Key Takeaways
The government claims substantial consultancy savings, but spending on professional services has risen. Changes in expenditure classifications make it difficult to assess whether reliance on external suppliers is genuinely declining.
KPMG and EY have secured civil service training contracts worth up to £456 million, even as the government seeks to bring more training in-house. The awards illustrate the difficulty of changing established supplier relationships.
Reducing dependence requires more than spending controls. Government must rebuild internal expertise, retain knowledge and develop the ability to challenge, replace or dispense with external advisers.
Britain wants to spend less on consultants, and yet, the Big Four continue to win substantial public sector work. Whether government can become less dependent on the firms requires a closer look at what it is buying, what it learns and what happens when a contract ends.
On 15 October, City AM reported Tussell figures showing that Deloitte, KPMG, EY and PwC had secured more than £1.5 billion in public sector contracts during 2026, compared with £1.04 billion throughout 2025. Deloitte led with £650 million, followed by KPMG with more than £413 million, EY with £260 million and PwC with more than £211 million.
Those figures describe contract awards, potentially covering several years, rather than money paid out during 2026. They also cover public sector work beyond consultancy narrowly defined. They cannot establish, on their own, that ministers have broken their spending pledge.
They do, however, invite a broader question. What would demonstrate that government is becoming less dependent on the firms?
The bill can fall while the relationship remains extensive
In November 2024, the Cabinet Office announced new controls intended to curb unnecessary consultancy expenditure. Georgia Gould, then parliamentary secretary at the Cabinet Office, promised to “halve government spending on consultancy in future years”. The announcement also connected the reforms to improving civil service skills and reducing reliance on larger firms.
The government has subsequently reported reductions. In a 31 July 2026 response to the Public Accounts Committee, the Cabinet Office put combined spending on consultancy, contingent labour and professional services at £12.595 billion in 2024/25, down from £13.508 billion in 2023/24. Consultancy and contingent labour expenditure fell, while professional services expenditure increased. The reported total reduction was approximately £913 million.
The government has separately said that consultancy spending fell by more than £600 million in 2024/25, exceeding its £550 million savings target, although that claim uses a different measure from the three-category expenditure figures reported to Parliament.
An important point that needs to be emphasized is that the £600 million figure represents savings against what departments had originally planned to spend, not a reduction in actual expenditure compared with the previous financial year.
As the Financial Times reported on 23 October, year-on-year consultancy spending fell by £129 million, while professional services expenditure increased by £581 million.
These figures concern spending across the reporting departments and include data limitations. They are neither a Big Four total nor directly comparable with the newer contract-award figures. Nevertheless, the reduction deserves acknowledgement.
The distinction between spending and dependence remains. A department might reduce its consultancy bill yet still need external suppliers for its systems, implementation and continuing operations.
The government has also begun addressing dependence more directly.
In September 2026, the Cabinet Office announced a new Insourcing Unit to coordinate departmental efforts to bring services back into government, remove barriers and identify opportunities for internal delivery.
Under procurement guidance published in June, a Public Interest Test will apply from 1 April 2027 to most new central government service projects and re-procurements valued above £1 million. Organisations with annual contract spending of at least £100 million will also be required to develop five-year insourcing strategies, with publication requirements taking effect in April 2027.
The test is intended to ensure that departments consider in-house delivery before proceeding with procurement. It does not automatically favour insourcing, nor does it replace a detailed assessment of delivery options. Departments will still need to demonstrate that internal provision is feasible and represents value for money.
These measures acknowledge that reducing expenditure and rebuilding internal capability are related but distinct objectives. Whether they succeed will depend on implementation and whether departments develop the skills and resources needed to undertake work previously performed externally.
The government’s supplier spending annex provides an illustration. For 2024/25, the Home Office reported approximately £115.15 million in professional services expenditure with Deloitte, £13.54 million in consultancy and £0.73 million in contingent labour. The reported consultancy expenditure was therefore substantially smaller than the professional services expenditure recorded for the same supplier.
The example shows that the relationship extends well beyond the category targeted by the consultancy pledge. A reduction in one category does not tell us what has happened to the relationship as a whole.
Knowing what government buys
The National Audit Office’s November 2025 report on external consultants explains why measuring expenditure is difficult. Departments can purchase packages combining advice, professional services and temporary personnel. Different definitions and inconsistent reporting make it difficult to isolate consultancy costs and compare them over time.
In March 2026, the Public Accounts Committee questioned whether the government could demonstrate that halving consultancy spending would effectively reduce expenditure on external providers. It warned that inconsistent classification could allow expenditure to be reclassified rather than reduced.
The concern has since acquired greater significance. The Financial Times investigation of 23 October identified actual accounting reclassifications, as well as substantial differences between movements in consultancy and professional services expenditure.
