Transport Investment under Uncertainty: Transport Knowledge Hub workshop 12 December 2017

Image: ExpectGrain
Transport Knowledge Hub logo Published on: 15th December 2017 by David Fowler.

The world is poised on the cusp of a transport and societal revolution which, over the next 10 to 15 years, “will drive change on the scale of the Industrial Revolution”.

So said KPMG global head of public transport and UK head of infrastructure Richard Threlfall, introducing this week’s Transport Knowledge Hub workshop. TKH’s second workshop, entitled Transport Investment under Uncertainty, addressed the question of how to approach transport investment at a time when disruptive technology was likely to have a profound effect on travel patterns, and the wrong decision could lead to millions being wasted on assets that became worthless or “stranded” midway through their life.

Mr Threlfall believed the issue was even more fundamental.

“Do we just let the market get on with it and see what society we get as a result? Or is it a fantastic opportunity to involve policymakers to gently nudge the way this technology plays into our lives, so that we get a better society out of it?”

An “avalanche of announcements” in the last few months illustrated the accelerating pace of change, he said. The UK and France had announced dates by which they intended to ban sales of petrol or diesel cars. Volvo had said all models it launches from 2019 will have an electric motor, either as the sole power source or in a hybrid configuration. Volvo had also entered into an agreement with Uber to supply up to 24,000 sport utility vehicles by 2021 as the basis of a driverless vehicle fleet. And last month Waymo began trials of a fleet of autonomous cars in Phoenix, Arizona.

“The sheer pace of change is extraordinary,” said Mr Threlfall. “For anyone involved in any way in public policy, we’ve got to be on it now.”

Policy implications

Charlie Simpson, head of KPMG’s global mobility 2030 business, said three distinct but complementary disruptive forces were coming together: electric vehicles, autonomous vehicles and mobility as a service. “Each of these is a significant trend in its own right, But when you layer them together, they are truly transformative,” he said.

He added: “It’s easy to imaging a utopian outcome, but dystopia is just as likely,” as the recent controversy between Tfl and Uber in London had shown.

Edwin Kemp, an associate partner in KPMG’s global strategy group, outlined expected trends in mobility as autonomous vehicles and mobility as a service (MaaS) became widely adopted.

KPMG’s 2017 Global Auto Executive Survey identified that industry executives believed that up to 50% of current car owners would no longer own a car by 2025 as new mobility services were introduced. Passenger miles travelled would increase by 10%, while the cost per mile could decrease by up to 40%. Because cars will be “shared” rather than owned, the miles covered by each vehicle will increase by a factor of five. And the number of companies participating in the sector will decline as consolidation take place.

Autonomous vehicles could make possible waves of untapped mobility growth, especially among the youth and senior markets, for whom AVs will provide a way of travelling safely and independently.

Mobility as a service will be up to 40% cheaper per mile than the cost of owning a car.

With new powertrains and without a driver, taxis will become very competitive.

Opportunities for the economy will be created through three main benefits. In 2015 KPMG estimated the value of consumers’ freed travel time up to 2030 at over £20bn; the value of more efficient journeys contributing to greater productivity at over £15bn; and the value to UK industry at £16bn through new revenue streams and services. While travelling in a connected autonomous pod, passengers would be free to consume “a vast array” of entertainment and media content and on-demand services. Customers would will buy mobility subscriptions, pods will be “consumer-centric” and personalised, and services will be available directly to pods.

But although this could be the norm in urban areas, in suburban and rural areas the story could be different, with lower take-up of Maas. In rural areas there could be little coverage by on-demand autonomous vehicle services. There could also be differences from city to city.

Mr Simpson considered the automotive sector perspective, and said the challenge would be not to focus too much on the supply side viewpoint. If it was cheaper and easier to access mobility through MaaS rather than owning a car, “a lot of misapplication of capital is likely to occur”.

Regarding the availability and introduction of autonomous vehicle technology, he said “the perspectives of the major car makers have converged over the last 12 months”, with a clear timeline for introducing AV technology.

Five levels of autonomy have been identified on a spectrum from none to full autonomy. Level 5, with no human input, was expected to be achieved between 2025 and 2030, “and we are already at level three”, he said.

