a two-phase approach: near-term operational efficiency to close the current productivity gap, followed by longer-term technology-enabled productivity that genuinely shifts the frontier of what is possible (NHS England, 2026). The risk embedded in the current workforce plan is that the headcount assumptions — the 380,000 fewer staff — are locked in based on the expectation of transformation that has not yet been demonstrated at scale, and may not materialise within the timeframe the arithmetic requires.

Fourth, and perhaps most fundamentally, it requires honesty about sequencing. Efficiency gains tend to materialise quickly; genuine productivity improvements, particularly those that improve outcomes, take longer. If the workforce plan is structured in a way that delivers the efficiency savings first and defers the investment in transformation, there is a serious risk that we find ourselves, a decade from now, with a leaner NHS that has not become a better one — having banked the headcount reductions before the conditions for genuine productivity growth were actually in place.

None of this is an argument against the ambition. The goal of a more productive and more affordable NHS is not only legitimate — it is the only sustainable future the service has. But ambition without the right architecture for delivery is not a plan. It is a hope.

What This Plan Actually Needs to Say

The NHS has carried the weight of two conversations at once for too long: the one about what it should be, and the one about what we can afford it to be. The workforce plan risks conflating them — and in doing so, it may obscure the real choices being made and the real conditions required to deliver on both.

If the plan is genuinely committed to both affordability and productivity — to building an NHS that costs less to run and delivers better outcomes — then it needs to be explicit about what that requires. It needs an investment framework that funds not just technology procurement but the organisational capability to deploy it well. It needs a measurement framework built around patient outcomes and health value, not just activity volumes and headcount ratios. And it needs a sequencing framework that ensures the efficiency savings are not banked before the productivity gains have been demonstrated. Without these, "completely substitute for a role" remains not a transformation strategy but a cost-cutting instruction — and the gap between the plan's ambition and its architecture will be felt, eventually, by the patients it is supposed to serve.

The difference between productivity and efficiency is not just an economist's distinction. In a health service, it is the difference between a better NHS and a cheaper one. Both are legitimate aims — and both, with the right plan, are achievable. But achieving them together demands more than bold language and a favourable set of workforce projections. It demands the kind of honest, rigorous, outcome-focused planning that the NHS — and the people who depend on it — genuinely deserves.

References

Biddle, M., Bryson, A., Corfe, S., Fox, A., Green, F. and Wilkinson, D. (2024) 'The UK Productivity Puzzle: A Survey of the Literature and Expert Views', The Service Industries Journal, 44(9–10). Available at: https://www.tandfonline.com/doi/full/10.1080/13571516.2024.2367818

Digital Health (2025) 'We Need to Act Fast to Close the NHS AI Safety Gap', Digital Health, 20 August. Available at: https://www.digitalhealth.net/2025/08/we-need-to-act-fast-to-close-the-nhs-ai-safety-gap/

EY (2025) Mind the Productivity Gap: The Public Sector Potential. London: EY. Available at: https://www.ey.com/content/dam/ey-unified-site/ey-com/en-uk/newsroom/2025/08/ey-mind-the-productivity-gap-report.pdf

Financial Times (2026) 'NHS Plans to Cut Recruitment and Use AI to Avoid "Financial Ruin"', Financial Times, May. [Leak of draft NHS workforce plan, as reported.]

Health Foundation (n.d.) Efficiency and Productivity of the Health and Social Care System. London: The Health Foundation. Available at: https://www.health.org.uk/what-we-do/sustainability-of-health-and-social-care/efficiency-and-productivity-of-the-health-and-social-care-system

Health Foundation (2026) Attitudes to Technology and AI in Health Care. London: The Health Foundation. Available at: https://www.health.org.uk/reports-and-analysis/analysis/attitudes-to-technology-and-ai-in-health-care

Institute for Fiscal Studies (2025) The Outlook for Public Sector Productivity. London: IFS. Available at: https://ifs.org.uk/publications/outlook-public-sector-productivity

King's Fund (2019) 'The NHS Needs To Be More Productive – Or Is It More Efficient?', King's Fund Blog, 12 March. Available at: https://www.kingsfund.org.uk/insight-and-analysis/blogs/nhs-productive-or-efficient

Lilley, R. (2026) Roy's Blog: The NHS Workforce Plan — AI, Substitution, and the Challenge No One Is Naming, May. Available at: https://www.nhsmanagers.net

