Seventeenth Report - Cabinet Office functional savings

Different functions are still at different levels of development, and different levels of maturity in their approaches to calculating and reporting savings. The 14 functions are different shapes and sizes and have different approaches and methodologies to calculating savings. It has taken nearly a decade for the government to begin quantifying and reporting the efficiency savings that functions deliver. Those that do report are at different stages of development in terms of their maturity as organisations and how they report their savings. While most have a clear methodology in place, some still do not, meaning the savings reported are inconsistent. For example, the Commercial function has a detailed methodology for claiming savings, while the Property function did not include the appropriate level of detail and documentation for its savings to be included in the most recent Cabinet Office exercise. Recommendation 2a: Cabinet Office should, within six months, take stock of the different functions to understand their respective progress, and outline how it will support those who need remedial action to improve. b) By the end of 2024, the Cabinet Office should publish a balanced scorecard for each function which includes savings of all types made by the functions.

Cabinet Office and HM Treasury do not have a full picture of the performance of functions. Cabinet Office’s current exercise to quantify functional savings has reported for the last two years, but it does not include all functions. In addition, 6 Cabinet Office functional savings functions deliver different types of savings and each sets its own methodology for tracking and claiming savings. This can make it hard to compare progress between the functions and means that there is a risk that efficiencies are being under reported. One of the barriers to doing this more consistently is access to the necessary data. Government has already committed to creating more successful data exchange, so that locally created data can be shared more effectively across government. Standardised metrics, including establishing accurate baselines, are needed to address gaps in the existing data, fully capture the savings made, and allow comparisons to be made across functions. Recommendation 3: As part of the Treasury Minute response, the Cabinet Office should set out how it will work with functions to develop consistent methodologies that report the totality of the costs, benefits and savings delivered by functions, using metrics that can be compared across time and different areas of government.

Cabinet Office and HM Treasury have not ensured that functions have fully reported the efficiencies they achieve. Overall, government expects departments to make efficiency savings equivalent to 5% of their day-to-day budgets by 2024– 25, and each department has its own efficiency target. Currently, functions do not report all the savings they produce, and not all functions report their savings. There are no targets reported in the Government Efficiency Savings report Cabinet Office publishes, although they are set, for example, for the counter fraud function. Without clear targets for most functions there is little incentive for them to report more savings, particularly non-cash releasing savings, given the current focus within departments on the need to reducing their overall budgets. In 2023, GIAA noted that the amount of support the Cabinet Office could provide to the functions created a risk that efficiencies might be underreported. The Cabinet Office does not share examples of good practice between functions, although this would be welcomed by the functions themselves. By sharing successful approaches to calculating savings across a function, such as the case within the finance function, and between functions, other functions can be incentivised to achieve greater efficiency savings. Recommendation 4a: The Cabinet Office and HM Treasury should set out what the targets are that functions are working towards in the next functions’ savings exercise. b) Once the new targets have been set, the Cabinet Office and HM Treasury should work with all the functions to more robustly test the amount and scale of savings they report, and challenge functions to report wider savings where necessary.

Cabinet Office and HM Treasury have not finalised how they will manage the risks of double-counting and cost-shunting. Avoiding adverse effects is a key element of good practice in making and reporting efficiency savings. HM Treasury and the Cabinet Office expect there will be natural convergence between the separate efficiency reporting exercises for departments and functions. HM Treasury expects that ultimately functional efficiency information will become part of the wider government efficiency data set. But it recognises that its new Efficiency Framework could increase the risk that savings are counted by both central teams and departmental teams, as departments will be more rigorous in reporting efficiencies. It is essential that savings are not double counted, costs are netted, and that costs are Cabinet Office functional savings 7 not shunted from one area of government to another. Reporting functional savings and wider efficiency savings in parallel increases the risk of duplication in terms of savings claimed and bureaucratic effort. Recommendation 5a: HM Treasury and the Cabinet Office should incorporate into the 2023–24 efficiencies exercise: • Examples of best practice on calculating, recording and reporting of savings with departments and functions; • Clearer guidance on how to avoid cost-shunting and double-counting, as well as any other adverse effects; b) By the end of 2024, the Cabinet Office and HM Treasury should report back to the Committee on what assurance they have received from departments that the savings claimed have not led to costs elsewhere in government. 8 Cabinet Office functional savings 1 Achieving efficiency savings

