Henry Overman is Professor of Economic Geography at the London School of Economics. He is also Research Director of the Centre for Economic Performance and the director of the What Works Centre for Local Economic Growth.
This interview was conducted on 13 March 2023.
Q: Could you tell us about yourself and your role in shaping regional growth policy?
I’m a Professor at the London School of Economics. I do academic research on the causes of spatial disparities.
I’ve also had a policy role, for the last 20 or so years. I’ve been involved with various efforts in government to deal with our spatial disparities. That’s ranged from being an expert advisor – on things like the Eddington Review; the HS2 analytical challenge panel; and various other things – to running projects funded by a mixture of government departments and Research Councils. This started with the Spatial Economics Research Centre, which I was involved with from 2007 to 2013.
Since 2013, I’ve run something called the “What Works Centre for Local Economic Growth” and I continue to do a mix of academic research which I seek to publish in academic journals, and a whole set of policy advice.
Q: When you look back over UK policymaking since 1979, what do you see as being the chief success?
The chief success, if we’re talking about spatial disparities, is that our core cities were in a terrible place at the end of the seventies – London, Liverpool, Manchester. In fact, this carried on clearly into the Eighties. We know what was going on in the macroeconomy; places like Liverpool lost a million people during that time period. London was also struggling. One of the successes is turning that around. We were dealing with a massive structural decline.
We have had some success stories beyond that. London’s noticeable, but places like Manchester – relative to what was going on in the 1980s, early 1990s… we’ve at least stabilised things. I think that’s the success.
The flip side of that is that our overall performance isn’t great. The Centre for Cities had something, for example, showing London’s overall poor performance relative to some international comparators, which showed . We’ve also struggled to narrow the spatial disparities back down in any meaningful way. We stopped things from being terrible, but they’ve not exactly improved a lot.
Q: Does it surprise you that we’ve seen rising spatial inequality over the past few decades?
I think that, everywhere, we’ve seen that the convergence process has stopped and gone into reverse. The fact that convergence has stopped is common. I don’t think we’re a huge international outlier, but our spatial inequalities are reasonably large. Part of that is a manifestation of our levels of personal inequality; but some of that is clearly to do with the fact that we’re not getting urban and regional policy right.
Q: How much of that is due to London’s role in the UK economy?
That’s partly driven by London. If you abstract away from the high-performing London and the South East, then our spatial inequalities don’t look huge. For our regional inequalities, a big part of that is a London story: London success and the end of convergence. So London’s part of the story, but not the only part.
Q: What are the other factors holding back convergence?
There’s a whole set of things we’ve not got right.
We’ve struggled to effectively diagnose the problem. Some say that industrial structure doesn’t matter very much, but I do think that one thing that matters quite a lot is the fact that we’ve become so specialised in services. To be competitive in finance, insurance, real estate – and a whole bunch of things that we do – requires big cities. It is hardly surprising that we’ve got big city success in London. If you don’t understand the problem, starting to formulate policy responses is difficult.
Then there’s a whole set of national actions which play out differently spatially. We’re not great on personal income inequality, and that’s always going to manifest spatially if high-income people live in one place and low-income people live up in another. So some of the story is rising individual inequality. (Although we’ve managed to draw a line on that; the Deaton Inquiry suggesting it hasn’t gone up so much over the past couple of decades.) Interpersonal inequality is still pretty high in the UK.
Then we’ve underinvested in infrastructure. We’ve under-invested in Research and Development. We have short-termism: our companies don’t invest properly in training. Our planning system is not fit for purpose and causes all kinds of problems. That broad set of problems are part of the reason why our collection of cities isn’t doing very well. They’re an important part of the macroeconomy. We’ve had macroeconomic mistakes or haven’t fixed challenges. We haven’t spent enough, probably, on dealing with regional disparities.
We were overly centralised. I’m not a ‘silver bullet’ person on decentralising, but when you’re one of the most centralised countries and things aren’t going very well there’s clearly a case for being more decentralised.
From short-terminsm, we have all kinds of instability, as we haven’t stuck to programmes. We had the Regional Development Agencies, which had some good aspects and bad aspects. Then we scrapped them. We’ve done Devolution Deals. They cover only part of the country. It is just a whole host of “muddling through” responses, which tend not to deal with the policy problems.
Q: What do you think is the policy problem that regional policy ought to aim to address?
Understanding the extent to which spatial disparities contribute to our underperformance in the macroeconomy is a priority. There’s an argument to be made that having these second-tier cities underperforming isn’t great for the macroeconomy. I think part of it is about macro performance. The other part of it, which has always interested me as much as the macro performance, is the extent to which it matters, and has consequences for individuals. There, I think the narrative is quite complicated. I’m interested in the extent to which we could improve outcomes for poorer families and households through effective urban and regional policy.
