2050 Road Traffic Forecasts: Starker Choices

The recent publication of the Department for Transport’s National Road Traffic Forecasts to 2050 throws up some major challenges to the field of transport planning. Whilst this is a set of forecasts for England and Wales it provides a set of questions that go well beyond that scale. Some initial reflections, drawing on the work of the Commission on Travel Demand are set out below, although there will no doubt be more to debate.

Greater Transparency

The first point to make is that it is only possible to have a debate about the strength, purpose and future of road traffic forecasting because of the high level of transparency which the Department for Transport has brought to this. This practice is a further development of the 2015 report and includes much greater reflection on the performance of previous forecasts, variations across areas and road types.  For example, the report notes that three of the scenarios from the 2015 forecasts consistently overpredicted road traffic levels between 2010 and 2017 as shown below.


More generally, it is worth observing that none of the sets of assumptions appeared to track the outturn particularly well.

There is also a high degree of transparency as to what has been done to change the model, where to find more details of this, what the assumptions in each of the scenarios tested in RTF2018 are and how they work in different regions. This includes a broader array of tests than those in RTF2015, most notably high and low migration scenarios, a scenario which includes 100% ZEV sales by 2040 and another scenario which includes a continuation of the decline in trip rates.

Starker Choices

The purpose of RTF is clearly stated as not to be indicative of desirable futures or likely futures. It is here then that the range of outcomes from RTF2018 defines the debates that should be at the forefront of transport policy today. For me these include:

  • How are any scenarios being run which are not consistent with our Climate Change Obligations? Only one scenario has decarbonisation of the car fleet by 2050 with a reduction in CO2 of 76%. However, the other six have just 25% of miles travelled by ZEVs by 2050. This is not even as ambitious as the goals set out in Road to Zero. The smallest reduction implied by these scenarios is a 17% reduction in CO2 which simply cannot sit alongside the reduction pathways the CCC has set out. As such, they should not, in my view, be being conisdered as part of a meaningful future scenario set.
  • Do we want to allow motoring costs to decline and how should we manage the transition to electric vehicles? In the absence of any other assumption on changing electricity prices, the scenario with decarbonisation of the fleet has the largest increases in road traffic by 2050 – 51% – a massive 16% higher than the reference case. If this is not remedied then, looking at the other metrics on congestion and journey times (increases of 75% in lost time), the pressure would be for a major roads programme to reduce these impacts. This would amount to massive infrastructure investment and associated environmental damage in order to compensate for additional demand which results from the reduced cost of driving. Where would the tax take come from in order to fund this kind of programme given that fuel duty would largely disappear from general taxation? This has been much trailled as an issue – but if RTF is to inform policy, rather than just roads investment decisions, then it is difficult to see how this can be ignored.
  • Should we create lower demand futures? One of the intriguing quotes in the report is “research was unable to provide an explanation for the reduction in average trip rates over time leading to significant uncertainty about future trajectories”. One scenario runs with the assumption that the past two decades worth of trip reductions continue. The other six presume they stop. That seems like an unfairly stacked assumption (stabilisation deemed more likely or at least of greater interest to decision-makers) with no middle ground. However, the scenario with declining trip rates shows up some very important results. Only 8% of traffic in congested conditions compared with 7% today (and 16% in the highest growth scenario!). The Commission on Travel Demand was unequivocal in stating the need to understand the change in trip rates around which so many other assumptions hang. Is it possible to develop policies that support a decline in trip rates and deliver greater well-being? Given the difference the answer to this question makes to perceptions of future investment need it is now a priority to address.

Whilst it is not the place of RTF2018 to set out the policy response, it seems necessary for bodies like the Transport Select Committee to make sure that this follow through does happen.


Longer term relevance

If, as is contested in the Future of Mobility grand challenge, we can anticipate a radical reshaping of our transport networks then the very structure of the relationships between income, technology, activities and mobility will themselves be quite different. What then for the insights from RTF? The report states that RTF continues to produce “reliable and robust forecasts provided the most up to date evidence is used, the model is calibrated to a recent year and government QA procedures are followed”. It seems to me, more than ever, that whilst a well calibrated and up to date baseline is a good thing, the pursuit of accuracy in the short run does not help with dealing with uncertainty in the long run.

The RTF2018 makes an initial attempt to review the potential impacts of CAVs on the future travel demands on our network. It is systematic in setting out the many assumptions that need to be made and draws on the scant evidence that is available to do so. The answers, with all their health warnings, are in line with preliminary findings from elsewhere (which could be group think!) – CAVs could see traffic grow from between 5% and 55% depending on the amount of sharing that occurs. However, the policy framework matters to where we might end up here, just as it does for the pricing of transport with ZEVs.

At the end of this, we remain faced with a huge range of relevant uncertainty – from CAVs, from ZEVs and from disruptive mobility models as well as broader social change which we have only partial insights on. If we are serious about making decisions with a 2050 time frame in mind then we cannot take these changes as exogenous in the same way we might with GDP or population. The future has to be shaped and we require different policy frameworks and options to be set out so that there can be proper debate about what are looking like increasingly starker choices between future demand pathways.


Tags: , , , ,

2 responses to “2050 Road Traffic Forecasts: Starker Choices”

  1. Alison Irvine says :

    Good article as always – but it still leaves me with the question. What policy options and frameworks do we need and how do we collectively get there?

    • drgregmarsden says :

      Well, that is indeed a good question. When it comes to what the future pricing envelope for road travel – how much would politicians be comfortable in seeing car travel diverging from public transport? That would be one range of boundary conditions. Another might be what range of technological change is deemed necessary by different points to be consistent with climate change targets. Another might be how much of our travel needs to involve ‘active’ – some of the scenarions here seem to me to imply a continued shift away from walk and cycle which might not be consistent with health and well being goals. Without boxing politicians into picking a scenario I think there are some which fall outside the realm of ‘acceptable’ and therefore interesting/relevant to debate. There are only so many options and trade offs that can be presented in an exercise like this and I’d prefer to see the debate focussed around ones which are consistent with overarching policy statements. Whilst there is always a claim that the scenarios presented are policy neutral, I find it hard to agree – what is included and what is excluded is always political – even if the analytical team are rigorous and robust in their work (as I believe the DfT are).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: