The following is a transcript of a keynote speech given in Parliament by Gareth Davies, the head of NAO, to MPs and civil servants on 4 February 2025. You can also watch a recording of the speech.
Introduction
Good afternoon everyone, and thank you Sir Geoffrey for that kind introduction, and thank you also for the productive working relationship between your committee and the National Audit Office, which you have ensured was quickly put in place in the new Parliament.
Every year I give a speech in Parliament on the big themes from the NAO’s independent perspective, drawing directly on the work that Sir Geoffrey was describing we’ve done in support of the Public Accounts Committee and other Parliamentary committees.
Last year I identified the scope to liberate billions from better run major projects, from properly managed assets, better procurement, digital transformation and reduced fraud. I said that a concerted cross-government effort was needed to do that, focused on robust data; innovation and evaluation; an improved approach to planning and spending; and the leadership, skills and culture to succeed.
So I was pleased to see both the previous government and the new one respond to the efficiency agenda I outlined last year, and we will be tracking progress in all of these very significant areas through our reports for Parliament, because they remain critical in cutting out waste and achieving better results for the available funding.
I welcome the creation of the new Office for Value for Money, with its brief to drive systematic improvement in value for money from inside government.
One of the opportunities it offers is for government to act more quickly and effectively on the evidence from NAO and PAC reports and other sources. And to that end, I am meeting regularly with the Office for Value for Money’s chair, David Goldstone.
The NAO is independent of government and has two main responsibilities: ensuring that government accounts properly for its income, expenditure, assets and liabilities; and assessing the value for money of public spending. We don’t comment on the merits of policy objectives but, alongside our vital role in supporting parliamentary and public accountability, our evidence-based reports are available to assist those making and implementing policy.
This year as Sir Geoffrey just said, we have a new Parliament and a new Government, but many of the same problems of rising demand and not enough money to quickly fix the gaps in key public services. We also face other challenges that risk causing widespread disruption, from global instability and climate change to public health emergencies and cyber threats.
And it’s clear how difficult it’s proving to address these issues. Yet finding effective solutions is vital and the upcoming spending review is the opportunity to do so. The NAO’s work can support the government in addressing these challenges, and I’ll draw on a range of that work as I speak.
This year I want to focus on two of the deeper issues – productivity and resilience – that must be confronted if we are to get public spending on a more sustainable footing, delivering for people at a price society can afford. And crucially, how government can use innovation to make faster gains.
Productivity
So my first theme is productivity – the benefits we get from every unit of input. Since the 2008 financial crisis, we’ve seen lower productivity growth across the whole economy than in the decades before.
That’s well known – and Covid 19, of course, didn’t improve the picture, particularly in the public sector.
Take the example of acute hospital productivity. Figures from NHS England in May last year showed it was still 8 per cent lower in productivity in 2023/24 than before the pandemic and much work is underway to address this.
While more recent data from December shows encouraging signs of improvement, we know that the Health system is currently battling through a difficult winter. Demand on services continues to grow, and not just in the NHS. And therefore, focusing on productivity across the entirety of the public sector is essential.
As our work shows, improving productivity in public services requires action on many fronts. I’ll pick out just four of the most important from the evidence from our work.
First, applying new technology including Artificial Intelligence. This is rightly at the top of the agenda at the moment with clear potential for reducing the time taken for routine tasks, augmenting the work of skilled experts and making public services easier to use.
Our 2024 report on government’s use of AI showed that almost three-quarters of departments and national bodies were either already deploying AI tools or were piloting them.
The question is not whether AI will make a difference to productivity in the public sector but how to maximise the benefits whilst managing the risks. I’ll return to this shortly when I talk about innovation.
But it would be a mistake to rely only on new technology to drive higher productivity.
Second on my list under productivity is system reform. In the last few months, our reports on supporting children with special education needs and NHS financial sustainability both identified the need for fundamental reform in the face of rising demand and costs, alongside unsatisfactory outcomes. This means tackling the causes of avoidable demand and allocating resources in a redesigned system where they can have maximum impact on outcomes.
Our recent report on planning and spending for long-term value for money sets out eight lessons from our work across government, to help ministers and officials as they carry out the Spending Review, with a focus on tackling the systemic barriers to better productivity.
Thirdly, investing in people, their skills and how they are organised, and delivering higher productivity through a culture of improvement. This I can illustrate with a simple example from our work on operational delivery.
The Independent Office for Police Conduct, which reviews complaints about local police investigations, encouraged staff to come up with better ways to manage workload. Previously, leaders set targets for casework teams to complete cases, and shared out more complex cases evenly across teams.