I don’t necessarily think that those two questions [improving individual outcomes and supporting the macroeconomy] were always perfectly aligned. The answers to those two questions are quite different. One of the real problems that we’ve got is that I don’t think most politicians are very comfortable grappling with the fact that what might be good for the national economy is not necessarily what might benefit individuals.
To give a concrete example: if politicians want Manchester to contribute to driving national growth in the UK, they should do a set of things there to benefit higher skilled workers: invest in research and development; move the BBC; etc. That helps Manchester contribute to catching up with London and helps with the macro problem. But none of that benefits people at the bottom of the income distribution. In fact – given what we know about the planning system and what’s happening to house prices, and the fact that the education system means that kids from poorer households are much less likely to get a degree or be able to participate in those kinds of activities – arguably it could make the poorest worse off in real terms. Despite all the rhetoric around levelling up, I still think that some of our most disadvantaged communities are found in our largest, most successful city [London].
Q: There are people who would advocate for a target of simply trying to equalise performance across regions. What do you think to that?
It’s not for me if the policy is all about improving the macro performance, and it’s not about helping individuals.
The only reason for me to care about narrowing the disparities between places is either that it’s going to improve the performance of the macroeconomy or it’s gonna help individual households and improve their well-being.
You can have policies to target both. However, if you came to me and said, “I can pull a lever that would narrow the gap between Manchester and London”, and I asked, “will it benefit the macroeconomy or make anyone that lives in Manchester better off?” and you said no, I would be sceptical. I’m not saying it’s always and everywhere a trade-off between those two things; I’m just saying that those are the two questions I asked myself about every single intervention.
Q: How much of the poor performance of UK macroeconomic productivity is about the underperformance of places rather than individuals and firms?
I don’t think that’s the right way around to think about it. If you wanted the UK economy to be performing better, I think you have to turn around the performance of some of our cities outside of London. That’s partly an economic statement; but it’s also a realist political statement. A strategy for the UK that just says “we’re just going to go with London and the South East” is politically unsustainable. The productivity of Manchester is about 30% lower than the productivity of London, depending on what metric you use. If you go to France, the gap between Paris and Lyon is about 20%. I’m sure there are things we could do to get the productivity of Manchester closer to London. That would be good for Manchester and good for the overall macroeconomy.
I think that the trickier issues are the very small, very rural communities where a relatively small number of people live. I don’t think that changing the performance of them is a big part of the story. That’s why I always come back to the fact that it’s turning around some of our big cities outside of London is going to be important. I think putting an exact number on it is a bit tricky. I probably could start to decompose it, but a big chunk of it is just the fact that London underperforms Paris and the whole distribution of place performance has shifted to the left. Our productivity has also got a wide variance. Precisely what the balance is between those two things is a bit tricky to say: you want a bit of both.
Q: If we were in a macroeconomically spatially efficient equilibrium, then we shouldn’t be trying to reshape activity across the country; we should be trying to make sure wellbeing was equal across communities. Do you have a sense of whether wellbeing is more equally balanced across the UK, even if economic outcomes aren’t?
In our Deaton Report, we look at wellbeing disparities. They’re much narrower than productivity disparities. What drives that is that our most productive places are also our most expensive. People care about their real standard of living, not just their nominal wages.
You need to be careful because, on individual inequality, the spatial stuff hardly matters at all. It’s all who you are. If you’re a highly educated person living in Manchester or London or Blackpool or Hull or wherever, your real income and wellbeing are relatively high. If you’re a poorer household, less educated, struggling with mental health challenges, etc, it doesn’t matter where you are: it’s just terribly unpleasant to be poor and vulnerable. For individual inequalities, place doesn’t matter very much at all. I know that’s surprising, but if you ask any London MP who’s a representative of a poor London constituency, problems here are every bit as profound as they are in some of these places that we claim to need “levelling up”. Our relative child poverty numbers are terrible in London and the South East.
I definitely think there are things we could do in Manchester, for example, that drive up employment opportunities in Manchester. I think that there are things that we can do in Oldham to make sure that kids growing up in Oldham are able to access those opportunities. I think the much more difficult thing to achive is when I want the opportunities generated in Oldham.
Q: What has held back productivity in the UK?
You’ve got a classic investment problem. We’ve got a skills problem: We’ve got a large university educated population; but we’ve got a terrible problem with the 50% who don’t go to university. We’ve got a school problem, in terms of individual educational outcomes. We’ve got not so much an innovation problem, but we’ve got an implementation problem. We’ve got an under-training problem. That’s a whole set of things. We’ve got a planning problem; the spatial side of it, it’s quite interesting. I don’t think it just comes down to the transport infrastructure.
If we want to improve the performance of Manchester and a few of those other second-tier cities, we’ve got to tackle the combination of public infrastructure and the private investment that it facilitates, and skills and human capital. They’re short on all of them.
