Demand forecasts for cycling projects ‘are stabs in the dark’

Appraisals of cycling projects would be better off discussing the investment’s potential to increase cycling, rather than relying on inherently unreliable demand forecasts, according to a new report.    

Justifying investments in cycling infrastructure is the third report on transport decision-making and appraisal by Yaron Hollander, a former modelling manager at Transport for London who left last year to set up his own consultancy, CTThink!. His first report examined modelling misuses (LTT 13 Nov 15) and the second looked at transparency in decision-making (LTT 08 Jan).  

Hollander’s new report presents ten lessons for cycle project appraisal, drawing on his own experiences as a practitioner and  a long-running dialogue with planners working on cycling projects in the UK and Ireland. 

He says appraisal practice for cycling projects is “still quite young” and knowledge is not widely shared because scheme funding is often awarded through competitive grants that see councils bidding against each other.

Professionals involved in developing cycling schemes often find the work “exciting and engaging”, he says. But this creates a danger of the “visionary narrative coming at the expense of providing reliable evidence on the need for investment and the likely impacts”. 

“The faith that planners have in improving infrastructure for cyclists is a great thing, but there is a gap between the optimism that radiates from our strategic documents and the real­-world difficulty of getting people on their bikes.

“The people that read your business case know this, and they look for signs that indicate to what extent you’ve applied realistic adjustments which take this gap into account. It’s in your interest to show that you’ve done this yourself, because otherwise they’ll apply their own adjustments.”

Hollander says there is a genuine difficulty in forecasting the demand impacts of cycling schemes because of factors such as the partial understanding of trends and what makes people cycle; a lack of data; statistical issues resulting from the low proportion of trips made by bike; and the seasonality of  cycling demand.

“Many recent business cases for cycling programmes have followed a practice which I wouldn’t recommend. They give a very brief explanation of the assumptions they made when estimating cycling demand, and then they dedicate dozens of pages to calculations and conclusions, which are all based on these assumptions. 

“The demand assumptions are often difficult to find when you browse through the business case document, but they are implicit in every figure showing health, ambience and other impacts. If you do find the description of how demand was estimated, and you’re not sure it’s robust, then the entire logic of the economic case might collapse.”

The best cycling scheme appraisals, he says, adopt a ‘fail safe’ strategy. “They remain convincing even if the reader doesn’t accept the demand assumptions. An effective way of applying a ‘fail safe’ approach is by analysing the potential demand for cycling in addition to?the analysis of the expected demand. A high potential means that the investment has good chances of success even if the forecasts are wrong.”

The ‘flagship’ study to adopt this approach was TfL’s Analysis of cycling potential in 2010, he says. It also lies at the heart of the DfT’s National Propensity to Cycle tool, which is in development (a first version is available at http://geo8.webarch.net/master/)

Hollander says assessing an area’s cycling potential doesn’t require transport modelling at all. “Most towns and cities have all the inputs they need for this analysis.”

He describes a number of inappropriate uses of transport models for cycle scheme appraisal, adding: “Over the last few years I spoke to many local government officers who worked on cycling projects, and were under pressure to squeeze from their models some insight that the models simply weren’t suitable for.” 

Hollander says that even if his ten lessons are followed, he retains “a degree of cynicism about the whole process”. “I’d recommend that you do the same, because the assumptions we make aren’t true. If we really have to limit the benefits of each project or programme only to those that result directly from this specific investment [which he suggests], then cycling infrastructure doesn’t have much value.” 

Many of the investments made to cycling networks throughout the UK during the last decade “haven’t shown yet a clear impact”.  “In many places where cycling levels have grown, it’s not clearly attributable to a specific investment.

“The main benefit from most of our cycling work is meant to be the way it contributes to a major cultural change. The millions invested in cycling are meant to gradually add up into a critical mass which will, at an unknown point in the future, turn England into a cycling nation, with all the huge benefits this would finally bring. 

“We don’t know how feasible it will be in practice and when it will occur. And if it does happen, then the benefits will probably take a different form to what we currently know how to quantify.”