On the basis of a report by the Comptroller and Auditor General, we took evidence from the Cabinet Office and HM Treasury (the Treasury) about functional savings across government.1 The importance of efficiency savings

Government officials need reliable data on where efficiency savings are identified to free up resources for other priorities and to make decisions about where to best allocate resources to maximise return on investment. Managing Public Money emphasises that effective decision-making is reliant on regular, high-quality information about costs, performance and efficiency.7 We have found throughout our work that timely and comprehensive data is essential to support informed decision making. For example, we have repeatedly raised concerns that delays to the Whole of Government Accounts (WGA) and missing data were reducing the usefulness and reliability for Parliament, local authorities and the public. In 2024, we found that delays to publication of the WGA meant that Parliamentarians and other users were having to make important decisions without full information about public finances.8 We therefore asked when government could expect the reporting of efficiency savings to be sufficiently improved so that the information could be used to inform policy decisions. The Cabinet Office told us that the reporting of efficiency savings in 2024 would be “so much better” as a result of the Government Efficiency Framework.9

We also asked whether departments were expected to include their efficiency savings as part of their Annual Report and Accounts. The Treasury told us that while some chose to, they were not required to as standard, but that this was something it was currently looking at. For example, it explained that the Ministry of Defence provided details of its efficiency savings that were audited and reviewed annually, but that the approach across departments was inconsistent. It noted that as part of developing its new Efficiency Framework, it hoped to be able to provide publicly audited information on the efficiency savings made by departments, but that this would take “several years to be done properly, thoroughly, and well”. The Cabinet Office was not able to say when it thought that it would have a “forward Spending Review long look over what efficiency frameworks to this level of audited standards are going to come in” but that it would be a multi-year commitment.10 Calculating the savings achieved by functions

Since the first functions were established in 2013, there are now 14 government functions, covering activities such as procurement, major project delivery and finance. We asked the Cabinet Office how successful it thought functions had been in improving efficiency. It told us that it thought that the functions had been very successful and provided a high level of specialist expertise to departments, as well as external challenge.11 The NAO found that the functions were different shapes and sizes, which meant they had different approaches and methodologies to calculating savings. The Cabinet Office recognised that the functions were at different stages of development. It explained that some functions, such as the Commercial Function, are “very well-established … [and have] been going for 6 Qq 43, 68 7 C&AG’s Report para 1.11 8 Committee of Public Accounts, Whole of Government Accounts 2020–21, Ninth Report of Session 2023–24, HC 65, 26 January 2024 9 Q 79 10 Qq 77–81 11 Qq 2, 4, 27; C&AG’s Report para 3, 1.2, Figure 1 10 Cabinet Office functional savings a long time”, whereas others, such as Security, Property or Communications, were much younger and “less mature, but obviously very good in their way”.12 It also recognised that the performance of the Digital Function, which had been in place for over 15 years, had not been as strong as it should have been. It noted that the function had “lost some of its excellence” and was “a recovering function”. In comparison, it noted that the HR function was performing well, but that more was expected of it than used to be the case.13 The NAO report found that the Treasury expects that the relationship with functions will be “even more mature by the 2024 Spending Review”. We therefore asked what this meant in practice. The Treasury told us that it worked in partnership with the Cabinet Office, and that it was their joint endeavour to make sure that the functions continued to mature and that they raised the standard overall.14

Each function sets its own methodology for tracking and claiming savings from its work, to reflect the different types of work they undertake. Functions are responsible for verifying the supporting data and confirming appropriate levels of governance to ensure confidence in the efficiencies before reporting. The GIAA examines the process functions use to calculate and assure the savings claimed by functions. In 2022, as part of its assessment of the 2020–21 savings, it found that most functions had clear methodologies in place to support the savings they report, and had gathered evidence to support them, but not all. In addition, it found that standards varied widely between functions, and that even within some functional teams, methods and evidence for savings were inconsistent. As part of its review of the 2021–22 savings, the GIAA again reported that the majority of functional teams had clear methodologies in place and evidence packs to support the headline savings claimed, but not all. Overall, the GIAA gave the reported savings for 2021–22 a ‘moderate’ assurance rating and found similar weaknesses to the previous year in the approaches taken by functions. In 2023, the GIAA similarly found that the Commercial Function had a detailed methodology for claiming savings. In comparison, the Property Function was new to the process of benefits tracking and its approach did not include the appropriate level of detail and documentation required, and so did not progress with the audit of the savings and they were not included in the most recent Cabinet Office exercise.15

