National Road Traffic Projections 2022

On Monday 12 December 2022 the UK Department for Transport shared its latest look into the future for road traffic. Its last look had been in 2018 – a pre-pandemic world. Greg Marsden and I took a close look at the history of road traffic forecasts from the DfT - see our free to access paper (which is now rendered out of date by this latest publication!). There is also a LinkedIn article about the paper.

The latest forecasts – sorry, ‘projections’ – are based upon a set of seven ‘Common Analytical Scenarios’ that DfT has produced and which form part of its Uncertainty Toolkit. The Toolkit was first published last year and at the time I provided a LinkedIn article offering my reflections on that. I titled the article “Judgement, judgement, judgement - the key to accommodating uncertainty” – because judgement play a key part in looking to the future (alongside what we have learnt empirically from the past).

In fact, in this report there are not seven scenarios but eight – one is called the ‘Core’ scenario. We’ll come back to that.

The report is 88 pages long and if you really care about being able to make robust decisions in transport planning then I would say it’s well worth a read. I am also finishing writing this the day after Yasmin Fox (Head of Uncertainty and Scenarios, DfT) ran a free-to-attend webinar to introduce and explain the Common Analytical Scenarios. Over 140 people were there. This speaks well of DfT’s willingness to engage with the wider sector on the challenging business of examining the future.

The article below is not going to give you a full summary of the content of ‘NRTP2022’. I want to draw out some of the issues that caught my attention from reading it.

What’s in a word?

In scenario planning, some words can be source of confusion – either because a given word means different things to different people or because there are different words that could be synonyms or which could be subtly but importantly different in meaning. The word ‘scenario’ itself has different connotations, as does ‘vision’. Is a ‘plausible’ future different to a ‘possible’ future?

To this is seems we may now need to add the question over when is a forecast a ‘forecast’, and when is it a ‘projection’? In 2018, DfT published its ‘National Road Traffic Forecasts 2018’ report. The new report is the ‘National Road Traffic Projections 2022’. A subtle but perhaps important change I thought to myself. But Google tells me that the definition of ‘projection’ is an estimate or forecast of a future situation based on a study of present trends – gosh, that almost makes it sound like we are being presented here with a series of trend extrapolations. Yet we’re not, that’s for sure. The report itself in the Introduction says “our last road traffic projections were published in 2018”.

So, are far as I can tell, ‘projection’ is just being used as a synonym for ‘forecast’ – maybe for a bit of variety? I must check!

7+1 scenarios for everyone

DfT has worked hard on making available a set of scenarios that can be used in quantitative modelling but also in qualitative dialogue to allow the whole profession, to some degree, to be able to explore possible futures. That they are ‘Common’ is also an important signal that they constitute a form of ‘baseline expectation’ from DfT that they should be considered across all proposed schemes such that comparison across schemes has some consistency in the process of appraisal.

There are seven scenarios that look out to 2060, summarised briefly as follows:

• high economy (strong economic growth coupled with higher population growth – 76 million in 2060 for England and Wales);
• low economy (weaker economic growth coupled with lower population – 58 million in 2060);
• regional (people leave the wider South-East of England to move elsewhere in Great Britain);
• technology (50% of private cars on the roads are fully-autonomous (Level 5) by 2047 and near total penetration by 2060 – and are also electric);
• behavioural change (new ways of working, shopping and travelling embraced, with trends accelerated by COVID-19);
• vehicle-led decarbonisation (rapid take-up of electric vehicles which are cheaper to run than petrol or diesel equivalents with downward effects on public transport use);
• mode-balanced decarbonisation (rapid take-up of electric vehicles but without downwards effects on other modes’ use).

And then there is the eighth scenario, the Core scenario: this assumes ‘firm and funded’ government policy plays out in future and “[r]elationships between the key drivers of demand and road traffic are broadly assumed to continue in line with historical trends and evidence”.

Getting to the Core

The Foreword to the report is, as one would expect, full of carefully and well-chosen words. It begins by saying “It is critical to understand future road traffic demand and the uncertainty around it to provide the basis for informed and resilient transport policy” and recognises that “[t]here is considerable uncertainty around future travel demand”.

When the report later then says “[t]he Core Scenario is the starting point for most of the Common Analytical Scenarios” a warning signs goes off in one’s head – hang on, ‘core’ scenario sounds rather like the previous report’s use of ‘reference’ scenario and in turn potentially comparable to earlier reports’ use of ‘central projection’ or ‘most likely’ forecast. Are we back to ‘one scenario to rule them all’ with the other scenarios reflecting sensitivity testing and making up numbers?

No is the answer. One must embrace this report with a realisation that it is attempting to encourage the reader to recognise the need to accommodate the uncertainty felt about the future by considering “range of different plausible future scenarios”. The Core is a stepping off point reflective of ‘if we were to carry on as we have been…’ which allows possible alternative futures to be brought into play and compared against one another.

So, repeat after me, Core does NOT mean the most likely ‘do nothing else’ scenario against which to consider the effects of introducing a ‘do something’ scheme or new policy. If you’re tempted to head in that direction you are concealing the uncertainty and diminishing the support you’ll be able to provide for robust decision making. This is not the DfT’s intention – which is why it expects the Common Analytical Scenarios to all be used to some extent when considering the merits of candidate new schemes and measures.

Exploring the possibilities

The first step in accommodating uncertainty is acknowledging it. The next step is to represent the uncertainty in a pragmatic way. This takes the form of determining a set of possible alternative future scenarios. There are infinite possibilities here. The challenge which DfT has risen to is to offer up a finite set of possibilities that may be seen to reasonably reflect the nature and extent of uncertainty. But when it comes to deciding upon that set, we may each have different opinions. This is partly why DfT is keen on having Common Analytical Scenarios.

However, this does not stop us each having opinions regarding the scenarios on offer. I’ll be honest, at a personal level I struggle with the idea that we are going to see mass penetration of Level 5 self-driving cars that are privately owned (the Technology scenario). That’s partly because I don’t put a lot of stock in what Elon Musk has to say these days – after all, he said earlier this year that induced demand was the dumbest idea he had ever heard in his whole life (a slap in the face for nearly all transport planners).

Yet these projections invite us to open our minds to possibilities – that this scenario is one of the set reflects that others’ can readily entertain such a future.

How low can you go?

Some of you will be aware that several years ago I led a project for the New Zealand Ministry of Transport in which we developed a set of scenarios for the future of motorised road traffic – in three of the four scenarios road traffic went down, not up. This is unheard of in DfT’s history of road traffic forecasting. So, what of this new set of projections? “The projections illustrate that a wide range of traffic growth is possible in the long term, with the scenarios suggesting an 8% to 54% increase in distance driven between 2025 and 2060”. For comparison, the change in total traffic between 2015 and 2050 was 17 to 51% in the 2018 National Road Traffic Forecasts. It seems the forecast fan has tilted and extended downwards.

So for England and Wales, DfT is saying the lowest scenario is 8% growth. In this scenario, total car distance travelled levels off (while growth continues for light goods vehicles). Yet Wales has a publicly declared target of reducing car distance per person per year by 10% by 2030; and Scotland of reducing total car distance by 20% by 2030 compared to pre-COVID. In both cases this relates to a need to address the climate emergency. So, might we have hoped for a negative growth scenario within these latest projections? It’s important to note that they reflect published plans and funded policies – they are not scenarios of where we must or might like to get to but where we could get to given other uncertain drivers of changes.

This said, I might argue that if the technology scenario is not considered too extreme to include in the set, why was there not a scenario considered which involved low economy, low population, and behaviour change which could have gone lower still, and perhaps into negative growth? Indeed, none of the scenarios appear to consider whether failure at a global level to bring climate change under control could have seriously disruptive effects for travel in future (something Phil Goodwin has written about).

So yes, DfT could have gone lower than 8%. But not doing so doesn’t make DfT ‘wrong’, or me ‘right’. This is about exploring uncertainty in the interests of making better decisions. I would be at liberty – as you are - to consider further scenarios beyond the Common Analytical Scenarios, but DfT asks that at least the latter are considered.