When the teams took on collective responsibility for allocating cases, they found that some caseworkers preferred working on more difficult cases. So they re-allocated to allow people to specialise in different case types. As a result, teams reported increases in every teams productivity, in one case by over 30%.
And fourth on my list is maintaining assets to support productivity and productive service delivery. Our recent work on school and hospital buildings highlights the negative impact of closed classrooms and closed wards on productive operation of those buildings.
The Office for National Statistics is working to improve the measurement of public sector productivity, tackling long-standing technical challenges in measuring it. We will seek to work closely with them to target areas of interest, track progress on this vital element of value for money and share good practice as we find it.
Resilience
My second big theme is resilience – or how robust the UK and its public services are to acute and chronic risks. Resilience is a word that crops up repeatedly in our work, particularly since the pandemic. We have much evidence that money is wasted and services to citizens compromised when we’re unprepared for what are now increasingly likely events, whether that’s pandemics, extreme weather or cyber-attacks.
All this points to the same thing: we need to be better prepared nationally and locally; to have sound risk management in place; and to be ready to adapt to new information and events quickly and effectively.
Recognising the importance of resilience in achieving value for money, we are systematically examining government’s preparedness for, and ability to respond to, the biggest risks on the national risk register. This isn’t an entirely new approach, of course. Arguably all our work on defence spending addresses the country’s resilience to military threats.
Last week we reported on the resilience of government and public services to cyber-attack, reaching the clear conclusion that insufficient priority has been given to protecting against this risk.
In November, we published an update on government’s work to ensure that owners are making multi-occupancy residential buildings resilient to fire risk in the wake of the Grenfell disaster. And this was the subject of an enlightening evidence session of the Public Accounts Committee just yesterday. We’ve also examined government’s preparedness for extreme weather events.
This work on resilience will take us into less obvious territory. For example, in the coming weeks we will be publishing our first report on the government’s work in the UK and with its international partners to address anti-microbial resistance. And this risk is as worthy of our attention as that of further pandemics.
I want to underline that efforts to build resilience are not in conflict with work to improve productivity. Adequate investment in infrastructure, better contingency planning, and agile and responsive services will serve both objectives.
A measure to improve short-term productivity that reduces resilience (like cutting back on essential maintenance) is unlikely to deliver long-term value for money.
Innovation
So all this brings me to the key to unlocking the gains we need in productivity and resilience. It’s innovation. The human brain is a wonderful thing, and we’ve been innovating since our emergence as a species.
Think of the technological developments we’ve seen in our own lifetimes, and we can celebrate the benefits of more recent – and successful – public sector innovations, from the literally life-saving development of Covid-19 vaccines to more recently reducing the time to process child benefit claims from three weeks to three days.
So what does the NAO’s work tell us about the fundamentals of effective innovation? I’ve picked out four key points:
- First, a clearly articulated risk appetite and a spread of investments, to maximise the chances of success in innovation
- Second, harnessing new technology as I’ve already mentioned
- Third, a culture of fast learning and evaluation, stopping failed experiments quickly and scaling up successes
- Finally – and close to home for us – an accountability and scrutiny framework that encourages well-managed risk taking
It’s no coincidence that innovation thrives in times of crisis, such as when lives are at stake. Organisations rapidly adjusted their risk appetites during the pandemic to meet urgent needs.
And our report on the MoD’s support for Ukraine showed how quickly and effectively procurement and other processes were adapted to that situation.
When the technical solutions we need to tackle big problems such as climate change are not yet fully developed or implemented, a portfolio approach to investing in research and development becomes essential. And this allows a risk appetite to be set for a spread of investments, recognising that not every new idea will bear fruit.
When we reported on innovation to deliver net zero in 2023, we found that government hadn’t defined the level of failure it could tolerate across the 31 research areas in the portfolio. The department was relying on risk management at the individual programme level. This made it harder for those spending the money to match their risk appetite to government’s, potentially constraining the level of innovation.
The clearest example of this thinking in action is The Advanced Research and Invention Agency, or ARIA, which has been set up with the brief, in their words ‘to empower scientists and engineers with the resources and freedom to pursue breakthroughs at the edge of the possible’. ARIA explicitly acknowledges that many of its investments will fail to meet their targets, but the learning from them will inspire those that follow.
And the investments that do succeed are intended to generate massive social and economic returns. When the NAO looks at ARIA’s use of resources in delivering its mission, we will do so against its stated risk appetite and approach.
I spoke earlier of the role of new technology including AI in driving productivity improvement. We’ve started seeing relatively straightforward applications of AI in government across the board.
