Thank you and good morning. My normal role when speaking at Conservative events is to be the warm-up man for the raffle, but I’m delighted that today they don’t have a raffle so I’m not doing that. Thank you for joining. Since I came to government just under three years ago, the commitment and hard work of civil servants to make government happen has been so impressive. So I want to begin by thanking you for all you do, and what this is about making your efforts go further so that we can have an even better and more effective government.
Opening today’s session of this Policy that Works conference, I had some thoughts on why evaluation and evidence-based policy design are vital for delivering value for taxpayers whom we serve. This great bugbear of mine is there is no such thing as public money. It’s always taxpayers’ money. My combined responsibilities for Brexit opportunities and government efficiency are one and the same with total governing independence outside the European Union.
It has never been so important to ensure the Whitehall machine is as finely tuned as possible – the Whitehall machine, that great Rolls Royce! We are to be one of those fantastic, classic Rolls Royces that purrs away and the loudest noises stop. The government currently spends around £1 trillion of taxpayers’ money a year . It’s a vast sum of money and it’s crucial to know how well this is spent and whether it is funding effective and meaningful policy. This should be at the front of all civil servants’, and indeed politicians’, minds. You are custodians in this regard. We must not waste taxpayers’ money or your own energy on ideas that won’t ultimately work.
Analysis undertaken by the Cabinet Office, the Treasury and Behavioural Insights Team found only 8% of government spending on major projects in 2019, £35 billion of £432 billion total expenditure, had robust evaluation plans in place. This isn’t where we want to be. And it’s now attracting criticism from the National Audit Office into the evaluation of government spending. For evaluation to work in policymaking, it must have a clear plan for results. There must be scope to deliver things in a new way to scale up or to stop altogether. Evaluation must be informative every step of the way, not just at the end of the process, or indeed some years after the process is completed. Let me look at these 3 points individually. When embarking on a project in your role as a civil servant, you need to plan for evaluation at every stage from the beginning.
The first question you should ask when evaluating your own work is the most important. How will the policy I am developing make people’s lives better in this country? That is what we’re here for. And some departments have established organisations that sit alongside them to drive evaluation, like the Education Endowment Foundation created by the Department of Education or the role played by the National Institute for Health and Care Excellence for the Department of Health and Social Care. But there is still more to do to ensure all departments are operating as effectively and efficiently as they can.
Hence, the focus on evaluation in the context of decisions seen at the Spending Review 2021 is here to stay. It is to be increased by challenging departments to tackle evaluation without being afraid of the results. As for those departments that have established What Works Centres, have you now learned what does not work and what are you doing about it? It is not only about what works but also what doesn’t work.
We now move on to thinking more about how you do what we are asking you to do: how to implement effective and robust evaluation. We can do this through testing small innovations or variations of policies or programmes to identify which aspects of our proposals are effective and which are not. This is a valuable process as it enables us to explore whether some ideas might fail while minimising costs for the taxpayer and saving money in the long run from the experience.
One useful example of this is an early experiment HMRC ran to review how they write to people who were late in paying their taxes. HMRC compared different wording in letters to late-paying taxpayers using a Randomised Control Trial, it was found that adding the line “most people pay their taxes on time and you are one of the few yet to do so” into correspondence, it led to a 15% increase in the payment rate without further prompt. It was also associated with fewer complaints to HMRC, compared with our previous, more complicated and aggressive letters. As we all know, taxation is about plucking the goose with the least amount of hissing. By saying most people pay their taxes on time and you haven’t, by not being rude and aggressive, more taxes are being paid. It was fantastic and that’s a fantastic way of evaluating things. And HMRC didn’t stop there – it took on board the deeper lesson that it could improve the performance of its services for the taxpayer by making their letters easier to understand. They began using Anglo-Saxon rather than Romanised words, but not all Anglo-Saxon words are suitable for correspondence from HMRC. They improved comprehension, in turn, reducing frustration caused from confused taxpayers while improving payment rates and even increasing satisfaction with their service – a mean feat indeed.
We must all start with small pilots and work up to a bigger rollout. We must not be afraid of failure. If the pilot is not successful, go back to the drawing board and try again. No one will criticise you for that. And there is support available to you at this crucial point on the back of our commitments in the 2021 Spending Review – which I’ll come to slightly later. So often in government combinations of inertia and sunk cost fallacy leads to colossal waste and failure. One of my biggest bugbears is the sunk cost fallacy, if £10 was wasted we shouldn’t waste another £10 to justify it. Proper evaluation and honesty can stop this.
While pilot projects are important, it is helpful to reflect on what evaluation methods we can apply when we are happy that a project can be rolled out at full scale. An impact evaluation will mean we catch and fix problems early on in the rollout. We would rather do this than have an extensive list of lessons learned at the completion of a project, many years later, along with the vague hope that we can do better next time. Instead, we must do better at this time and use impact evaluation as a live tool to improve the customer experience in public services at the time people use them.
This will also help us tackle the problem we have in that we aren’t generating enough evidence of what works to help inform future projects. You might ask: what support is available to ensure I can meet my evaluation obligations across the three areas set out above? Emphasising that we recognise this way of working may require change in some departments’ traditional working practices, our commitment to put the outcomes of real-world spending impacts at the heart of decision making is why we established the Evaluation Task Force. The Evaluation Task Force (ETF), a joint Cabinet Office and HM Treasury unit, aims to improve people’s lives by ensuring robust evidence on the effectiveness of policies and programmes sits at the heart of government spending decisions. Already, they are doing a fantastic job of promoting and furthering the use of robust evaluation. They played a significant role in the recent spending review, ensuring spending decisions were informed by evidence. The team has also launched a £15 million Evaluation Accelerator Fund. Over the next three years, we will accelerate evaluation activity, creation of actionable evidence, tackle evidence gaps, and provide vast evidence of financial or efficiency savings from new interventions. I would strongly encourage you to consider how we might make use of this fund to answer pressing questions in priority policy areas.
The ETF website is also helping to drive our commitment to government transparency. In the last few weeks, the ETF has been starting to feature some of the government’s top evaluations. You will see a number of these top evaluations featured at this conference. Many of these from complex policy areas, where departments have been told evaluation would not be possible. Evaluation is always possible. Using new analysis techniques, high-quality evaluations have been developed, counterfactuals established and the impacts of our interventions have been learned and improved. Therefore, I would encourage you to draw on their services and drive them to provide you with techniques for evaluation you can implement in your pilot projects and impact evaluations as you deliver for the British public.
All of this links back to the notion I introduced at the beginning: we need to be bold, not afraid of failure and have more evidence-based approaches to deliver an effective service to the taxpayer. I hope today has given you some sense of why I feel this agenda is so important and how you can go about implementing evaluations yourselves. Not just to deliver the same service or policy advice you gave yesterday, but to experiment, evaluate and improve the efficiency of government. We must find out what works and what does not work. Not only what works overall, but what works for whom, where and why. This is how we better deliver for citizens. This is how we build a modern civil service that supports us to be our best. So, let me close by thanking you once again for coming to this event today. For showing your willingness to ensure that the civil service can be as good as possible. Thank you for all that you do on a daily basis to ensure that the country is well-governed. The real Rolls Royce of Britain.