Insights
With the Autumn statement published today, there is concern that the Government might cut infrastructure budgets that are viewed by many as critical for increasing local economic growth and helping level up struggling parts of the country. Ongoing disruption to rail services for many places has served to increase the focus on performance of the infrastructure network.
At the same time, two recent reports on spatial inequalities from the IFS Deaton Review on inequality and from the Economy 2030 Inquiry – both of which I helped author – continue to place an emphasis on the central importance of education and skills for increasing local growth and narrowing disparities.
Comments from our new Prime Minister suggest a renewed focus on skills, further raising concerns among those on the infrastructure ‘side’. However, while it’s tempting to characterise this as a debate between two different sides, as I have just done, the decision facing central and local governments is more complicated than having to pick whether to focus on infrastructure or skills – there is no ‘silver bullet’.
Infrastructure and local economic growth
On some measures, London receives a disproportionate share of infrastructure investment. Many argue that this is a key driver of spatial disparities in economic performance, and that investment in infrastructure – particularly in transport – should be the number one priority for local and central governments looking to drive local growth.
Infrastructure matters for economic performance. London, and other big cities, could not function without their underlying infrastructure. Commuter towns rely on transport infrastructure to connect residents to jobs elsewhere. Evidence suggests that the UK has under-invested in transport infrastructure and that increased investment would improve national economic performance.
However, whether transport investment can improve local economic performance in places that need levelling up depends on the choice of investment, because the economic returns to infrastructure vary a lot across places. Even if money is spent well – for example, to tackle congestion in cities outside London and the South-East or to tackle local bottle necks – the direct impact on spatial inequalities in income will be small. This is because differences in infrastructure play a relatively small role in explaining disparities in income, compared to differences in education and skills. Analysis for the IFS Deaton Review suggests that the spatial concentration of more educated and higher skilled workers, together with the wage and employment premiums they enjoy, explains between 60 per cent and 90 per cent of area-level disparities in wages in the UK.
Infrastructure, skills, and local economic growth
While the direct impact on spatial disparities of infrastructure may be small, the indirect effects – via changes in the composition of the workforce or of local businesses – can be large, especially if infrastructure makes places attractive to more educated and higher skilled workers, or to the firms that employ them. At smaller spatial scales, we often see such effects – for example when a new railway station makes an area attractive for households in neighbouring areas looking for cheaper housing or easier commutes.
At larger spatial scales this is much less likely. The only way for infrastructure to have a big effect on spatial disparities in economic performance is if it improves local educational outcomes or leads to the relocation of large numbers of skilled workers and the firms that employ them, away from London and the South-East to other parts of the UK.
From the perspective of an individual business owner this focus on the indirect effect of transport through skills can sometimes seem puzzling. If infrastructure was better, they argue, they would be able to recruit the people they need from across a broader area. But this also applies to all other business that are looking for skilled labour. If you connect areas with similar skill levels – say, Manchester and Leeds – then the demand for skills will tend to rise at the same rate as the supply of skills, meaning this won’t make much difference to overall skill shortages. It is true that a larger labour pool should allow for the better matching of specific skills to specific workplaces, but the impact of this will be relatively small.
Empirical estimates for the UK suggest that on average the indirect effect of infrastructure on spatial disparities in income, working through changing the composition of the labour force, is about six times as large as the direct effect that comes through things like better matching. A good example is provided by some research I did for the Northern Way. Our largest estimate for a 20-minute reduction in train journey times between Manchester and Leeds has wages increasing by between 1.06 per cent and 2.7% per cent. However, nearly all these wage effects come through the changing composition of the workforce. The effects for any given individual who does not increase their education or skill levels are small at somewhere between 0.20 and 0.50 per cent.
The silver bullet?
This discussion leaves open the possibility that infrastructure could still be the silver bullet if it is the key to improving educational outcomes or attracting new higher skilled and educated workers to underperforming areas. As discussed in our Deaton Review chapter there are two main factors driving spatial disparities in education and skills:
- Differences in educational outcomes for people growing up in different areas.
- Differing patterns of mobility – in terms of both tendencies to move and choice of where to live – for people with more education and higher skills.
You can construct arguments where transport infrastructure affects educational outcomes – for example by changing the local returns to education and the incentives for children to acquire education – but it’s a push. The available evidence suggests these effects are likely small relative to the effect of intergenerational transmission (that is, the fact that highly educated parents are more likely to have highly educated children).
Transport clearly affects differing patterns of mobility, but other factors, including amenities and the costs of living, matter too. In short, transport infrastructure has a role to play but it is no silver bullet.
