ICAEW.com works better with JavaScript enabled.
The transcript for ICAEW Insights podcast episode 27: 'Data, Devolution and Levelling Up'.

Philippa Lamb: “Hello and welcome to another ICAEW Insights podcasts with news and analysis from the world of accountancy, business and finance. I’m Philippa Lamb. Today we’re looking at the statistics behind the government’s Levelling Up agenda with Ed Humpherson, who leads the Office for Statistics Regulation. And we’ll be discussing the pros and cons of a more devolved tax system with ICAEW’s own Head of Taxation Strategy, Frank Haskew. 

“The Levelling Up agenda is all about changing Britain’s economic geography and narrowing inequality between the regions. These are substantial tasks involving huge infrastructure projects and changes to everything from taxation to regional funding. So to grasp what needs to be done, we need a comprehensive understanding about demography, and where those imbalances currently exist. But where does that data come from? And how do we know it’s sufficiently robust? Ed Humpherson is here to discuss the importance of good data in moving that Levelling Up agenda forward, and how statistics can be used to measure progress. For those who aren’t familiar with the Office for Statistics Regulation, can you just give us an idea of exactly what the OSR does?”

Ed Humpherson: “We are the UK’s standard set-up for all governments statistics and data. We set the standards that any government department must meet when they publish statistics on anything. It could be crime, or health or the economy, or housing – all human life is there in the work that we cover, and we check that they uphold those standards. We do various things: the most well-known is we step in where statistics are misused, right up to the Prime Minister; we will be willing to step in and highlight misuses of statistics.”

PL: “Levelling Up is all about doing away with historic economic disparities right across the UK. What does existing government data tell us about those imbalances?” 

EH: “The imbalances are really striking. I’ve got a few examples to illustrate how significant they are. At the regional level, you see that London and the South East have the highest GDP per head and quite a lot above the other regions of the UK. I’m sitting here in Camden, and since 2004 productivity per working hour in Camden has gone up by 42%; in Merseyside and Sefton it has actually fallen by 13%, so that’s a huge widening of the gap. GDP is up in Camden, over 200% since 1998, but in the Midlands – places like Dudley and Wolverhampton – it’s only up by 65%. So you see not only gaps, but widening gaps. 

“And it’s not just those economic numbers, it’s also health numbers. I always think the most striking and perhaps depressing imbalance is healthy life expectancy: how long somebody can expect to live disability-free. In Camden, the average male can expect to live to be 65, in Hull it’s 53. So 12 years less healthy life – a really striking imbalance.”

PL: “That’s a huge gap, isn’t it? But data is only useful if it’s good data. How do you go about monitoring government data to assess its quality?”

EH: “We have a framework which we call TQV – trustworthiness, quality and value. The starting point is that it’s not just the quality of the data that matters, it’s how trustworthy it is – whether it’s been produced by people who are free from political spin or political interference, and whether it’s valuable to the public. The reason why that point is so important, is that I think it’s very easy to imagine data just go into a darkened room in Whitehall and senior decision makers – civil servants and ministers – look at the numbers and come up with clever policies. But of course, that’s a really elitist way of thinking about data. Really, data are collected from society and should be available to everyone in society to understand. So we also look at accessibility and insight. 

“We check trustworthiness – who’s deciding what numbers are collected, and how; quality – what the numbers actually are, whether they’re reliable, where they’re from, the methods used, and then also the point about value how accessible they are. Do they provide the data in a way that answers the questions people want to have answered, like how much imbalance is there in the UK?”

PL: “So you introduce a lot of rigour into the checking process. But what happens if you’re not happy with a particular study or dataset?”

EH: “We make a noise about it. We publish a report, we will write publicly to the department, and we will be very clear that we expect improvements. If they don’t comply with our requirements, their statistics will lose the status of being an accredited statistic. That signals to everyone, to the government itself, but also to all the users, that the numbers have not met the high standards I’ve just outlined.”

PL: “So you have real organisational teeth?”

EH: “We certainly do. We find that we’re effective in getting government departments to improve their data collections. Data collection is never still – society is always changing – so statistics need to evolve and adapt. When we step in and we say something needs to be improved, we are listened to.”

PL: “What are the biggest data gaps and challenges that you see when it comes to Levelling Up?”

EH: “I think there’s four issues to address. The first and the most important one is quality. A lot of the regional statistics that we have, particularly the economic statistics, are not directly observed, not collected from observations; they’re modelled. What the Office of National Statistics does is build a model and say, this is what we think is happening, for example, for regional GDP. The trouble with that is that it’s not very good. If a particular area starts to really struggle, or indeed surge away, that model won’t be picking it up. So really addressing that by using more data sources, tax data, which tells you about business activity in a local area, or Companies House data, using that much more is a really important first step in quality. 

“Once you’ve got that, if you want to think about Levelling Up, you bring together lots of data about people and spaces, spatial economic data. While we have good granular data on people – health, education and so on – and we have reasonable spatial economic data, it’s quite hard to find it all together so you can say, this is what that place is like, because you can look at one source. So bringing it together is the next thing. 