The other big issue is I think is the effective size gap between Manchester and London or between any of those cities in London. London has an effective size of eight million. Being generous, Manchester has an effective size of two and a bit. Birmingham is probably smaller than that.
All of these things matter, it’s not just one thing.
Q: If we care about national productivity and London is our best bet, should we be prioritising public spending towards London and the South East?
The ‘volumes of expenditure’ debate is so deeply frustrating to be honest. I don’t find it very helpful. The key is spending money effectively. For example, if you’re putting in place effective transport infrastructure – say in somewhere like Manchester– then ask what do you think the mechanism is for how that turns into improved productivity in Manchester?
Q: I would’ve thought that by raising the effective size of the city, it improves efficiency of labour and product markets – an agglomeration effect?
Our central estimate from the academic literature is that doubling city size gives you 2% on productivity. People struggle to grapple with that. So there’s no way that you can close the gap between regions through the direct size effect on productivity. That just won’t happen.
The way that it will work is if Manchester becomes much more attractive to the most highly skilled graduates. I think that what happens is you improve the functioning of the public transport system, and the housing market, and it attracts skilled workers. That then makes Manchester a more attractive place for inward investment for the private sector. It’s got to work through attracting skilled workers and investment. That tells you that the fact that the skilled workers aren’t there is a major problem for the Manchester economy.
Q: Couldn’t this be about the supply of high-paying jobs from businesses, rather than the supply of skilled workers? That is, it’s a skills-demand problem in places like Manchester, which otherwise has a ready supply of graduates.
I think it is problematic for you to get that conclusion, for two reasons. One is a narrow point: observable characteristics, in our Deaton review work, don’t explain much of the spatial inequality that we see. I would actually say the problem is the unobserved characteristics: not all degrees are equal. If you take the upper end of the graduate distribution – the most labour marketable – they’re getting pulled into London and the South East. If you look just at those with STEM degrees, there is little regional difference; whereas if you look at those with management or education degrees (all those things, it’s quite broad), you can’t observe the differences in quality across the regional inequalities. If you want to narrow the gap from within Manchester, you need a different set of people living there.
Two consequences flow from that: one is that you can’t do it [narrow regional inequalities by changing the population profile] everywhere. The other is that you have to think hard about the consequences for the lower-income families. If you just go down that route of saying, “well it’s about infrastructure spend and maybe some higher skills”, you’re not forcing people to face that tricky trade-off.
Q: The trade-off is that you end up displacing activity from either London or from one city to another, and you don’t actually have a net improvement?
There are two trade-offs, this first of which is: narrowing the gap is about attracting people that would have gone to London to stay in Manchester. You’re stopping someone from moving from a place which has a 30% productivity gap. The average graduate or highly skilled person that moves from Manchester to London increases their productivity by some percentage.
The other trade-off is the fact that because you’re talking about graduate retention in Manchester, you’re not talking about kids at the bottom of the labour market.
Q: Does the second point need to be a trade-off?
It’s a consequence, that’s for sure. Resources spent on trying to attract more graduates to Manchester could be better served on supporting the wider Manchester population.
When people say that what we need to sort out Manchester is more investment in public transport, or for the R&D budget to be skewed towards Manchester, I just don’t think that that grapples with the deep changes that you need to happen to Manchester for it to narrow the gap with London. You need quite a fundamental shift in the graduate retention rate to retain more of the people that currently do move out of Manchester. I would say maybe they could do with being 10 percentage points higher with some of the highest performing graduates, being part of the graduate labour market. That’s a huge shift for an economy.
If we pretend we can do it all without the population composition changing, then the policy might seem “easy”: all you have to do is up the transport and R&D spend. But I think that is only one part of the mix. I just don’t think it will narrow the gaps very much without these bigger population changes.
The other thing is that you then get into thinking you can do it everywhere (by increasing transport spending or R&D). I don’t think that can be true either. I think it will improve the productivity performance in Leeds or Sheffield or Bristol if we improved public transport spend. I think it’s an exaggeration to pretend that it’s going to massively narrow the gap to London, because the big narrowing of the gap requires a change in the composition of the workforce.
Q: When you look back over the past few decades, have there been periods where our policy regime has been more effective?
I think New Labour had policy stability but the wrong spatial scale (larger regions). By accident, the Coalition got closer to the right spatial scale.
I think that the move from RDAs to LEPs had the opportunity to get us to the right spatial scale; but doing it before specifying what their powers were going to be – or what you wanted them to do – is not good government.
I think the government have been grappling towards getting it a bit more right with the “Deals” approach. I actually think it’s not a bad way of going about it; you try to incentivise them to form sensible geographies. There’s a balance between the top-down and the bottom-up. New Labour were a bit too top-down with the regions, and LEPs were just a bit crazy bottom-up.
