This section is updated monthly.

Land-use uncertainty and transport model horizons

Professor Alan Wenban-Smith, October 2009

Professor Alan Wenban-Smith, a highly-regarded independent consultant of Urban & Regional Policy, appeared for and presented written evidence on behalf of the North West Transport Roundtable at the Mersey Gateway public inquiry (see 'documents' page of this website). He used that case as an example of some of the short-comings in the way transport schemes are appraised in the lead letter he had published in Local Transport Today (LTT 527, 28 August-10 September 2009). We reproduce his letter, which begins by referring to previous correspondence in 'Local Transport Today' here.

As a land-use planner with a long-standing interest in transport planning, my eye was caught by the article describing the recent Atkins/Simmonds/Katalysis/Tyms study for DfT on the continuing differences of perspective between land-use and transport planners, and by David Metz's letter on the same subject (LTT 526, 14 August 2009).

There is a tendency amongst transport planners to criticise their land-use colleagues for their less quantitative approach to predicting future distributions of development (and by implication to blame them for the poor performance of transport models in predicting travel demand more than a few years ahead). The study quoted seems to follow this well-worn track. But as David Metz points out, this is too polarised a view: we are all (planners and engineers) dealing with a highly uncertain world, and the demand for appraisals to be supported by 60 year transport forecasts simply does not engage with this reality. For example, at the recent Mersey Gateway Bridge Inquiry it emerged that 98% of the net economic benefit arose in years 30-60 - after the end of the transport model, and far beyond the range of any crystal ball!

This is not uniquely a transport planners' problem: land-use planners are being driven up a similar cul-de-sac by the Government's insistence on trying to implement 20-year housing forecasts which bear no relationship to real-world events. The problem for both land-use and transport planners is not with the models and forecasts as such, but with the planning systems they are plugged into. DfT and CLG are converging on a 'predict and provide' model of planning, in which infrastructure investment and planning permissions are driven by year 1 forecasts, which can only be revised by starting again (- perhaps more accurately described as 'fire and forget'?). As I noted in a Viewpoint article back in 2003 "... while engineers are from Mars and planners from Venus, policy comes from Uranus".

The problems of transport forecasting do not primarily arise from lack of precision in specifying planning scenarios: planned new development is seldom (realistically) on a scale which makes much difference to transport demand. Existing buildings change hands (through 'churn') at ten times the rate of new construction, with thus much greater potential for wider economic, social and transport impacts. 'Pure' transport models cannot deal with the changes in patterns of activity that transport change itself causes. Land-use/Transport Integration (LUTI) models, which do attempt this, are still some distance from precision in representing the highly complex economic and social changes that flow from transport. There is thus a crisis in transport appraisal, which neither better 'planning data' nor the WebTAG 'fixes' introduced as a result of TASTS can resolve.

As David Metz rightly says, we all need therefore to be less appraisal-driven and more conscious of the strategic outcomes that we are seeking. In the case of High Speed Rail this means thinking about how better connectivity, allied to complementary action in other fields, affects regional disparities and national productivity (my recent report for Greengauge* deals with this issue). I am not saying that value for money does not matter, but I am saying that it is not going to be the output of a transport model, regardless of land-use input, any time soon.