The Cabinet Office told us that while it thought that the reported savings figures were “pretty robust”, it recognised that the figures could be more comprehensive and more consistent in the measurement of savings. The Cabinet Office recognised the need for consistent methodologies for calculating the savings achieved by functions, but asserted that these did not need to be identical because “the ways in which we actually achieve the savings across the different functions are genuinely different”. When asked which functions were making the most efficiencies, the Cabinet Office told us that the functions provided a lot of value, but that while some of that could be best described by savings, this was not the case for all functions, such as Security.16 12 Q 4; C&AG’s Report para 10, 12, 1.9–1.10, 2.3 13 Q 4 14 Q 68 ; C&AG’s Report para 3.14 15 C&AG’s Report paras 12–15, 2.3, 2.7–2.12 16 Qq 2, 6, 41 Cabinet Office functional savings 11 2 Completeness of the functional savings reported Capturing the performance of the functions

The Cabinet Office has measured and reported on the financial efficiency savings and wider benefits made by cross-government functions since 2021. In March 2022, it reported that functions, departments and other central government bodies had achieved £3.4 billion of cash-releasing savings in 2020–21. Of this, £1.4 billion was attributable to the Counter fraud Function and £1.2 billion to the Commercial Function. In July 2023, it reported that the cross-cutting central functional teams had delivered £4.4 billion of savings in 2021–22, consisting of £3.4 billion cash-releasing savings, and £1 billion non- cash releasing savings. Of this, the Counter Fraud and the Debt Management functions has the largest savings, at £1.3 billion each.17

We noted that, of the reported £3.4 billion savings, the overwhelming majority was attributed to a small number of functions. We therefore asked what work the Cabinet Office has undertaken to identify whether more savings could be achieved from elsewhere. The Cabinet Office explained that a lot of work was being done by functional experts and departments was regarded as business as usual, so was not counted as part of the efficiency savings achieved. It recognised that savings were heavily dependent on the Commercial, Counter-fraud and Debt Management Functions, and “there are more savings than that out there”. It told us that it thought there was a lot of potential to make savings in other functions and that it needed to calculate these savings, ensure they met the audit standards, and publicly report them.18

The Cabinet Office is ultimately responsible for the accuracy of the figure for the total amount of savings reports. It co-ordinates, assures and reports the efficiency savings achieved by functions, and provides guidance on what they should report. But the NAO found that the Cabinet Office had not consistently reported the efficiency savings delivered by the functions. The Cabinet Office reports savings from those functions headquartered in the Cabinet Office for the period it is reporting on – which was 10 functions for the 2021–22 report. In its March 2022 report on the savings achieved in 2020–21, the Cabinet Office recognised that the £3.4 billion figure was an understatement of the total savings and benefits delivered by the functions. The report also included ‘unaudited wider benefits’ from the Legal, Property, Analysis, Security, and Finance functions, as well as Government Business Services and the Crown Commercial Service. The GIAA concluded that the processes in place for 2020–21, did not sufficiently mitigate the risk of claims being under- or overstated.19

We noted that the NAO report stated that it was important that the Cabinet Office had a tighter definition of what a saving is, and that this needed to be consistently applied and have the same methodology and baseline. In its July 2023 report on the savings achieved, the Cabinet Office noted that the savings reported related to central government functional teams – the parts of functions which sit in the centre of government, rather than within 17 Qq 2, 4, 8, 28; C&AG’s Report paras 4, 6–7, 1.6–1.7, 1.9–1.10, 2.2 18 Qq 2, 39 19 C&AG’s Report, paras 10, 1.8–1.10, 2.4, 2.9 12 Cabinet Office functional savings departments. But in its technical notes to the report, it stated that around £1 billion of the Counter-Fraud savings were delivered by department-led activities supported by the central function, leading to inconsistency in what savings are included in the £4.4 billion figure.20