The right approach needs to be somewhere in the middle of those two. The position we have gotten to is not a bad one. It has taken ten years. The national policy stability hasn’t been there. I don’t think we have a route map for what we think the devolved powers should look like. Nor what we think would be sensible things to put at the Combined Authority level and where we would like ourselves to be by 2030. It is also getting pulled around all of the time by the need to deal with the politics of the towns and the coalition that elected Boris. I just don’t think we have a concrete vision of getting to a more decentralised system.
Q: How important do you think it is to have a devolution map which covers everywhere?
I can live with it evolving. I would try and fix the places where it’s obvious that that level of government would be a good thing: Southampton and Portsmouth is an interesting example, where you’ve got large, functional urban areas which have the capacity.
I see that that leaves rural places behind. I just don’t think that they’ve got the same challenges; they’re not going to play the same role in the macro performance; and it affects fewer people. I’m sure there are things we could do about the way that we provide funds, for example. There are obvious ways of doing stuff that might then benefit the more rural places, such as less competition-based funding. We could have more rules-based funding, some decentralisation of power. For most rural places, if we could reverse the terrible austerity that we’ve seen, I can live with the governance being uneven.
Q: Is it inevitable, given England’s politics and economy, that there won’t be any effective fiscal devolution?
I know people have very strong views on this. I think, increasingly, that a long-term funding settlement should come first. If I was Andy Burnham or Andy Street, I would eventually be wanting some tax-raising powers, but I’m not sure that I ask for that first, in a world where I’m 30% less productive than London.
The number one thing you want is a ten-year fiscal envelope. Then, you go away and tell us how you’d prioritise your infrastructure spend and all of these things that they would say. We try to narrow the disparities. I think I would start on that side of it rather than the fiscal powers.
I think you want to have them eventually, to get the incentives right. But if you do it now, they’re working off of a weak fiscal base. My feeling is that you’ve got to do something about the fiscal base before you start giving them the tax powers – but I’m a little bit outside of my comfort zone on that.
Q: How important is the quality of local leadership?
I think that it is important, which is why I quite like the devolved model. But when you’ve had a system that is so incredibly centralised for a long time, jumping across to strong local institutions from weak ones is tricky.
I think that that current devolved model – where you see Manchester taking some response, and begin slowly investing; then you hand them powers; and then they’re able to take responsibility and more powers – feels to me like a good way of building up institutional capacity at the local level. Quite a few places would struggle if you suddenly handed them over. Something that manages the transition carefully is going to be important.
Infrastructure prioritisation, the local education system, a realistic plan for commercial and residential space, running public services that attract talented people in…we need good provision and good leadership at the local level. We need the same at the national level, and have seen the problems of undermining those two things.
Q: How attuned is Whitehall to those dilemmas?
It varies hugely. All of them are caught in some version of the conversation we’ve been having, of the chicken-and-egg problem. There is probably quite a lot of recognition that we’re overly centralised – across both officials and politicians. But moving from a centralised political system to a somewhat more decentralised political system is quite tricky. There’s lots of disagreement on how to get there.
I think the Treasury is right to worry that we could have a lot of ineffective infrastructure spend, if we just went with very localised decision-making; but having said that, we’re pretty useless at doing that on a national level – as with HS2.
Let me try and give a more concrete example. Reversing the Beeching Cuts (to local rail in the 1960s) is a colossally expensive piece of pandering to a small political electorate. With devolution, you could end up with a whole bunch of projects that look like that. We’ve got tonnes of examples of constituency-based politicians in tension with national priorities; local government is a big screen version of that. To be clear, this is a problem for national and for local government. To get around it, we need some national prioritisation of decisions, and that is in tension with decentralisation.
Q: What’s the one innovation or policy that you think would make a difference?
Prioritisation. We’ve got to bite the bullet that it’s about turning around one or two places outside of London, as an immediate priority for 2030. That requires spatially-focused investments in those places, to try to generate opportunities; alongside a set of investments to make sure as many people in those areas can take advantage of the opportunities that you’re going to generate. I would do that within the devolved model.
Q: There is tension between saying the solution is devolution and needing to prioritise by taking a macro-strategic view?
I said I’d do the expenditure side before I did the revenue side. Because it does require a transfer of funds. If you’re going to move Channel Four out of London, why aren’t you putting it where the BBC is, if what you want to do is create a counterbalance to London, agglomerate? If you want to move public sector jobs out of London, why are you putting some in Darlington? We need a balance to London. Manchester and Birmingham are two obvious places to try and drive that. There are all kinds of other expenditure decisions that you’d make for all places – trying to improve mental health support, trying to improve the NHS. But on the economic growth, I think that we just need to focus some of it.
ENDS