Thursday, 26 January 2012
For some time now it has seemed to me that there is a fundamental, but unresolved, tension concerning the economic objectives for UK cities. On the one hand, government wants to maximise their economic potential. On the other government wants rebalancing both in terms of a shift away from financial services and geographically (from south to north). Of course, some will argue that these policy objectives are not in conflict. Most urban economists would disagree. If we do need to choose (at least in the sense of prioritising one over the other) what should we do? Overall, at least for the UK, I think the evidence points towards prioritising growth in our more successful cities even if this leads to more uneven spatial development. Let me explain why.
Disparities across local areas in Britain are pronounced and very persistent but much of these disparities are driven by ‘people’ rather than ‘place’
Let's start with the underlying drivers of spatial disparities. In recent SERC research we assess the extent of and persistence in wage disparities across labour market areas in Britain. We examine to what extent these area differences arise because of differences in the characteristics of people who live in different places – ‘sorting’ – versus different outcomes for the same types of people living in different places – ‘area effects’. We also consider the extent to which these differences across areas contribute to overall individual wage disparities. Our research finds that between 1998 and 2008 there were few changes in area disparities, despite many policy interventions. It also turns out that who you are is much more important than where you live in determining earnings (and other outcomes). Area effects only play a small role in the overall wage dispersion.
Earnings disparities are uninformative about differences in people’s overall wellbeing unless we take account of differences in the cost of living and the availability of amenities
Such disparities between different cities and different labour markets concern policymakers because they seem to imply differences in standards of living and economic welfare. In fact, however, this is not the case. Our research shows that across Britain increased living costs (particularly of housing) tend to offset completely increased wages for the average household. In terms of real earnings (conditional on skill) it doesn't make that much difference where you live in Britain.
In short, looking at area differences greatly exaggerates the importance of place in determining individual wellbeing. Of course, the evidence says that place does play some role in determining wages so there is a question about whether or not we should try to address these area effects. If, say, Hull's economy is doing relatively badly because of the combination of lower skilled workers and bad area effects, shouldn't we try to address both?
It is very hard for policy to change area effects
Unfortunately, evaluation of specific policies suggests that it is very hard to change area effects. Details on that will have to wait for another day, but for the purposes of this post let me simply point to the fact that our research suggests that a decade of fairly significant intervention left underlying area effects essentially unchanged. It's asking a lot for the current government to achieve any bigger impact with much less money.
It is very easy for policy to drive up the cost of living
In contrast to the difficulties in reducing area effects it is very easy for government to do things that drive up the cost of living in our more successful places. The most obvious way in which we do this is through constraints imposed by the land use planning system (SERC's Max Nathan explains the evidence here) but transport and other policies all play a part. If you can't do much to tackle underlying area effects, then an alternative is to allow people to move to areas where they will do better. Of course, you might think that this is a second best option. But the evidence suggests that it is likely to be the first best feasible option.
The need to focus spending
But what should government do in terms of the limited amount of expenditure that it is able/willing to make? Given the difficulties in addressing area effects, as discussed above, there's a strong case for trying to build on success. Of course, success can be relative so this might call for focusing investment in, say, Birmingham, Leeds, London and Manchester. Again this is a matter of prioritising, rather than a call for all expenditure to go to a select set of places.
More uneven spatial development: good economics, bad politics?
Investing in more successful cities to either enhance the economy or reduce cost of living clearly exacerbates uneven spatial development. But I have tried to argue that this may make for good economic policy in a world where who you are matters more than where you are and the government can't do much to offset the market forces that make some places perform worse than others. Of course, adopting such a course, and prioritising growth over rebalancing makes for very difficult politics for constituency based politicians.
If you are interested in hearing more on these issues, I'll be considering them and some concrete policy issues in my LSE London seminar. The seminar takes place this Monday at LSE on 'What should urban economic policy do? Lessons for London'. Details on time and location available here.
Wednesday, 25 January 2012
While broadly supportive of the government's localism efforts I am sceptical about the claims that this will drive economic growth. I outlined the details of my arguments last June. In short, I don't think there is much evidence that mayors increase economic growth. That said, there is not much evidence against this proposition either and I am in favour of experimentation when it comes to government policy in this area.
For some of these mayors, the success of negotiations on city deals will be crucial in determining their power. For those of you that are interested in this particular aspect, SERC's Max Nathan discusses some of the central issue surrounding city deals here.