The Department for Business and Trade acknowledged transferring £14.9 million of costs from consultancy to professional services through an accounting adjustment. According to the FT, the transfer was equivalent to roughly three-quarters of the department’s reported £20 million reduction in consultancy expenditure.
Elsewhere, the Home Office reduced consultancy expenditure by approximately £124 million while increasing professional services expenditure by £86 million. The Ministry of Defence recorded a £4.4 million reduction in consultancy spending alongside a £236 million increase in professional services.
These movements do not, by themselves, establish that departments deliberately disguised consultancy expenditure. Professional services encompass activities that may genuinely fall outside the consultancy category, and a transfer between classifications does not necessarily involve any change in the work being delivered.
Nevertheless, the figures raise questions about how spending reductions are measured. The Institute for Government had previously identified departments retrospectively reclassifying expenditure and described some of the revisions as unusual.
The Cabinet Office rejected allegations that the figures had been manipulated. It maintained that consultancy expenditure had fallen by more than £600 million against departments’ original plans and said that government bodies were improving data collection and correcting historical reporting errors. It also pointed to a £1.4 billion reduction in contingent labour expenditure as evidence of lower overall spending.
The government has now commissioned its internal audit agency to examine whether consultancy spending is being correctly classified across 15 public bodies. The findings of that review will be important in determining whether existing classifications provide a reliable picture of expenditure.
In the committee’s accompanying press release, deputy chair Clive Betts said that “government does not yet have a grip on precisely what is being spent”.
The committee also highlighted a separate weakness: although government collects information on consultancy spending, it has a much less complete understanding of how consultants are being used. Its March report noted that the Consultancy Playbook had not been updated since 2022 and still referred to a Government Consulting Hub that closed in 2023.
A government that cannot consistently identify what it buys, why it buys it and what capabilities it acquires may struggle to determine which activities should remain outsourced.
The July spending response and supplier annex represent subsequent disclosure and should be read alongside the committee’s criticisms. They provide more information, but neither amounts to a measure of government’s ability to perform recurring work, assess suppliers or change arrangements.
Buying expertise and retaining it
The question of capability is central to Mariana Mazzucato and Rosie Collington’s The Big Con. They argue that repeatedly outsourcing core work can undermine the learning through which organisations develop expertise. Buying a completed output is not equivalent to developing the knowledge needed to understand and adapt it.
Chris Hurl and Leah B. Werner’s The Consulting Trap extends the argument to firms’ overlapping roles in advice, administration, technology and service delivery. Their concern is that relationships can become embedded in the systems through which government operates, making them difficult to unwind.
These are explanations to test against evidence, rather than conclusions that follow automatically from a large contract total. An agency may buy specialist assistance because it lacks expertise, because it needs temporary capacity, or because external support offers genuine value. Repeated contracts can reflect good performance as well as limited alternatives.
The NAO itself found that 86% of the 48 officials answering its question about value regarded consultants as valuable contributors. This was a small survey of officials, rather than an independent assessment of contract outcomes. It nevertheless underlines why an argument against all external advice would be unconvincing.
Academic research also complicates a simple account of consultants replacing civil servants. A 2023 study of 125 English public hospitals over six years found no evidence that higher consultancy expenditure subsequently reduced management employment. Larger management functions were associated with more consultancy spending. A greater presence of specialist internal consulting managers was associated with less external consultancy use.
That study concerns hospital management functions, rather than every form of public sector capability. Its findings still suggest that the type of expertise retained internally matters, and that public managers are active purchasers with their own objectives.
Adding staff alone is therefore an incomplete response. Departments need to identify which skills they repeatedly purchase, which should remain available internally and which are sensibly obtained from outside.
There are also economic limits to insourcing. Some specialist capabilities may be needed only intermittently, making permanent internal teams difficult to justify. Others require technical expertise that government may struggle to recruit and retain. Bringing work in-house can involve substantial upfront investment in staffing, systems and training.
The objective need not be to eliminate external expertise, but to ensure that departments can distinguish between services that are economical to buy and capabilities they cannot afford to lose.
When the contract ends
The NAO’s good-practice guide on using consultants supplies a practical test. It says: “Civil servants need to learn enough to keep an initiative going after consultants’ contracts end.”
That requires more than receiving a presentation or a folder of handover documents. Officials need to understand the analysis, assumptions and tools well enough to use the work, challenge it and adapt it. Knowledge transfer needs time, staff participation and clear expectations throughout an engagement.
A commercial supplier benefits from repeat business and additional services. A public body benefits from useful delivery while retaining workable alternatives. Those interests can align: a firm may win further work because it performs well and strengthens its client’s capabilities. They can also diverge if an engagement leaves the customer unable to manage the next stage without further assistance.