He added that “there is a clear view that humans are pretty bad at driving”, illustrated by the fact that insurance companies are offering Tesla drivers a discount if they are seen to be using automated drive mode. Automated systems are seen as better at avoiding accidents such as rear end shunts than humans.

“The pace of adoption is very dramatic,” he said. Substantial investment is being undertaken by companies such as Tesla, Waymo and Otto (in self-driving trucks). Goods miles travelled are increasing exponentially in the US, he said. Many truck drivers are in their fifties or sixties, and they are not being replaced by young people. “The gap is already being filled by autonomous vehicles,” Mr Simpson said. Otto’s self-driving rig undertook a delivery covering 120 miles of US highway in 2016.

Partnerships are being formed in the AV field to collaborate on experiments and to share risk, bringing city authorities together with the private sector.

Mr Simpson said the pace of adoption of AV-MaaS could be expected to be comparable with other disruptive technologies. “We are expecting it to mirror the uptake of smartphones and e-commerce,” he said.

But while new entrants such as Uber and Tesla have low fixed costs, the big car makers will be faced with trying to dismantle “a 100-year-old business model while investing in a new one – that’s difficult to pull off”. Business valuations of Uber and Tesla (their 2017 market capitalisations are similar to, or greater than, GM and Ford) demonstrate the market’s view of their potential.

Commercial considerations

Simon Craven, a special adviser to the Go-Ahead group, considered the future of mobility from a commercial perspective. He began by quoting James Gleave, of Transport Futures, who said: “One nagging question keeps coming up [about MaaS]. How on earth will anyone make any money out of it? It is very difficult to make any money out of transport. Transport is a high cost industry, with a lot of money tied up in vehicles and infrastructure.”

Investment is a word often used loosely, said Mr Craven, but in its true sense investors accept a risk with the incentive of a return on their capital. Success allows reinvestment for growth and to replace assets. Without a reasonable prospect of bringing in a return, the activity is not investment but speculation, or gambling, and a business based on it will sooner or later fail.

“Mobility industries are massive capital investors in infrastructure, fleet and facilities,” he said. These are long-lived assets, which need to generate a return over a lifetime of, for example, 14 years for a bus, over 40 years for rolling stock, and over a century for a road or rail track bed.

“When we invest shareholder money or public funds in an asset, we’re failing in our duty if we don’t manage the risk of it being stranded – becoming economically unviable,” he said. “So we need to manage our business and our stakeholder relationships so that the commercial use of these assets is likely to be defensible for their economic lifespans.”

Mobility systems need three kinds of sustainability, he continued. In addition to environmental sustainability they also needed political/social sustainability, where there are questions, for example, about the monetisation of data derived from passenger mobility. A robot taxi would be in effect “a mobile online surveillance platform” constantly harvesting data. Was this acceptable?

There also needed to be economic sustainability.

“This is being glossed over in many discussions of MaaS,” he said. “Some people don’t really talk about the extent to which new types of MaaS are only being proved to work with big subsidies – from the taxpayer, from investors, or in some cases from gig-economy workers who aren’t really making the numbers add up.

“Whatever systems we build, they will ultimately let down our communities if they are not built on honest, sustainable economics. If you’re a company which wants to embed itself in the heart and lungs of a living city, then you’d better have a great deal of confidence about your ability and willingness to be there in the long term.”

2030 is 13 years away. Looking back over the last 13 years, technology companies have lived through disruptive changes. By contrast, for over 50 years change in transport industries has been incremental. Transport is now entering an era in which substantial change is under way. There is a “cultural risk” of managers in the these two industries, rooted in different commercial experiences, “failing to understand each other, and failing to achieve win-win solutions”.

“Mobility in 2030 has to be good for our customers, for the cities, market towns and rural areas we have to serve, in their totality. Or this whole thing is not worth doing,” Mr Craven said. “We have to find win-win solutions which work commercially… pay the workforce something they can actually live on, and meet the cost of capital as well as operating costs for the fleet.”

He added: “Maybe that’s a reality that some recent entrants into mobility need to embrace, just as more traditional mobility providers need to keep up with the times.”

