San Francisco versus Birmingham Part II

Just to clarify, in my post earlier this week I was talking about Birmingham, England not Birmingham, Alabama. It appears that muddling the two is an easy mistake to make.

I gave a talk at the San Francisco Federal Reserve Bank yesterday and something that came up concerned the impact of land use controls on fluctuations in prices. Conventional wisdom is that more restrictive supply leads to greater price fluctuations. So, if you take San Fran (with very restrictive land use) and compare it to Columbus OH (with plenty of open land) you see price fluctuations around trend that are far more marked for San Fran.

But some of the highest price falls in the US have been seen in places (like Las Vegas) that have seen huge building booms. Presumably this is because housing supply is inelastic once houses are built (unless you knock them down again) but I am not totally clear why this should have lead to larger house price falls in those markets (unless speculative demand played more of a role there than elsewhere and that component of demand has now collapsed).

[Thanks to Christian Hilber for the figures]

San Francisco versus Birmingham

I am in San Francisco for work. I am reminded of research by my colleagues Paul Cheshire and Christian Hilber that reports that commercial rents in Birmingham are twice those in San Francisco (mostly because Birmingham's planning restrictions are more restrictive than San Francisco - itself a highly regulated city for the US).

Rather than risk offending anyone, let me simply finish with a couple of pictures for you to compare and contrast for yourselves.

Birmingham:

San Fran:

Manchester-Leeds Linkages

We launched our research for the NW on links between Manchester and Leeds yesterday.

The research finds that closer integration between Manchester and Leeds could bring economic benefits to much of the North. However, any impact would depend on changes in the industrial structure and in the composition and skills levels of the population. My worry is that such changes may be mostly compositional (i.e higher educated workers moving in to the two cities) in which case existing residents who don't up-skill could actually be worse off (because rising average incomes might drive up e.g. house prices).

That worry brings us back, yet again, to the question of whether people or places are the most crucial objective for policy. For some policy makers raising average incomes in the North is a suitable objective even if that just comes from moving already better paid workers around. As I have said before, I think the main focus should be on directly helping the disadvantaged.

But if you are someone who believes in helping places per se, focusing on relatively successful places within less successful regions at least has the advantage that you are working more with market forces than against it. Manchester and Leeds certainly meet those criteria so focusing interventions on improving the performance of those two economies may not be such a bad thing, conditional on existing government objectives.

Kickstart

Having got terribly confused the last time housing numbers were discussed it was nice to see so much data being presented by CLG on projects funded through Kickstart.

Once again, however, the amounts of expenditure involved for relatively modest amounts of house building convince me that government intervention of this kind is not a long run fix for the underlying problems of high house prices and poor supply responses.

New Deal for Communities?

New Deal for Communities (NDC) is a 10 year Area Based Initiative (ABI) spending £400 per household per year in 39 of the most deprived areas of the country. I think that new evidence on its impact to date is disappointing.

According to the summary: "NDC areas are experiencing positive change, some of which is over and above that occurring in the comparator areas." However, "once individual-level socio-demographic factors and also starting position are taken into account, those in NDC areas saw significantly greater improvement than those in comparator areas for only one indicator: thinking the area has improved in the last two years." What the summary doesn't tell you but the full report does (see p.91) is that (controlling for socio-demographic factors) individuals in NDC areas did worse on somewhere between 2/15 and 7/15 indicators (depending on the time period).

What I take from this is the following: Based on the best evidence that we have available a reasonably well funded ABI has not, on average, improved individual outcomes in targeted areas.

The report gives supporters of ABI wriggle room. Perhaps there are ABI's going on in the comparison areas that are just as successful (although it seems unlikely that they could be as costly as NDC so we should be doing whatever they are doing instead of NDC). Alternatively "There can be no assumption that 'success' is best measured in relation to what happens to individuals as opposed to what happens to these areas over time". I find that argument simply baffling (I thought 'no place left behind' was means to an end - i.e. helping poor people - not an end in itself). Next, 'it's still early days'. Fair enough, although the research suggests that the largest gains came first. Finally, ABI's might be good delivery vehicles. I think this last one has legs, but raises questions about whether the holistic nature of NDC fits with the need to target spatial concentrations of particular problems.

As the report says "assessing the success of neighbourhood level interventions is contested territory" but my feeling is that for the moment the evidence emerging from NDC is more negative than positive.

The educational divide

Catching up with news after a long period of unexpected leave. The first thing I came across was a story about the educational divide: "Britain is becoming increasingly divided along educational lines with degree blackspots springing up in the poorest areas of the country as graduates flock to the capital" according to the Guardian reporting on research from the lecturer's union UCU.

The analysis involves looking at the proportion of working age population with degrees in every parliamentary constituency. What it tells us is that graduates choose to live in some areas (Richmond Park 63% graduates) and not others (Hodge Hill, Birmingham 9.9%). Whatever the newspapers say, it doesn't tell us anything about widening participation (that is about flows into education, not location decisions afterwards) and on whether this pattern is a good or bad thing for people who do not graduate.

High Speed Rail

Last week Network Rail made its business case for a new high speed line. Greengauge 21 will follow shortly, while HS2 will report next year.

Here is what I think we will know after these exercises. The direct user benefits (i.e. the benefits to people making journeys) are potentially quite large. Unfortunately, so are the costs. Both costs and benefits are highly uncertain. The wider benefits - e.g. 'regenerating the north' - are even more uncertain. I suspect the only thing we can say with any certainty is that they are likely to be overstated (there's a long history of this in terms of 'transformative' transport schemes).

Two things where we have a bit more certainty. First, the environmental impacts are not large and could well be negative. Most of the journeys on high speed rail will be additional (or transfers from other lines). Extra/faster journeys generally require more energy not less. You get some savings on car trips and air trips (offset if you use freed up slots for other flights; or if you make getting to Heathrow easier and so generate additional numbers on other routes). The overall impact will depend on what you assume about how electricity is generated. Expect to see generous assumptions using average, not marginal, carbon figures. The second certainty is that any new route will not be commercially viable and will need large government subsidies.

A final thought. Congestion charging has fairly certain costs that are outweighed by clear and certain benefits (the social costs of congestion are very high; and it also helps with carbon). Road pricing has more uncertainty over costs but the benefits are also fairly clear. Yet the strong political support from all three parties is for high speed rail not road or congestion charging. Go figure ...