The appropriate question is whether contracts, staffing and evaluation arrangements make the development of government capability an explicit objective where that capability is needed repeatedly.
The ability to change course
The government’s efforts to bring civil service training in-house provide a particularly revealing example of the practical difficulty of changing established supplier relationships.
In August 2026, the Cabinet Office awarded KPMG and EY contracts with a combined maximum value of £456 million to provide training across the civil service. KPMG’s contract was valued at up to £319.2 million and EY’s at up to £136.8 million, both including VAT.
The agreements run from 31 August 2026 until 31 March 2028 and cover training in areas including digital skills, artificial intelligence, leadership and management. Their stated values are contract ceilings rather than guaranteed expenditure.
Both were directly awarded to incumbent suppliers. The Cabinet Office explained that the arrangements would support an accelerated transition to the National School of Government and Public Services, which is intended to bring more civil service training in-house. Changing suppliers during that transition, it argued, would have been disruptive.
The department also emphasised that the contracts covered training rather than consultancy services. They therefore illustrate how substantial Big Four contracts can sit outside the expenditure category targeted by the government’s consultancy reductions.
There are two possible interpretations. The awards may represent a necessary transitional investment, enabling government to build internal training capacity without interrupting existing provision. They may, however, also be an example of how difficult it can be to unwind relationships with suppliers when continuity depends on their continued involvement.
Reducing reliance on external providers may involve more than deciding not to renew a contract. Government must first be capable of delivering or transferring the underlying service.
Digital services make the consequences particularly tangible. Replacing a supplier can involve technical integration, data transfer, retraining and disruption to essential operations. The ability to choose a different supplier depends partly on decisions made when the original arrangement was designed.
The NAO’s January 2025 report on technology suppliers found weaknesses in government’s digital procurement skills and its use of existing technical expertise. It also identified insufficient attention to managing suppliers after contracts had been awarded. The findings concern technology suppliers broadly, not allegations against the Big Four specifically.
In the official press release, NAO head Gareth Davies said:
“A lack of digital and procurement capability within government has led to wasted expenditure and lack of progress on major digital transformation programmes.”
Effective oversight requires people who can interrogate proposals, recognise unrealistic promises and understand what switching would entail. Even where government continues to outsource delivery, it needs sufficient knowledge to govern the relationship.
That makes workforce planning central to the spending pledge. Yet on 9 October, ministers confirmed that publication of the Civil Service Strategic Workforce Plan had been delayed, citing alignment with wider transformation proposals and ambitions for an AI-enabled state. The answer supplied no publication date. It does not establish that internal planning has stopped, but the public explanation of how the workforce will change remains incomplete.
The government’s new insourcing framework offers a route towards more systematic decisions, but it remains to be seen whether departments will have the funding, skills and organisational capacity to act on those decisions. A requirement to consider internal delivery cannot, by itself, create the workforce capable of providing it.
Britain can buy useful expertise from the Big Four. The harder task is to ensure that those purchases leave government able to make informed choices about what comes next.
A smaller bill would be welcome. Evidence of reduced dependence would also show that officials can challenge the analysis, retain and reuse useful knowledge, manage delivery and change supplier without unacceptable disruption. Where recurring work belongs inside government, it would show that departments have rebuilt the ability to do it.
That requires more than demonstrating that consultancy expenditure has fallen. It requires understanding the full range of services purchased from external firms, identifying which capabilities government needs to retain and ensuring that outsourcing decisions remain reversible where practicable.
The test is whether continued reliance on a firm is a choice government can justify—and a choice it can change.
Follow the money. Examine the influence.
Big4News investigates the Big Four’s role in business, government and public life. Explore our global scandal timeline and regulatory case database for the evidence behind the headlines.
Subscribe to receive independent reporting and analysis on Deloitte, PwC, EY and KPMG.
About Claudine Cassar
I’m a corporate anthropologist and former Deloitte equity partner. I sold my technology business to Deloitte in 2016 and led the Malta Consulting team for five years. I am the founder and editor of Big4News, which provides independent, clear analysis of PwC, Deloitte, EY, and KPMG — free from corporate spin.
Find me on LinkedIn, X, Instagram, or my author website.
Feel free to reply to this newsletter — I read every reply.
The pattern of spending going up despite promises and dependency are global. The fact is the civil service is in part to blame. The idea that it would be disruptive to have new suppliers while the government builds its own in house leadership and management capability is risible. As if they cannot manage a transition like that or just find people and bring them in to run it in house
We find the public sector is the same in Australia. The growth in public sector numbers continues and so does consultancies. Why? And why is that in all the scandals surrounding KPMG in Australia they were awarded constracts to run ethics programs.
Weak, weak, weak. Time the public sector is held to account for these choices. No wonder the consultancies grow.