The NAO concluded that functional savings report may therefore give an incomplete picture of savings generated across the functions. The Cabinet Office provides challenge on an ad-hoc basis to functions where it believes that they have secured savings but not submitted them as part of the reporting process. In 2023, the GIAA noted that this approach meant that there was a risk that efficiencies might be underreported due to the amount of support the Cabinet Office could provide to the functions during this exercise.21

We noted that a crucial area to improving reporting is ensuring that functions can get the data they need to identify savings, and that to do this effectively data systems needed to be updated or redesigned so that data sets were comparable. The Cabinet Office explained that two types of data were key; operational data, which is used to support the delivery of public services and analytical data which informs future policy. It told us that the crucial thing was ensuring that different local data systems were able to speak to each other so that long-term patterns can be found. The Cabinet Office told us the Central Digital and Data Office had been championing better data exchange so that “when you want to access other people’s data or have data yourself, that is now entered into a kind of registry” and the terms for sharing data were according to pre-agreed, standardised Memorandums of Understanding and templates.22 Ensuring functions report their savings

As part of the government’s ambition to make savings equivalent to 5% of overall spend across government by 2024–25, every government department has both an efficiency target and an efficiency assumption “baked-in” to the funding allocated to it as part of the Spending Review. The Treasury explained that different departments had slightly different targets, depending on what it considered was deliverable. In its March 2022 and July 2023 reports on Government Efficiency Savings, however, the Cabinet Office did not state any targets for the savings achieved by functions.23 However, as part of our inquiry into Tackling fraud and corruption against government, we found that Cabinet Office and HM Treasury expected every major department should have an outcome target for their counter-fraud investments, which should provide a return of at least £3 for every £1 spent. During our evidence session in May 2023, HM Treasury told us that the move towards targets showed that investments should be underpinned by making a measurable difference to the risk or problem being addressed.24

We asked the Cabinet Office how it was ensuring that there was sufficient direction and leadership in departments and functions’ approaches to achieving savings. The Cabinet Office told us that it was getting “terrific value” in areas such as land disposals, where it had set a clear target for each department and public body to help achieve savings 20 Q 41; C&AG’s Report, para 1.9–1.10 21 C&AG’s Report paras 11, 2.6 22 Qq 59–62 23 Qq 33, 36; C&AG’s Report, para 2, 3.2; Cabinet Office, Government Efficiency Savings 2021, 28 March 2022; Cabinet Office, Government Efficiency Savings 2021/22, 19 July 2023 24 Committee of Public Accounts, Tackling fraud and corruption against government, Sixty-Ninth Report of Session 2022–23, HC 1230, 8 September 2023 Cabinet Office functional savings 13 of £500 million and had achieved savings of £1.1 billion. The Cabinet Office explained that efficiency savings and targets had become lower priority in recent years because of issues such as EU exit, response to Covid and support for Ukraine. It told us that it believed that the efficiency reporting exercises and allowing departments to keep the money they save through their initiatives had helped drive good behaviours and incentivise departments to look for savings, including making savings that can be quite difficult.25

We asked the Treasury whether, as the lead for the Finance Function, it had shared the methods it had put in place to develop savings with other functions. The Treasury told us that the heads of functions met regularly as part of the Government Finance Function Efficiency and Performance Group, which focused on sharing best practice within functions, between functions, and across departments. However, the NAO, concluded that the Cabinet Office did not share examples of good practice between functions, although functions would welcome this. The GIAA recommended that the Cabinet Office produced “principles-based” guidance for functions about how savings should be calculated and presented.26

We asked the Cabinet Office how accurately it could predict what savings would be achieved in future. It told us that this had changed over time and depended in part on the extent to which the methodologies to calculate savings had been developed by the functions. The Cabinet Office explained that the methodology used to calculate efficiency savings was working well in relation to areas such as counter-fraud, debt management and commercial transactions, but in others they were still being developed. It told us that it expected to be able to develop a methodology that met the required audit standards for areas such as shared services and property management, and some others, which is expected would change the composition of the savings as they developed. It told us that, having reported overall savings for the last two years, it was confident that it would continue to be able to report savings in “that kind of ballpark” in future.27 25 Qq 18, 29, 31, 33, 36 26 Qq 13–16; C&AG’s Report paras 18, 2.11 27 Qq 7–8 14 Cabinet Office functional savings 3 Future reporting of efficiency savings The Government Efficiency Framework