But how low do we NEED to go?

I reproduce below, Figure 11 from the report. Look closely. There are eight scenarios – but can you see all of them? Regional and Core are the same in terms of CO2 emissions. Likewise, Vehicle-led Decarbonisation and Technology are the same (I think!).

But that’s an aside. Look at the CO2 emissions for five of the eight scenarios – they are woefully short of achieving a total removal of direct (‘tailpipe’) emissions from road transport by 2050 (even though the report says “Carbon Dioxide equivalent (CO2e) emissions are projected to fall significantly in all scenarios”). If these are possible scenarios based on published plans and funded policies then surely this is pointing quite alarmingly to the need for Government to do something more – which it may see as accelerating the take-up of electric vehicles or a need to reduce the distance travelled by carbon emitting vehicles, or both. That’s a matter for serious consideration elsewhere.

How are we going to net Net Zero?

As an aside, let’s remind ourselves that the Government was found wanting?by the High Court earlier this year because its Net Zero Strategy was considered to be unlawful – it had “illegally failed to include policies it needed to deliver the promised emissions cuts”.

Five of these projections indicate over 40 megatons of carbon would still be getting emitted by road transport annually in 2050. Meanwhile, the Jet Zero strategy for aviation sees emissions (for domestic plus international aviation for the UK) go down from 38 megatons pre-COVID to 19 megatons by 2050 (based on pursuing a 70% growth in aviation). Somewhere in the economy there will need to be a means to magic away all these megatons standing in the way of Net Zero!

I don’t like to grumble, but...

You might consider it a matter of personal taste, but for me I thought the report was bordering in places on false precision – when we’re talking about a possible range in traffic growth between 8% and 54% that’s a reminder of a lack of confidence in the future – I would have liked to see ‘between about 10 and 50%’ in the narrative and possible % changes out to 2060 reported to the nearest 5 or 10%. That’s incidental you might say and you might be right. But I also think that optics matter and we need to remind ourselves that numbers are powerful and in turn also put them in their place when it comes to an unknown future. You’ll see above that I have brazenly rounded population projections to the nearest million for 2060.

I also found myself noting down as I read through the report the question of what has changed from the 2018 National Road Traffic Forecasts? I was then pleased to see a section specifically on this – but then somewhat disappointed that the section focused mostly if not entirely on comparing the Core scenario of the 2022 report with the Reference scenario of the 2018 report, rather than the two sets of scenarios from the two exercises. I’m not sure this helped me much in getting a sense of how our sense of uncertainty has changed over the last four years.

Lastly, if you are looking for a table in this report that sets the scenarios side by side and shows how each of the key drivers of change (economy, population, fuel price, EV take-up, trip rate, Level 5 self-driving vehicles fleet penetration etc) play out in the scenarios, you won’t find it. Your best place to look I think will be the Common Analytical Scenarios Databook.

Getting to grips with COVID-19

Will we ever really be able to consider COVID-19 and the national behaviour change experiment it triggered to be ‘over’? The effects of the pandemic continue to play out – in uncertain ways. It is a credit to DfT that it has nevertheless got to grips with this and, yes, made assumptions to do so. We are therefore presented with a well-considered attempt to ‘see through’ this global shock and offer up a way of making sense of possible futures for road traffic.

Signing off

In closing I should declare an interest in that I am a member of the DfT’s Joint Analysis Development Panel (referred to in the report) and have had opportunity to contribute thinking and constructive challenge to the work DfT has been undertaking. It has been no mean feat for DfT to prepare and account for this set of projections.

As the report alludes to itself, it is how they are now used that matters. In this respect, DfT reminds the reader that “the projections are not anticipated to be directly used to appraise individual road schemes, nor are they intended to be used to consider capacity changes on a specific road or solutions to specific local issues”. I hope people will take them seriously – as a set – and with open and critical minds consider the implications for road investment and expenditure. That should not mean that we do not continue to challenge them and how they will be applied over time – as indeed I’m sure DfT itself will con