The benefits of autonomous vehicles (AVs) are very well established including the potential to improve safety, reduce emissions, capture lost time, improve quality of life, save money, improve equitable access to transportation etc. Morgan Stanley estimated that the US might save $1.3 trillion a year when AVs are fully deployed and that global savings might be $5.6 trillion. For the Canadian context a report by the Conference Board of Canada, the Van Horne Institute and the Canadian Automated Vehicles Centre of Excellence (CAVCOE) estimated more conservative economic benefits to Canada of $65 billion per year. Whatever the numbers they are on the scale of full percentage points of GDP.
But these estimated benefits are not necessarily directly related to revenue generation and profits for the private sector. Yet AV technology is being developed at an increasingly rapid pace within the private sector and that more investments, partnerships, mergers and new players are being announced on a regular basis.
So, what are the compelling business cases that has prompted this acceleration in investment in AVs? How big is the prize that they might be fighting for?
Within Five Years
First it is important to understand how close we might be to seeing the first AVs on public roads.
When asked, in January 2016, when we will see fully autonomous vehicles (AVs) on the road US Transportation Secretary Anthony Foxx responded:
“I think we’re going to see it within five years. That doesn’t mean 100 percent penetration; that just means market availability. But I actually think we’re going to see it within five years.”
However, we may see some form of road-going AV even sooner than that. For example, Astro Teller has stated in April 2016 that the Google self-driving car project is “close to graduating from X.” At this moment in time we do not know what vehicle or service that Google might initially deploy, or its capabilities, but we do know from past presentations that they anticipate both incremental deployment and improvements. Google currently have a fleet that includes modified Lexus SUVs and their own prototype 2-seater electric pod that is limited to 25mph, as well as a collaboration with Fiat Chrysler to create 100 self-driving Pacifica minivans in the near future.
If initial AV deployment is with pods limited to 25mph on public roads this could still have a substantial impact on transportation within a geo-fenced area of a city. Here it is worth noting that New York City recently lowered the default speed limit from 30mph to 25mph for safety reasons, and in so doing have possibly helped create the ideal conditions for the first generation AVs to prove their value.
Market size and opportunities
Second it is helpful to understand how big a pie that this new mobility market may represent.
The reason that Google (now Alphabet), or any other AV developer might be keen to get to market first is because of the scale of the opportunity. Adam Jonas of Morgan Stanley summed it up thus:
“The auto industry is a century-old ecosystem being ogled by outside players hungry for a slice of a $10-trillion mobility market,”
“The car, on our estimates, is the world’s most underutilized asset.”
“We believe it’s the most disruptable business on earth.”
However, we can only begin to guess at the new business models and additional value that AVs will generate. Such value creation will come from novel business models around the movement of people, goods and services as well as the provision of new services by these vehicles. It is possible that the data generated by AVs could itself be a multi-trillion dollar market additional to the mobility aspect [more on this in a future blog post]. There will also be a huge app market develop around AVs, which could be as transformative for our vehicles as it has been for our phones [again, more on this in a future blog post].
AVs will take new forms
It is also worth noting that when considering an AV future that we move beyond a perspective that is limited to the possibilities of the current transportation paradigm. There will be new types of vehicles, new modes of transportation and new opportunities and business models.
Once the incredibly challenging problem of safe autonomous driving has been solved then the technology can be calibrated to work on any vehicular platform. Future modes could therefore also include pods that physically combine into road trains as they enter higher density traffic flows to provide a similar passenger density to a conventional bus; or self-delivering bicycles that overcome the single biggest problem of current bikeshares which is re-distribution of bikes following tidal flows; or sidewalk friendly delivery robots that will be very efficient at small package local deliveries and take much larger delivery vehicles off of the roads.
When Elon Musk, CEO Tesla stated:
“We have an idea for something which is not exactly a bus but would solve the density problem for inner city situations,”
and then added the following in his ‘Tesla Master Plan – Part Deux’:
“In addition to consumer vehicles, there are two other types of electric vehicle needed: heavy-duty trucks and high passenger-density urban transport.”,
then he might well be referring to smaller pods that physically combine into road trains that may well be a game-changer on top of the AV game-changer. Examples of this concept already in the public realm include:
Key principles of money flows around AVs
It can therefore be seen that the subject of AVs is so complex that we cannot forecast with any certainty what will happen with them, but certain key principles will likely dictate the general direction of deployment scenarios and the rate of market penetration. If we understand how the money might flow around this rapidly emerging eco-system it might provide greater clarity on the likely deployment scenarios that may develop on our roads. Although municipalities and regional and federal policy will also play an important role in shaping the future of AVs, in capitalistic economies we know that money flows can, and often do, play a dominant role in market development.
Principle 1 – AVs will do work and make money for their owners
Unlike conventional vehicles, which mostly sit idle 95% of the time, an AV can be put to work moving people goods and services – and providing services. This will make money for the owner. We have never had a product like this before – an autonomous robot doing work outside of factories. Applying basic capitalistic principles, then such a product that makes money for its owner can reasonably be expected to be in very high demand and will therefore sell faster than it can be made, until the market is saturated.
Or, as Elon Musk stated in his Tesla ‘Master Plan – Part Deux’:
“Enable your car to make money for you when you aren’t using it”
This raises the expectation that just as higher net worth individuals buy houses to rent out as an investment, so people will invest in owning multiple AVs. With economies of scale it makes sense that individual AV owners would hire out their vehicles via a cloud company that would leverage the benefits of having as many vehicles as possible within a given area to provide the highest level of service to customers and maximizing the possible revenue stream back to the vehicle owners.