“Even when you’ve done all that – you’ve got the quality sorted, you’ve brought the data together so you can look at a place – there are gaps. I think the most significant is that a lot of Levelling Up is about pride in place, it’s about people feeling that where they live and where they’re from is worthwhile – it’s got a good community and so on; that’s really hard to measure. I actually suspect that what government will need to do is to go outside the official data system, and use things from civil society, things which are more bottom up – surveys, or locally generated data – to get much more of an insight into what, beyond all the numbers, a place is really like, how it’s really doing, and how the people who live there really feel. It’s a little bit like the ghost in the machine.”

PL: “Tell us more about those gaps and challenges.”

EH: Well, the interesting development in data collection over the last decade has been this interest in wellbeing. So instead of just collecting numbers on the size of the economy, and unemployment, you ask people how happy people are and how satisfied they are with their life. That produces really interesting insights. Some of the areas which are wealthier, like Camden and Islington, don’t do so well on wellbeing. Some of the more rural areas, where incomes are less high, do really well. 

“Again, this is an issue where there’s a gap between that wellbeing insight, which gives you some sense of how people feel about their lives and where they live, and the harder numbers. I think what we need is to close that gap. If we really want to understand how to level up in a way that makes place a meaningful thing in people’s lives and something they can feel proud of, understanding these relationships between wellbeing and employment, wellbeing and health, wellbeing and crime, and wellbeing and education ¬– all of these things are really important.” 

PL: “Is there a tension between those different sorts of data? On the one hand, you’ve got hard statistical data, and on the other, you’ve got self-reported data about how people are feeling. Is there a difficulty in trying to combine those into solid insights?”

EH: “That’s what makes it interesting. It’s not something where you can simply write a formula, plug the numbers in and out pops your answer. It’s much more subtle. But I do think that if you drill down into both sets – the harder numbers that we’ve just been talking about, and these wellbeing numbers – you can start to see causal patterns and relationships between things, like how people’s health affects how they feel about their lives. But it’s not easy, and it’s certainly not something you can do in a very kind of glib way. It’s much more complex, and I think we’re at the foothills of understanding it. I think that’s really where understanding Levelling Up needs to go – to understand this kind of community wellbeing issue.”

PL: “So the skill is in the analysis as much as the collection?”

EH: “Absolutely right, and the inferences and where you get those inferences from and how you connect things up.” 

PL: “How will you be monitoring and measuring all the data that’s going to emerge from this levelling-up strategy? It’s a 10-year plan, so it’s going to be vitally important, isn’t it, to have an accurate picture of progress over time?”

EH: “I think a really bad outcome in terms of statistics and data would be for this process of measuring progress against Levelling Up to be a Whitehall activity – for all the data and statistics be funnelled up and at the end of the 10 years, for the civil servants to emerge blinking into the light to say, we can now tell you that we have successfully levelled up. I think that would be the antithesis of the whole project. The whole project is about saying we’re too centralised in the UK and one of the manifestations of that centralisation is in really, really huge imbalances. 

“So what I’m going to be looking for is that at every step along the way, the statistics and the data are not just hoarded centrally, but made available publicly in ways that are meaningful to those local communities and the people in them as things are changing. You might think that’s a bit of a dreamy fantasy for the data lover that I am. But we’ve just been through a pandemic, where exactly that has happened – where exactly this process of making data available on a daily basis to millions of people happened and people have engaged with it, they have wanted to know, because it’s been easy for them to engage with. That’s what I’m going to be looking for. 

“As we start to tease out these interesting relationships between how people feel and the kind of material dynamics of a place – as we tease that out, we’ll be taking people on that journey. I can’t believe I’ve just said ‘journey’ – my staff are banned from using the journey word!”

PL: “Are you expecting to get buy-in on that hope from Whitehall, from government? Do you think they’re going to be as keen to be open as you would want them to be?”

EH: “Right now, the signs are really good. The Department for Levelling Up has set up a new Spatial Analysis Unit and it’s very keen on open data and making things accessible. I think the test comes not when you set out, but further along, as maybe things start to get harder, or political interest moves on; is that same commitment to openness still maintained? That’s what we’ll be looking for. But right now the signs are good.”

PL: “Thanks very much Ed, it really great to have you with us. 

“Now let’s turn to the tax implications, because one of the big Levelling Up questions is whether to further devolve taxes, both between nations and at a regional level. This could give regions more autonomy and it would help address funding gaps that have caused problems in recent years. But it could also add more cost and complexity to the tax system. So let’s look at those pros and cons in more detail with ICAEW’s Head of Taxation Strategy, Frank Haskew. Tell me, what do you feel are the big benefits and risks around further devolution of taxes generally?” 

Frank Haskew: “That’s a very good question, and it’s a very opportune time for considering it because there’s quite a lot of change currently in the pipeline. There are things like the Levelling Up Bill, and also the recent publication of an independent fiscal commission report into further devolution of taxes in Northern Ireland, which is a very lengthy and well-considered report, which actually sets out a lot of the risks and rewards of tax devolution. In fact the chair of that was Paul Johnson of the IFS and it’s certainly well worth a read. 