Tuesday, 24 January 2012
As I have discussed before, much of this can be understood by looking at the underlying structure of places - particularly in terms of skills. Centre for Cities like to argue that this makes for a more complex picture than a north-south divide. I see the point, and there is some variation within regions (e.g. York does relatively well) but I still think the broad north-south divide is pretty clear. You can judge for yourself by taking a look at the very useful maps provided in the report.
The marked difference in city's youth unemployment rates identified in the report are a little more of a puzzle - although it's not clear to me the extent to which these persist if one were to control for underlying differences in skill levels. If city differences are not just structural, Centre for Cities argues that this makes the case for local variation in labour market policy. While supportive of much of the localism agenda, I have to say that I remain to be convinced on this particular aspect of it. It's not clear to me why individual level interventions are not the appropriate level at which to conduct most labour market policy. Centre for Cities suggest, for example, that in some places language might be a specific barrier. But presumably this is only because of a concentration of people with language problems. Individual level policies address that (and have the benefit of still addressing language issues for people who have those difficulties but in areas where others don't). I guess there might be some scale aspects on the provision side resulting from concentration of workers with specific characteristics, but it's not clear the extent to which that benefit of localising policy would offset the other disadvantages. As I say, I remain to be convinced.
There's plenty more in there to digest, but that will have to wait for a later date.
Monday, 23 January 2012
Posted by Dr Max Nathan, SERC and LSE Cities
Yesterday’s Observer ran a great piece on megacities and city growth in China and India. Both countries are urbanising rapidly: China has just become majority-urban, and India’s population is predicted to be 40% urban by 2030. The authors also make great play of two countries’ urbanising trajectories: top-down and massively resourced in China, slower and more organic / chaotic for India.
Despite the rather misleading headline, the piece also correctly points out that megacities will be the exception, not the norm. This is a crucial point for policymakers. Rather than a world of super-size cities, large and regular-size places will be far more common.
Some numbers. The count of megacities (with 10m people or more) is rising – from two in 1950, three in 1975 to 19 in 2007. By 2025, the UN predicts there’ll be 27. But the number of ‘large cities’ – five to 10m people – is already bigger, and growing faster. In 2007 there were 30: the UN suggests there’ll be at least 48 by 2025. More importantly, half the world’s urban population live in much smaller cities, of around 500,000 people. These may be the most common of all. So in fifteen years’ time we’ll see far more Liverpools (around 400,000 people) and Londons (8m people) than Tokyos (26m people).
Paradoxically, the very biggest urban settlements are now hard to recognise as cities at all. Across the world cities are merging into mega-regions: notably China’s Pearl River Delta, the US Eastern seaboard, even the Greater South East.
Some of these numbers are difficult to take in. An estimated 120m people live in the Pearl River Delta, the largest urban zone on the planet – China is now planning to merge nine cities in the Delta to create a single sprawl of 42m people. The Tokyo-Nagoya-Osaka-Kyoto-Kobe region may comprise 60m people by 2015, almost the entire population of the UK.
All this may suggest that urbanisation is accelerating. In fact the opposite is true. Globally, cities grew fastest in the 1950s and early 60s: growth rates have been slowing ever since, from 4.1 percent to 2.5 percent today, and a predicted 1.8 percent by 2030. Developing countries are also on the same downward trend.
Urbanisation runs in parallel with economic development, and so as developing countries industrialise, their urban systems tend towards steady state. Of course there is a lot of city by city variation. For example, the UN predicts Dhaka will keep growing – from 15.9m in 2007 to 22.8m in 2025. But Lagos, which has grown from less than half a million people in 1950 to over 13m in 2007, is predicted to reach just 16m in the next fifteen years.
Like many researchers, the Observer piece spends some time on the growth of urban slums. Again, the picture is complex. Over the past decade the share of urban slum dwellers has actually fallen from 39 to 32 percent, due to economic growth and policy interventions. But as people are flowing into cities faster than infrastructure can keep up, the absolute number of people in informal settlements is growing, and will keep growing. Ed Glaeser describes slum neighbourhoods as ‘private energy, public failure’: the development challenges of poor public health, chaotic infrastructure and urbanised poverty remain considerable.
Finally, we need to factor in the geography of climate change. Many megacities are coastal, and will be threatened by rising sea levels. Many will also be increasingly water-stressed in the years to come.