The Department for Transport is trialling an AI tool to prevent fraud when installers apply for subsidies for Electric Vehicle chargers. It checks photos as proof of installation against identical, similar or stock photos across multiple applications.
To reduce duplication across departments, the government’s grants management team within the Cabinet Office has launched an online portal that advertises government grants in a single location and handles online applications, which are automatically checked as part of the due diligence process.
These are the foothills of a huge coming change in the design of public services and administration.
There’s a lot to manage if the productivity benefits of new technology are to be realised, either as service improvements, cost reductions, or both. This includes training, procurement, data quality and review of business processes. Government will have to continue to tackle its overhang of obsolete IT systems and patchy data quality.
And as our recent report showed, a new approach is needed to government’s management of its technology suppliers, so they can respond to the fundamental changes in the market for digital services, where the global providers are larger than most governments. And good governance will be needed to manage the risks accompanying AI, to ensure fair treatment of service users and maintain public trust.
Most important of all, leaders will require the knowledge and skills to implement these changes successfully. But the immense opportunities for greater productivity, and more user-friendly services mean that these are risks and challenges are there to be managed, not avoided.
Another feature of successful innovators is their ability to learn quickly what works and what doesn’t, so that failed experiments can be stopped promptly and the resources redirected to more promising ideas. Being open about this can be challenging for government, with its ingrained worry that any failed project represents poor value for money.
So it’s good to see the Department for Education piloting innovative ways to reduce emissions and increase energy efficiency in school buildings, in order to assess which ones justify future investment.
DfE is using an ‘innovate, test and invest’ approach to learn what does and doesn’t work and what it should scale up with the money available. It’s running a series of projects to improve energy efficiency and resilience to the effects of climate change, as well as using ultra-low carbon construction methods.
There is a vital role here for the finance function across government, ensuring that robust financial information is used to inform decisions on stopping, continuing or scaling up innovations.
The NAO’s role
This neatly leads on to the role of the accountability and scrutiny framework as part of a pro-innovation culture in government.
There are those who say that the combination of the NAO and parliamentary scrutiny can stifle innovation, because civil servants are concerned about being criticised for something that hasn’t worked.
I take this point seriously, although my reflection after nearly six years of attending sessions of the Public Accounts Committee is that its challenge is far more likely to focus on why officials haven’t done more to improve long-standing performance issues.
For the NAO, our refreshed strategy from 2025 to 2030 takes fully into account the risk appetite set for the range of innovative projects. We will continue to look for and highlight positive examples of innovation, including where unsuccessful initiatives have been stopped in favour of more promising ones. As well as featuring these in our reports on departments and organisations, we will publish what we learn across government as part of our programme of lessons learned reports.
This approach doesn’t mean any change in the rigour and challenge offered by our work. On the contrary, we will continue to ensure we have the skills and specialist knowledge ourselves to provide that challenge credibly and robustly.
But it does recognise the importance of setting a clear tone that value for money requires well-managed risk taking. We are taking the same approach to our own organisation. We’ve replaced our audit software platform and are implementing new automation tools to reduce the time spent on manual processes, allowing our auditors to focus on the professional judgements that make the biggest difference to audit quality and the impact of our work.
For the next five years we have explicitly made it our ambition to have a positive impact on more productive and resilient public services, and better financial management and reporting. We’ll aim to increase trust and value for money through the impact of our audit work and the credibility of our insights and recommendations.
Conclusion
Thank you for listening today. Just to conclude:
Better productivity and resilience are in my view the most important value for money challenges and opportunities in the coming years.
We won’t achieve these without effective innovation in how society’s problems are being solved and how services are delivered.
This will require leaders in government to:
- clearly articulate their risk appetite and take a portfolio approach to investing in innovation
- harness the potential of new technology; and
- develop a culture of fast learning and adaptation.
We, at the NAO, are here to support Parliament to hold the government to account for its spending and help improve its value for money. In doing this, we want to contribute to a pro-innovation culture in public services.
We look forward to working with you all in the year to come. And I look forward to the discussion. Thank you very much.
Watch the speech
Author
Gareth Davies
Gareth Davies was appointed C&AG in June 2019. Before his appointment as C&AG, he was Head of Public Services at Mazars, a global accountancy firm specialising in audit, tax and advisory services. Prior to this, he was Managing Director of the Audit Commission’s Audit Practice. His experience spans financial and value for money audit, organisational leadership and board governance.
Gareth is a Fellow of the Chartered Institute of Public Finance and Accountancy and a Fellow of the Institute of Chartered Accountants in England and Wales. He is a non-executive Board member of the INTOSAI Development Initiative (IDI) which supports Supreme Audit Institutions in developing countries to sustainably enhance their performance and capacity.