But such individual ownership AV fleets for hire via the cloud will rapidly begin to compete with the major MaaS fleets that are also a logical result market competition as described following.
Principle 2 – Mobility-as-a-Service
Mobility-as-a-Service (MaaS), limited in this article to the context of AVs for simplicity, is the ability to order a ride on your smart device just before you need it, pay electronically in advance and then it takes you where you want to go, where you then off-hire it. To avoid the dystopian scenario of this resulting in even more vehicles and congestion on the road municipalities and service providers can promote ride-sharing of these AVs to increase average vehicle occupancy (AvVO) which is currently around 1.2 people per vehicle in peak periods. CAVCOE’s thought-development work estimates that if we can achieve an AvVO of 1.8 then this would have a very significant impact on congestion reduction, network efficiency and journey time reliability. An AvVO of 1.8 would also allow for some degree of induced demand including an increase in usage from the 30% of the population that currently don’t have driving licences e.g. many that are disabled, seniors, too young, poor and medically at-risk.
What MaaS does in terms of business models is to create fierce competition between mobility players that currently are rarely in direct competition. This can be explained by considering some of the mobility choices on offer to a visitor who flies into a new city (i.e. doesn’t have access to their own vehicle of any type). The visitor has a choice of numerous distinct business models for how they might wish to move around, including by; taxi, car rental, car share, ride share or Transportation Network Company (TNCs e.g. Uber, Lyft). However, once AVs are deployed then each of these business models will lose market share if they don’t have AVs, and possibly gain market share if they do. But the interesting thing is that once each of these different business models deploys AVs then they end up having the same virtually indistinguishable business model – that of the MaaS provider as described above.
Think about that for a moment; companies that might only be in limited competition at the moment, and have found a comfortable status quo, will end up in very fierce competition once AVs are deployed. They will literally be competing for their business survival. But they will also be competing for a much larger market than currently.
Of course the automotive manufacturers (the OEMs) also understand that their revenues are threatened by the evolution of MaaS. Many OEMs have been active in developing strategic partnerships with companies in the car share, ride share and TNC space, as well as bolstering their own AV R&D with the acquisition of AV development capability, e.g. GM recently invested in Lyft and bought Cruise Automation and plan to demonstrate their own fleet of AV taxis. Some OEMs like Ford and BMW have indicated an understanding that they may need to move from being car manufacturers to mobility service providers in the near future.
It is therefore quite possible that the OEMs would want to ‘cut out the middle men’ and operate their own fleets and become the MaaS providers directly. Why would they provide fleet vehicles at a lower margin than individual sales to the public to fleets that will be generating ongoing revenue and profit?
The Conference Board of Canada report also estimates that a household that gives up private ownership of a conventional vehicle and uses a MaaS service could conservatively save approximately $3,000 per year. Even higher savings for individual users were predicted by The Earth Institute, Columbia University in the paper ‘Transforming Personal Mobility’.
So it can be seen that this new autonomous robot product that makes money for its owner will inevitably end up in fleets providing mobility as a service. Whether via private individuals either choosing to hire out their own AV when they are not or via their investment AVs, or via the corporate fleet sector there will be very strong competition to provide mobility services to the public.
It is therefore possible that initial AV production will go directly into large corporate shared fleets and not be sold to individuals at all. With so many businesses facing loss of market share and possibly going out of business they will do all that is necessary to secure access to a technology that is expected to ensure their survival.
The competition will be very fierce indeed and we can expect to see a number of household names and global brands disappear as their traditional market sector becomes subsumed by the new MaaS business model.
Rapid Market Penetration
With a product that makes money for its owner, and demand exceeding supply, then it would seem to be that market penetration will be a direct function of how quickly the product can be made. Except this will not be the case, as AVs actually have a leveraged market penetration.
It should be noted that a single AV that is used in a shared fleet will replace multiple private vehicles. MaaS related studies have shown a very wide range of replacement ratios from approximately two (Kornhauser, Princeton), to ten (OECD ITF report) to 13 (Martin, Shaheen, Lidicker UC Berkeley). A more pragmatic approach is based on the fact that during a busy peak period that approximately 20% of all private vehicles are on the road at any one time that a maximum replacement ratio of 1 to 5 is probably more realistic. However, a significant increase in AvVO from 1.2 to 1.8 would allow for the potential for this replacement ratio to be higher, although allowance for a certain percentage of vehicles out of service for maintenance or charging/fueling must be allowed for.
Therefore with a highly leveraged market penetration, where a single AV is capable of replacing, say 5 privately owned vehicles, the impacts of AVs on the road network and various transportation business and operational models could be extremely rapid.
The author is developing a model to estimate how rapid this market penetration might be, which will hopefully be the subject of a future blog post. However early indications are that market saturation by AVs might not take multiple decades as suggested by other models from market analysts, rather a single decade or even less.
Implications for Government and Businesses
The compelling business case for AVs that results in highly leveraged market penetration will have profound implications for most businesses and most government departments and ministries at municipal, provincial and federal levels.
Because AVs present such a highly complex problem with far too many variables to be possible to predict with any certainty whatsoever then no forecasting method can provide a sufficient degree of comfort as to its accuracy. However, the author hopes that by raising the possibility that the compelling business case could result in very rapid market penetration and resultant impacts on business and operational models in the public and private sector, that this would encourage organizations to carry out their own due diligence and assess their level of risk exposure.
Disclosure: The author works for a consulting company (CAVCOE) that advises businesses and organizations on the potential impacts that AVs might have on operational and business models.