In his excellent book The New North, Laurence Smith explores the economic rise of the NORCS – cooler, resource-rich regions stretching across Canada, Scandinavia and parts of the US, Russia and China. He predicts new ‘hydrocarbon cities’ appearing across Canada and Russia, and new mega-regions like Cascadia – spanning Portland, Seattle, Vancouver and parts of NorCal.
Megacities are a great symbol of the global urban shift. But our urban future is going to be much richer and more complex than this. The sooner we recognise that, the better we’ll be able to plan for it.
A version of the piece first appeared on the squareglasses blog.
Thursday, 19 January 2012
On fairness, as I've explained before, I prefer to think of equal reward for equal work, rather than equal pay. The difference is that the former takes in to account local differences in the cost of living.
Chris Giles suggests that the main argument in favour of reforming public sector pay concerns the second of these issues - the impact on public sector services. As he says: "There is little reason why Britain should tolerate that a nurse, teacher or police officer should live towards the top of the local income distribution in the north-east or south-west of England, but be stuck well down the pay scale in London and the south-east." In the low real wage areas, equal pay translates in to low reward with subsequent consequences for public services (at the extreme, this difference kills people as explained by CEP colleagues investigating the impact of national pay in the NHS).
The FT is more dismissive of the impact on the wider spatial economy: "The simple argument deployed by Mr Osborne, that high public sector pay in poorer regions crowds out private sector enterprise, is also suspect." This is a bold claim given the lack of evidence in this area. The crucial issue here is whether the demand stimulus provided by higher public sector pay offsets the supply distortion in the labour market. Very little is known about either of these effects.
Preliminary finding from ongoing research suggests that there is some evidence of distortion in the labour market - at least from public sector jobs (we haven't yet looked at pay). Specifically, we seem to be finding that areas receiving more public sector jobs see no obvious effect on overall total employment, but do see lower manufacturing employment. This would be consistent with a story where public sector jobs crowd out 'tradeable' manufacturing while providing a demand stimulus to local services. A little more work to do till we are more confident on this finding, but suggestive of the fact that it might be a little early to completely rule out such labour market distortions.
Tuesday, 17 January 2012
The BBC describes the new voluntary scheme as working as follows:"Under the scheme, the council will arrange for elderly people to move into rented accommodation, and then take responsibility for maintaining and letting their property at an affordable rate. The rental income is then passed back to them [...]" Providing the scheme is truly voluntary, then lowering transaction costs in this way may be a relatively cheap way to increase the supply of space (a pilot in Redbridge suggests that this is indeed the case).
But, and it's a big but, this is likely only to lead to a small increase in effective supply. Why? Because it turns out that not that many people appear to be 'trapped' by moving costs - according to the Telegraph "Government analysis of the Redbridge project suggested that 200 people in the borough were considering moving, but felt that they could not afford to." And this presumably represents the 'stock' of people who feel trapped rather than the annual flow. If so, the scheme may be helpful, but doesn't offer a radical solution to the housing crisis. My preferred radical solution remains to build more housing.
Wednesday, 11 January 2012
Conceptually, it seems quite possible that there is something in this for the simple reason that most financial crashes follow booms and booms are generally associated with over investment in capital assets - factories, inventories, housing, so why not building height? That said, if someone wanted to seriously assess this claim, they could do worse than start with this wikipedia list of the worlds 250 tallest buildings. The correlation with financial crashes was far less obvious to me as I looked down that list.
Even if there tall building booms do precede financial crashes, it's important to remember that serious research tells us they play an important economic role - in helping make the best use of a scarce resource (land) and in generating agglomeration economies which help make firms more productive. These longer term benefits are likely to outweigh any short run effects so I remain convinced that UK cities (and London especially) could do with more tall buildings not less.
Monday, 9 January 2012
- Extra tunneling costs, the finance black hole and the problems with predicting demand growth from 6th December 2011.
- Why it's OK to support some large infrastructure projects but not others (following criticism of the ASI's position as purely ideological)
- HS2 and NIMBYism (some findings from related academic research)
- HS2 and the North-South divide
- Benefits and costs both highly uncertain.
- Environmental benefits limited.
- People are in favour of high speed rail, providing they don't have to pay for it.
- The opportunity costs of the project are large (i.e. we could better spend the money on other things)
[Disclosure: I sit on the HS2 Analytical Challenge Panel]