MHRA (2025) 'MHRA Seeks Input on AI Regulation at "Pivotal Moment" for Healthcare', GOV.UK, 18 December. Available at: https://www.gov.uk/government/news/mhra-seeks-input-on-ai-regulation-at-pivotal-moment-for-healthcare

Office for National Statistics (2025) Public Service Productivity, Quarterly, UK: October to December 2024. Newport: ONS. Available at: https://www.ons.gov.uk/economy/economicoutputandproductivity/publicservicesproductivity/bulletins/publicserviceproductivityquarterlyuk/octobertodecember2024

Pettinger, T. (2025) '10 Reasons for UK's Fall in Productivity Growth', Economics Help, December. Available at: https://www.economicshelp.org/blog/214546/economics/10-reasons-for-uks-fall-in-productivity-growth/

Productivity Lab (2025) Public Sector Productivity: Measurement Issues, Trends, Challenges and Outlook. Available at: https://lab.productivity.ac.uk/insights/public-sector-productivity-measurement-issues-trends-challenges-and-outlook/

Skills for Health (2026) 'AI in Healthcare: What Clinicians and NHS Leaders Need to Know in 2026', Skills for Health, 21 April. Available at: https://www.skillsforhealth.org.uk/article/ai-in-healthcare-what-you-need-to-know-in-2026/

Health Service Journal (2025) 'Government Delays New NHS Workforce Plan', HSJ, 31 October. Available at: https://www.hsj.co.uk/workforce/government-delays-new-nhs-workforce-plan/7040250.article

NHS Employers (2025) 'Understanding the Workforce Implications in the 10 Year Health Plan'. Available at: https://www.nhsemployers.org/news/understanding-workforce-implications-10-year-health-plan

Nursing Times (2024) '"Refreshed" NHS Long Term Workforce Plan Confirmed for 2025', Nursing Times, 9 December. Available at: https://www.nursingtimes.net/workforce/refreshed-nhs-long-term-workforce-plan-confirmed-for-2025-09-12-2024/

Nuffield Trust (2025) 'Plan B: What the Forthcoming NHS Workforce Strategy Should Not Ignore'. Available at: https://www.nuffieldtrust.org.uk/news-item/plan-b-what-the-forthcoming-nhs-workforce-strategy-should-not-ignore

Institute for Government (2024) A Preventative Approach. London: Institute for Government. Available at: https://www.instituteforgovernment.org.uk/sites/default/files/2024-05/preventative-approach-public-services_0.pdf

NHS England (n.d.) Virtual Wards Operational Framework. Available at: https://www.england.nhs.uk/long-read/virtual-wards-operational-framework/

NHS Support Federation (2025) Virtual Wards. Available at: https://nhscampaign.org/issues/staff-shortages-copy-9/

Stoye, G., Warner, M. and Zaranko, B. (2025) 'Affording the NHS: Estimating Demand Pressures and the Options for Addressing the Challenge of Fiscal Sustainability', Oxford Review of Economic Policy, 41(1). Available at: https://academic.oup.com/oxrep/article/41/1/195/8157941

Carnall Farrar (2025) Value in Health: Improving Productivity, Quality and Prevention in the NHS. Available at: https://www.carnallfarrar.com/value-in-health-improving-productivity-quality-and-prevention-in-the-nhs/

Health Foundation (2026b) 'We Asked How the NHS Can Boost Productivity: Here's What We Heard', Health Foundation Blog, May. Available at: https://www.health.org.uk/features-and-opinion/blogs/we-asked-how-the-nhs-can-boost-productivity-here-s-what-we-heard

NHS England (2026) NHS Productivity Plan Update. Available at: https://www.england.nhs.uk/long-read/productivity-plan-update/

NHS Providers (n.d.) Digital Transformation and the Productivity and Efficiency Challenge: Sources of Productivity from Technology in the NHS. Available at: https://nhsproviders.org/resources/digital-transformation-and-the-productivity-and-efficiency-challenge/sources-of-productivity-from-technology-in-the-nhs

Public Policy Projects (2026) 'Safer Care, Better Outcomes: Unlocking NHS Productivity', January. Available at: https://publicpolicyprojects.com/safer-care-better-outcomes-unlocking-nhs-productivity/

Pulse Today (2025) 'Refreshed NHS Workforce Plan to Include "Fewer Staff than Projected"', Pulse, 7 July. Available at: https://www.pulsetoday.co.uk/news/workforce/refreshed-nhs-workforce-plan-to-include-fewer-staff-than-projected/