“But in a nutshell, the advantages of devolution really stemmed from things like greater flexibility, the ability to respond more quickly to local conditions, the ability to potentially spur economic growth and effectually, to be able to make the best choices at the local level. But obviously, that comes with a lot of risks as well. So potentially, there’s obviously a downside – revenue risk – and potentially a lot more complexity could be added to the system. So it’s quite finely balanced; if we get it right, potentially, it’s good. But, you know, the long-term implications of getting it wrong could also be quite serious.”

PL: “If we’re thinking specifically about the Levelling Up programme, how might devolved taxes play into that?”

FH: “The Levelling Up programme, and the Levelling Up Bill that has been published, don’t really talk in great detail about the role of tax. The bill itself has provisions in relation to a thing called the Infrastructure Levy, which is effectively a levy on development. But other than that, the actual bill itself is quite silent in terms of how tax might play into this agenda. So obviously, we have devolved taxes already on stream in both Scotland and Wales. But how I think they’re going to play into the wider levelling-up agenda, particularly at the local level if you live within regions of say, England – I think at the moment it’s still really not entirely clear how it’s going to work.”

PL: “Why do you think they haven’t talked about tax?”

FH: “Because tax is a very complicated and difficult subject in its own right. We have one of the most complicated tax systems in the world and it requires a lot of administration and systems to get it right. So straightway you are into the problem of how to both design and administer taxes. Currently, effectively, most taxes are obviously done centrally with HM Revenue and Customs, so devolving that is actually incredibly problematic and time-consuming. Also, obviously, some taxes lend themselves much more easily to – effectively – devolution than others. For instance, both in Wales and Scotland, we’ve seen devolution of what I might call basically land taxes – effectively the UK equivalent of stamp duty land tax and landfill taxes. So it’s much easier to ‘ring fence’ those, if you like, and treat those separately. But if you’ve got something like income tax, national insurance and VAT, it’s much, much harder to do.”

PL: “From a national perspective, thinking of Scotland, Wales and Northern Ireland, do you see further devolution as inevitable?”

FH: “I think what we’re likely to see firstly is probably a period of reflection as to how the existing devolution of taxes is working out in practice. But I think in the longer term, we probably will see further changes in the pipeline. Going back to, for instance, the Northern Ireland Fiscal Commission Report, effectively we’ve already had that on the stocks since 2015, but the ability to devolve corporation tax rates to Northern Ireland has never actually been exercised. But the latest Fiscal Commission report, I think, has brought it very much back onto the agenda. I think what we will see is, across all the nations of the UK, a slow move towards potentially further devolution, but recognising that it takes time – they need to be designed. And there needs to be examples of, for instance, the block grant adjustments that will also need to be thought through. So I think it’s a long-term game, if I can use that expression, which will take many years to actually work through the system.”

PL: “So what would more tax devolution mean for the central tax system?”

FH: “In relation to the UK, we obviously have HM Revenue and Customs which is effectively the guardian of the UK tax system. That role, I think, is certainly likely to continue for quite a long time in terms of devolved taxes generally, because we’re likely to see that HMRC is still going to keep central records of, for instance, national insurance, which is currently effectively across the whole of the devolved nations. So there’s going to be a long-term role for HMRC in tax collection. 

“What we might see, and I hope we’ll see, is that developments in the devolved taxes might actually encourage, if you like, a move towards a more agile and responsive tax system that’s more accountable to local needs, and that that will feed back into the centre and HMRC, and make them more responsive to taxpayer needs. So I think it potentially is a good thing, and will encourage HMRC to develop and improve its tax systems.” 

PL: “As you say, that sounds positive. Do you have any specific tax changes you think we’re likely to see in the foreseeable future with regard to Levelling Up?”

FH: “As I mentioned, we could see further devolved taxes coming on stream, particularly in relation to, say, Northern Ireland, following the latest Fiscal Commission Report where, for example, the UK is looking to increase its corporation tax rate to 25%. I think this will renew calls to review whether we should devolve corporation tax to Northern Ireland given that, currently, the Republic of Ireland only has a 12.5% rate of corporation tax. I think that’s all going to come back onto the agenda. 

“VAT was always very much an EU tax. Now that the UK is no longer in the EU, obviously, it’s a UK tax, and we could possibly see now much more scope, I think, to consider regional developments, and possibly even rates of VAT. I think that could be quite an interesting area. We’re also likely to see, I think, further devolution of some niche taxes like passenger duty, which is certainly on the stocks. I think there’s going to be quite a lot of developments in this area over a number of years.”

PL: “Any sense of the timeframe?” 

FH: “We’ve already been into devolution of taxes for almost 20 years. It started back in 1998, so it’s already been continuing for many years. But my guess is, we could see these further developments, for example with Northern Ireland, if they can agree the funding of it. I think we could see that quite quickly – maybe in a year or two. I think we’re likely to see developments, certainly over the lifetime of this parliament and probably, I guess, further changes coming in a parliament after that. ”

PL: Many thanks to our guests Ed Humpherson and Frank Haskew.