Diane Coyle is the Bennett Professor of Public Policy at the University of Cambridge and co-director of the Bennett Institute. She is an economist and former advisor to the UK Treasury and was a member of the UK Competition Commission from 2001 until 2019.
This interview was conducted on 27 September 2021.
Q: What is the right unit of analysis when thinking about regional inequality? How is this tied to how we should think about economic growth? Does this vary by region – would a one-size-fits-all approach mean we miss some local challenges and some macro-ones?
It’s endogenous to policy decisions. Burnley could be a much more natural part of that Greater Manchester city-region if there’s more than one bus a day in each direction and if the Beeching Cuts hadn’t happened. Steve Gibbons’ work on cuts is really interesting in that respect.
I don’t know the answer to the question, but I think there is interest in thinking about places that are in different parts of the country, but similar to each other. I say that because of the work that I’ve done with Penny Mealy looking at complexity, which is really just a way of indexing what industrial structure different places have and comparing what would it be elsewhere. Thameside to Peterborough might be an interesting way to think about, both the causes and and the policy solutions to inequalities. The meaningful ones all seem pretty correlated with each other.
Penny Mealy and I are doing another piece of work now looking at health outcomes, particularly during the pandemic, but more broadly at public health outcomes and the Economic Complexity Index. It is a strong predictor of health outcomes. We found that in the Greater Manchester Update Review: these are interacting things.
I increasingly think about it as being just like an economic development problem for a low-income country – the backward and forward linkages that Ricardo Hausmann wrote about in that development literature. I think it applies very directly. You’re going to have to think about interventions, in some O-ring model. It’s all about coordinating and acting on different fronts at the same time, because if you don’t fix the health, then you’re not going to be able to upskill people to get into the jobs, and then you’re not going to fix the labour demand if the labour supplies isn’t there and on. (Although Ricardo Hausmann says the math doesn’t work for the O-ring model, intuitively it works.)
Q: Two questions follow from that: how do you tie this idea of complementarities into a mode of analysis when looking at individual projects pay off like cost-benefit analysis? Second, how do you then evaluate impact of a ‘big bang’ set of policies?
If you have a model of complementarities, then the only way you can evaluate the impact of individual interventions is by removing a part and see if the thing collapses. As we know from looking at infrastructure, you can’t prove the infrastructure increases economic growth. But equally if we didn’t have any electricity, we wouldn’t have an economy. That question is quite hard.
I’m doing a paper with Tony Venables and Garrett Brian at the moment to think about it the other way from the appraisal end and ask, can you think about a dose response function? Can you think about the bang for the buck from an intervention in this nonlinear way and what would be the thought process that you go through? In some ways the Green Book has made a lot of headway towards that. For this happenstance reason they picked on a paper I did with Marianne Sensier – which offers a partial, rather than rounded view, I should say – pointing to public transport decisions. They have now said in the 2020 update, ‘you need to think about the strategic context for any intervention’.
I think it needs to go one step further and say there’s an active, market shaping role here where the public authorities or the government need to make sure that the things happen that will deliver the higher return that’s consistent with the strategic gains. Ultimately it takes you back to not only picking winners but making winners.
There’s no analytic answer to this. It’s a question of having enough granular information about the strength of different places that you can identify the grit that becomes the oyster. The pearl will grow around. There needs to be something. Then you have to assemble all the components of the system around that initial strength. I don’t think there’s any solution to trying stuff and seeing what happens.
There you get into the real UK problem, compared to anywhere else – because these are universal issues, these are not just the UK specific. The UK problem is that we don’t let things run. We change things every year. I’ve got a paper looking at the vast number of changes in industrial policy over the years, it’s just absolutely a pig’s ear. The complexity of the institutional landscape and the fact that it changes constantly means that there’s just no developmental capability of the British state.
I think the missing link is, what is it about British political structures that make that happen? We’re not the only country that’s got a first past the post system. We’re not the only country that’s got centralist government.
In a paper with Bridget Rosewell, we wanted to make the argument both that national productivity would depend on other cities growing more, but this didn’t involve damaging London. The argument we were trying to make was that it was a positive sum initiative. Internal migration is the obvious thing to think about there, because most graduates move to London. I think Bristol retains on net some of its graduates, Manchester now does. But that’s pretty recent. To the extent that graduate skills are in short supply, fixed supply and there aren’t incentives, there’s that combined simultaneity of supply in a regional city and the kinds of jobs that hire graduates in a regional city. That’s going to be one of the key things that means London does disadvantage other places.
I used to think it was an issue of venture capital and bank finance as well, but there I now think it’s not so much resources as information. The kinds of people making financing decisions don’t know enough about the opportunities that they have. Capital here is just starting to go more to the northern cities than they used to. But that information that you need to decide whether to invest or whether to make bank loan… bank decision-making got centralised and the granularity of information and trust got lost from their decision-making processes. I don’t think finances are limitations in the same way as skills.