In July 2023, the Treasury issued a framework for tracking, monitoring and overseeing efficiency savings. The framework provides definitions, guidance and best practice examples on how departments should calculate and report efficiency savings to the Treasury. The Treasury expects departments to adopt the Framework for reporting efficiency savings for the financial year 2023–24 and arm’s-length bodies to do so in 2024–25. This, however, will be different to the process for identifying and reporting efficiency savings achieved by the functions.28 We asked the Treasury what progress it had made in implementing the new framework. The Treasury told us that it had instructed departments that they must comply with the framework, but the real test would be when it started to ask departments for their reports using the new Framework in 2024.29 The Cabinet Office expected that the introduction of the Framework would make “the most enormous difference to the value of savings that we capture, and will enable reporting to a really good standard”.30 The Treasury told us that one of the main reasons for producing the Efficiency Framework was because it didn’t have “consistent, comparable, bottom-up information for every single department”. It explained that the Framework would allow it to stress-test and benchmark the savings being achieved between departments and between functions.31

We asked what scope there was to accelerate the identification of savings. The Cabinet Office told us that the next year would be one of transition given the introduction of the Efficiency Framework and expanding this from the Treasury and Cabinet Office teams to all government departments. The Cabinet Office told us that it wanted to increase savings, but that it was simultaneously aware that savings were also indications that something could have been done at a lower cost in the first place, so it did “not necessarily regard more and more of them every year as a goal”. But it committed to ensuring, with the NAO’s advice, that it was comprehensively capturing, in a consistent way, the savings being achieved.32 The impact of efficiencies elsewhere in government

Avoiding adverse effects is a key element of good practice in efficiency reporting. Efficiencies should not adversely affect resilience planning or the experience of service users, and should not add costs to other part of government. This also includes being net of costs, where all transactional and ongoing costs should be netted off from savings and any adverse effects on other programmes should be recognised.33 As part of our examination of efficiency in government in 2021, we found that attempts to improve efficiency can inadvertently reduce the quality of services of increase costs elsewhere, known as cost-shunting. In particular, we noted that efficiency plans without a clear idea of the implications for service users had led to problems. We therefore recommended that the Treasury should ensure that, in their efficiency plans, departments considered 28 Qq 34–35, 42–47, 66; C&AG’s Report para 3.11 29 Qq 34, 47, 81 30 Q 9 31 Q 12 32 Qq 7–9 33 Qq 40, 75; C&AG’s Report Figure 7; and Comptroller and Auditor General, Efficiency in government, Session 2021–22, HC 303, National Audit Office, July 2021 Cabinet Office functional savings 15 what the potential impact might be on service users, and that data on this was tracked as programmes were implemented.34 In its response to our report, government accepted our recommendation and explained that as part of the 2021 Spending Review it had published an updated set of Priority Outcomes and Metrics to capture the real-world impacts of savings. It also noted that departments were required to report on their progress against these metrics, which enabled Ministers and officials to identify where outcomes were at risk of note being achieved and take action.35

The NAO found that while the Government Internal Audit Agency (GIAA) looked at the processes the functions used to calculate and assure the savings claimed, it did not consider whether the efficiency savings created adverse effects on other parts of government. The scope of the GIAA’s audit did not include assurance over whether savings affect resilience planning, service user experience, or add costs to other parts of government. In addition, the NAO found that the Cabinet Office did not ask functions to assess whether their savings might add costs in other areas, because it considered that this was not always possible or proportionate.36

We therefore asked the Cabinet Office how confident it was that efficiencies were genuinely cash savings, and had not resulted in additional costs elsewhere in government. The Cabinet Office told us it asked the heads of function to answer that question, but did not seek independent verification. It explained that it was “wary” about cost-shunting type activity, but that it thought that this was “probably not an area that is particularly susceptible to so-called cost-shunting”.37 We therefore asked what controls it could introduce to ensure that cost-shunting was not taking place. The Cabinet Office recognised that it did not currently have “that extra loop” to check with departments whether savings had resulted in a tertiary impact. It told us that it would be taking up the NAO’s recommendation to be more prescriptive in its guidance to ensure that the cost savings being put forward by departments were “net of cost shunting in other places”. It did not, however, propose to change the methodology of its central team or the GIAA, or to have an independent verification of whether savings had resulted in service implications.38

