Regional policy requires rethink as myth of London’s productivity debunked
Britain’s regional divide is well known and well documented. The richer bits of the country tend to be clustered below a line drawn from the Wash to the Severn estuary, while London is so different from everywhere else that it may as well be its own city-state.
There is also another divide: between the big cities and the smaller towns. The north-west may be less prosperous than the south-east but on average people are better off in Manchester than they are in Blackpool.
Economists have a posh way of describing the reason for this big-small disparity: agglomeration effects. This means that firms are more efficient if they are clustered together, with cities better locations for businesses than towns.
Regional policy, to the limited extent that it currently exists, has involved encouraging regional hubs in the hope that they will pull up the struggling towns on their peripheries. Current thinking is that there is not a lot of point trying to encourage firms to relocate to places such as Blackpool because they are less efficient than the big conurbations.
This notion, according to a new piece of research from Steve Fothergill and Tina Beatty of Sheffield Hallam University, is actually completely wrong. Once the official data for regional economic performance is broken down there is virtually no difference in economic efficiency between workers in one region and another, and workers in cities are no more efficient than workers in towns.
This is not the picture provided by the official method of calculating productivity – gross value added. GVA per capita is the result of taking the value of the output of a country, region, city or town, subtracting the cost of the inputs that went producing it and then dividing by the number of people.
There are vast differences between the various parts of the UK on this measure. Against a nationwide average of 100, London has a GVA per capita of 179, almost three times as high as southern Scotland (61), west Wales and the valleys (63), and Cornwall and the Isles of Scilly (65).
Intuitively, such wide disparities seem a bit weird. In any one part of the country, the range of jobs is generally pretty uniform. London has hairdressers and so does southern Scotland. There are supermarkets, hospitals and schools in the Welsh valleys just as there in the capital. There is a demand for solicitors in Bodmin just as there is in Barking.
Is it really possible that a doctor in London is three times as productive as a doctor in Merthyr Tydfil or that it takes three times as long for the Cornish lawyer to do the conveyancing on a house as it does for the one in Camden? Fothergill and Beatty’s research suggests there really is no real gap to speak of.
They come to this conclusion by gradually stripping away various elements of GVA per capita. An adjustment is made to divide GVA by the number of people of working age (16-64) rather than the population as a whole. Account is taken of the fact that employment rates among the working-age population are higher in some parts of the country than others because of the number of students or those on disability benefits.
Commuting makes a difference because the number of jobs in each area is not necessarily the same as the number of residents in work. The mix of industries also matters: some parts of the country have a high concentration of high productivity jobs – such as banking, finance and business services.
The split of head office and routine jobs within industries is not even: some regions tend to have a bigger share of white-collar and executive roles. GVA comparisons are also skewed by the number of hours worked in different parts of the country.
Once all that has been done, the authors say, you have a fairer measure of basic economic efficiency. And using this methodology, the big regional discrepancies almost entirely disappear. London has a smaller over-65 population than elsewhere, which brings its GVA per capita down to 165% of the national average.
Adjusting for the large number of commuters brings it down to 141%. London’s preponderance of jobs in banking and finance trims it further to 129%, adjusting for the capital’s high concentration of top jobs within each industry reduces it to 119%, while taking account of the tendency for average hours in London to be above the national average leaves a final figure of 111%.
Even that, the authors say, is probably an overestimate because of the difficulties of properly accounting for the concentration of the most senior and highest-paid jobs in London.
Adjusting in the same way for west Wales and the valleys (which have a high proportion of over-65s, large numbers out of employment, lots of people commuting to Cardiff and Newport, more jobs in low-productivity sectors and fewer people in head-office-type roles) and the GVA per capita figure rises from 63% of the national average to 94%. Fothergill and Beatty also show that once similar adjustments are made for Britain’s 11 main regional cities their GVA per capita comes down from 109% to 93% of the UK average.
This study should be read in the highest levels of government. While it does not dispute that some parts of the country are richer than others, it does challenge agglomeration theory and the notion that a more active regional policy would require sacrificing efficiency for equity.
Encouraging businesses to set up in the Welsh valleys would not mean that the UK as a whole would suffer. On the contrary, if underutilised resources could be brought into use through investment in infrastructure, research and development and skills, the economy would benefit.
Source: Read Full Article