Q: When you think about the last few decades, has the pattern of regional growth been as good as it could have been given exogenous factors? If we failed, to what extent was that inevitable?
I think we’ve got a particular deformation in being the only country that bought thoroughly into Thatcherite ideas, without having the institutions or political structures that mitigate against that. The U.S. has had the same ideology but – being a federal system and the politics of places getting infrastructure around the country, and the importance of the defence establishment and success of Defense Advanced Research Projects Agency – they haven’t suffered in the same way.
We are the only advanced economy that for 40 years hasn’t believed in the state having a guiding hand in managing the economy. We’ve had a state with no strategic function. For things where cross-subsidies clearly matter and network planning matters – like public transport – they have just been left to market forces. That means you get unfair and inefficient outcomes because they ignore all the network externalities. In any of the continental European countries, that tradition of dirigiste in the different languages, different places is still very strong. That was never the same. Obviously, attitudes did change. It was a broader change in the climate of ideas. But they never gave up on the idea of a functioning strategic state in the way that we did because of Thatcherism.
Q: Was that consistently true? There were periods of larger public investment, and of devolution?
Yes, I think you’re right. What I had in mind wasn’t just big infrastructure like Crossrail, but also the failure to reverse privatisation of bus services and bringing in bus franchising, or to accept that actually local transport schemes often won’t be good value for money but are important for non-economic reasons. Economic efficiency shouldn’t be the only only criterion. But you’re absolutely right, I agree that you’re layering devolution or the lack of devolution on top of, a particular world view.
Q: How much of this was ideological, as opposed to a mere political desire to maintain decision-making power?
It’s not just political. It’s really deeply embedded across the Civil Service, and at the Treasury in particular. Moving a bit of the Treasury to Darlington is just moving a bit of the centre; it’s not really decentralising decision-making or whether or not councils should be allowed to issue municipal bonds to build council houses. We might want those sorts of differences, to vary in different places, depending on local preferences.
Q: What is driving that distrust?
My hypothesis would be that it’s about the public school system and Oxbridge. Tim Beasley has a lovely chart about the proportion of MPs and prime ministers who have come from Eton and Oxford. If you’ve been to Eton, you automatically think you know how, know better than anybody else.
Q: Are there periods of devolution over the past few decades that you think may have shifted this?
The devolution to the three nations has been the most interesting. They have now taken really quite distinctive paths. They’ve got more granular information about their own countries, and they have a different philosophy. Scotland and Wales – and I think also Northern Ireland, I’m less familiar – have moved to models where jobs and productivity matter, but wider wellbeing measures and sustainability are a core part of what they think they’re trying to deliver for their people. That’s still quite a different frame from the Treasury. I would say the same for Greater Manchester and West Midlands at the moment.
Q: Scotland and Wales especially flowed from a constitutional process more than an economic process. How much do you think we need to focus on structures and accountability, as a way of legitimising decision making?
Mike Kenny thinks about this much more, but I do think this is a political economy question and they interact. We, as part of the Productivity Institute, set up regional forums. East Anglia is just a dog’s breakfast. It’s not a natural economic region. You’ve got the Fens, you’ve got Peterborough, you’ve got the Cambridge cluster, you’ve got the coast. They’re all very different functional economies, and transport has been planned around the region. What I am observing, though, is that there’s a process of change going on in people starting to talk to each other much more – partly through the forum that we’ve established. This is bound to be a very long process, but I think those identities and accountabilities can change over time. Better political boundaries and administrative boundaries would make it easier. It is a question of people figuring out that they want to cooperate with each other, if they want to achieve the coordination that will deliver economic benefits for all participants.
Q: The current institutional arrangement leaves places out, against the Regional Development Agencies (RDAs) which were comprehensive. Is it better to force a bigger front print, even if the relationships aren’t there?
My view about it is that the RDAs were working pretty well. It was a real shame that they got abolished in one of those policy reversals. Again, that was about people talking to each other. A lot of people in Manchester were talking every day to people in North West RDA.
Q: Is there an optimal level at which we should be setting policies? Does it vary by places or issue?
I think it depends on which aspect you’re talking about. They’re probably different geographies that you need to think about different types of policy. There’s not going to be an easy, clean solution to this.
Q: Although Scotland and Wales started at a bigger geography and then reformed internally over the past few decades?
The politics and the cultural identity matter. People in Scotland identify as Scottish. I don’t know if you heard that programme about Wakefield a while ago on Radio Four. What I loved about that was the informative example that because people couldn’t cross a busy road, they couldn’t go to the jobs in the new warehouse that was established. You need something providing large scale transport planning for the rail network, but also somebody local knowing that they need to build footbridge.
ENDS