The Treasury confirmed that it had incorporated the NAO’s principles for reporting efficiency savings into its Efficiency Framework, which included setting out the requirements for sustainability and avoiding cost-shunting.39 The new Efficiency Framework states that for reporting both cash-releasing savings and non-cash releasing savings there should not be an adverse impact on performance or outcomes. For cash- releasing savings it explains that savings must not adversely impact on the achievement of a department’s strategic priorities, and that departments should be able to demonstrate that because of reforms, the department and sector is delivering better value for money overall. It also states that, to report cash releasing savings, they should be “sustainable and should not be reallocating or deferring costs to future years”. The Framework notes that 34 Committee of Public Accounts, Efficiency in government, Twenty-Eighth Report of Session 2021–22, HC 636, 3 December 2021 35 HM Treasury, Treasury Minutes: Government response to the Committee of Public Accounts on the Twenty- Seventh to the Thirty-First reports from session 2021–22, CP 631, February 2021 36 Q 44; C&AG’s Report, paras 13, 2.7 37 Q 44 38 Q 45 39 Q 76 16 Cabinet Office functional savings cost reallocation or deferral takes place where there is a simple movement in cash across a year end which does not relate to a total net reduction in waste or inefficiency when the two years are taken together. It does not, however, specifically reference cost-shunting.40 Avoiding double-counting efficiency savings

As part of the key principles for claiming efficiency savings, efficiencies should be carefully calculated, evidence and reported. In doing so, savings should not be double- counted and should not be reported again in future initiatives.41 We asked what witnesses were doing to ensure that savings weren’t being double counted, and whether having one process for identifying and reporting functional savings, and a different process for identifying and reporting departmental savings, created a danger that there would be two bureaucracies both working towards the same aim. The Cabinet Office told us that double- counting was checked as part of the GIAA’s review, which looked at the methodologies used to ensure that savings were not being counted in other places. It recognised, however, that with the implementation of the new Efficiency Framework, there was “perhaps more risk that a commercial saving is counted by the Cabinet Office central team and in the departmental team, because the departmental team is doing more rigorous reporting of efficiency”.42 It explained that it had highlighted this as a risk, and that it would manage this as part of implementing the new framework, but that it thought that it was a “small price to pay for the overall increase in focus and reporting on efficiency”. The Treasury similarly explained that there will be a potential for overlap and inconsistency, and that this would be part of the work it would need to undertake with its auditors to make sure that it was “doing everything we can to clean the data and get consistency over time”.43

The Cabinet Office also noted that the introduction of the new Framework will mean that there will need to be an agreement between central government and departments about the proportions of savings that are collated to each, which is thought would become “a bit tighter”. It said that it was still keen to track the overall picture of savings across functions, because this was important to benchmarking activity – whereas if thought that if it only tracked savings by department, it would lose “comparability across the piece”. The Treasury hoped that over time it would be able to house the functional efficiency information in the wider government efficiency data available. It explained that the Efficiency Framework would initially start by focusing on cashable savings, as this was “the bit we are most interested in, can deduct off budgets and is the main incentive”, and non-cashable savings would be provided by the functions. But in the “very long-term” it expected that “absolutely everything” to be captured by the Efficiency Framework, and could lead to it not having to do two processes. The Cabinet Office similarly told us that it anticipated that “very quickly, there will only be one efficiency framework” but that on the basis of the NAO’s report, it recognised that there may be methodology reasons where it would want to make additional reporting, for example around fraud. The new Efficiency Framework states that savings should be scored once, net of any double-counting between 40 Cabinet Office, Guidance: The Government Efficiency Framework, 19 July 2023, sections 3.4 and 3.5 41 C&AG’s Report Figure 7; and Comptroller and Auditor General, Efficiency in government, Session 2021–22, HC 303, National Audit Office, July 2021 42 Qq 46, 66 43 Qq 46–47 Cabinet Office functional savings 17 different departments, and that double-counting must be avoided through “clear reporting of data”. But beyond outlining reporting requirements, it does not specify what this data should