Nor must the extent of the commercial investment be underestimated. The UK car fleet is 26 million strong. If autonomous vehicles became the norm, the current fleet could perhaps be replaced by an AV fleet of a tenth the size, say 3 million. If each AV cost £50,000, this would add up to £150bn, just for the UK fleet. This is equivalent to three HS2s, or the market capitalisation of the world’s most valuable car maker, Toyota, or about ten times the amount Uber has so far been able to raise.

“So for me, regardless of what technologies predominate, a good-looking investment scenario for Mobility 2030 is one where partnership thinking predominates,” Mr Craven concluded.


Prof Greg Marsden of the Institute for Transport Studies at the University of Leeds looked at the question of governance.

In planning for the future, he said, “there is lots of uncertainty even if the world carries on as it is – but it won’t.”

He set out three important contentions. First, smarter mobility is coming. Second, no amount of smart technology or big data will overcome the need for good policy, planning and governance. Third, there is a need to plan actively to try to bring about socially desirable outcomes from smart mobility.

He spoke about the role of the state and contrasted the enabling state, which stimulated others to action and then let them get on with it, with the “ensuring state”, an enabling state which in addition is “expected or obligated to make sure such processes achieve certain defined outcomes”.

The state intervenes in transport for “robust reasons”, he added. These include the fact that transport supports wider public policy; to establish common rules; to define allocation of space; because conditions for a free market do not exist; to provide socially necessary services; and to fund and invest in infrastructure. These will continue to be important in the future.

New mobility services must be seen as an opportunity, he argued. Driverless cars will probably be electric, and would contribute to better air quality. They would need less space (compared with today’s cars, which are parked most of the time), which could free space in cities for parks and expanded public areas.

“But there are big questions about whether we will get to this end state, or whether there will be a messy transition in different places,” he said.

Conventional thinking in transport policy assumes smart mobility will be more efficient, but this is not guaranteed, he warned. Passengers in autonomous pods will be a captive audience, giving the operator “control over their time and choices”. There is a risk of a monopoly or oligopoly being created, which would use its power to exact higher charges.

“The ensuring rather than the enabling state would be able to capture the advantages better,” said Prof Marsden.

There would be continuing reasons for intervention, he continued, in most of the same areas as now, including establishing rules concerning transport modes and access; allocation of space; to provide socially necessary services; and for funding and investment.

It is unclear exactly how the government or state’s role would be performed. At present there is a fragmented regulatory structure: for example there are 353 taxi regulators. “Do we need to consolidate [regulation] into fewer, big organisations?” Prof Marsden asked.

And there could be a range of outcomes, along the twin axes of “hands-on” versus “hands-off” governance and between social acceptance and social resistance of aspects such as data sharing and automation in various forms.

“How we approach the transition will make a difference,” he said. “Effective and extensive deployment can be accelerated by good governance.”

For planning and regulatory authorities, uncertainty could be reduced by considering what outcomes new mobility systems could help to achieve; how regulating their introduction could facilitate this; by avoiding damaging spatial competition; and by sharing knowledge and planning adaptively.


In discussion after the presentations Mr Threlfall asked what should we be doing now. Mr Craven said local authorities should be thinking about how to integrate mobility services with social objectives. “I don’t’ see that work going on,” he said.

From the floor TfL’s David Christie pointed out the difficulty of planning when technology in some forms, such as e-commerce, reduced the need to travel (for shopping). Prof Marsden said that in general MaaS was expected to encourage more travel because it would reduce the cost per mile, but the travel could be for different reasons from now.

There was a suggestion that the new forms of mobility could be managed better at city-region level, for example, than nationally, because at that level it would be easier to align objectives between transport and areas such as health.

In response to a question from a representative of Nexus about how to incorporate smart mobility into creating a business case for expanding the Metro, Prof Marsden suggested approaching the problem by thinking about how the catchment area could be significantly expanded through integrated services to make the service as widely accessible as possible.

Chris Mills of TfL suggested that widespread adoption of AVs could be at odds with the London mayor’s objective of increasing walking and cycling. Mr Craven stressed the need for an integrated transport system and argued that TfL was in a position to regulate autonomous taxis in such away as to fit in with other objectives – an approach taken by Copenhagen.

Image: ExpectGrain

About the Author

This post was written by David Fowler. David Fowler is a freelance journalist, covering transport, business and technology. He was Editor of Transport Times for ten years and has previously written for New Steel Construction, The Engineer and New Civil Engineer.