Uber Regulated?

How We Should Regulate ‘Smartphone’ Taxi Apps

The emergence of smartphone-based apps to order and pay for taxis could revolutionise how these markets work – and reduce the need for onerous regulation. But not all regulation is bad, and some of it might still be required to keep these markets working smoothly. In this bulletin, we discuss why the key challenge for Australian policy makers is knowing where to stop (over) regulating.

UPDATE: In October 2015, Frontier Economics published a follow-up piece to this bulletin, Levelling Up, which discusses why it is important to understand the economic value likely to be created by these services.

Companies offering platforms that enable different forms of selling or sharing are having a real impact on market transactions in areas as diverse as hotels and house sharing (AirBnB) to discounted bulk buying (GroupOn). These new platforms are raising thorny questions about why and how we regulate transactions in these markets.

An area of considerable platform growth – and contention – has been in smartphone apps for hiring taxis and other small commercial passenger vehicles. Uber and other taxi apps – Australian competitors include ingogo and goCatch – are all subtly different, but their use of sophisticated technology means that each offers a far superior customer experience to the old method of waiting in line at a rank, or calling and hoping that a vehicle arrives. Notably, the apps themselves are not transport providers; like a booking agent, they are merely intermediaries between transportation service providers and consumers.

The consensus view among consumers (and certainly most economists) is that the case for regulating taxi apps is not strong. Existing taxi regulation appears to protect the interests of those in the industry, not those reliant on taxi services. Apps only seem to be vilified because they threaten to break this entrenched order down.

These sentiments might be true, and the ability of these apps to break down the existing order is almost certainly true. It is therefore little surprise that many in the taxi industry are antagonistic towards these apps. But why are regulators and policy makers concerned about these apps? Regulators have expressed two main concerns about the way these apps are operating.

One concern is that the app technology makes it feasible for passengers to more easily hire or share private vehicles. UberX, Uber’s ‘low cost’ service, allows drivers of private vehicles to ‘ride share’ for payment, as do other competitors like RideSurfing and Lyft. Allowing these vehicles would undermine the taxi and luxury vehicle regulatory licensing structure that operates in most Australian jurisdictions.

Another concern is whether these apps insist on the use of licensed taxi or hire car drivers. Uber’s policies are to require criminal and driving history checks, but not compliance with legal licensing requirements. In Victoria and NSW, regulators have launched operations to crack down on unlicensed drivers and subject the apps to the same regulations that apply to existing booking networks.

These apps are re-igniting some old regulatory issues. Why should we regulate passenger vehicles like taxis? Will less regulated markets deliver the services that consumers want? And can smartphone booking apps change the answers to these questions?


We first need to recognise that some regulation can make markets work better. In the insightful Reinventing the Bazaar: A Natural History of Markets, John McMillan argues that markets work smoothly on five conditions or ‘platforms’:

  • Information flows smoothly
  • Property rights are protected
  • People can be trusted to live up to their promises
  • Side effects on third parties are curtailed
  • Competition is fostered

In many situations, markets can evolve to deliver these things from the ‘bottom up’, with no or minimal government intervention. EBay, the ubiquitous auction site, is a good example; it has developed mechanisms such as seller and buyer ratings to deal with the critical problem – the lack of trust – that would otherwise hinder transactions on its market platform.

For taxi markets to work effectively, consumers must have adequate information and be able to trust drivers. It’s hard to assess the quality and safety of drivers and vehicles before we travel, and hard to assess what is a reasonable fare – particularly if we’re not from the area. In most jurisdictions, we’ve relied on regulation to mitigate these problems; to vet drivers and vehicles so they are safe and comfortable, and to set reasonable fares.

While we have safer drivers and vehicles on the road, this regulation has come at a high cost — namely reduced competition. Not only among taxis (with fare regulation and restrictions on the numbers of taxi licences), but also between taxis and potentially substitutable forms of transport, such as small buses. Further, in many jurisdictions, we’ve produced a lot of regulations that go well beyond basic safety requirements – such as the colour of vehicles, dress codes and other vehicle markings. This has helped produce a stagnant industry with a strong interest in maintaining the existing regulatory arrangements.


The reason why Uber and its competitors are so promising is not merely because they threaten to undermine the existing order, but because they can reduce the scope of regulation. They can do this by solving the problems that markets have had in ensuring that information flows, and people live up to their promises. Further, and most importantly, they can foster the competition that is needed to get the most out of taxi markets.

To understand how apps can solve competition problems, it’s helpful to distinguish two types of taxi trips. Pre-booked trips are those that are booked by consumers prior to travel, from home, work or their mobiles. Rank and hail trips are where taxis are taken from a queue at a rank, or hailed down in the street.

Analysis of these markets has shown that competition and market performance problems are far more prevalent for rank and hail trips. Unfortunately, this accounts for up to 70 per cent of trips in the dense urban areas. The main reason competition doesn’t work well in these markets is the lack of ‘repeat business’. Consumers will rarely use the same taxi twice. This means that reputation is not important; if I have a bad experience, I will not get the chance to punish the taxi operator by not using their service again. Where there is little chance of repeat business, there is little incentive to deliver a quality service.

Uber and other apps can help solve quality problems by allowing users to book a specific taxi with knowledge of its current location and the quality rating of the driver. This allows these vehicles – where markets work better due to the importance of reputation – to compete against rank and hail taxis – where markets don’t work as well. Information flow is better (fares can be estimated and then metered on users’ smartphones), and bad performance can be punished as immediate feedback of drivers and vehicle feedback can be given.

By its nature, a platform like Uber must ensure that affiliated operators deliver good service. It’s all too easy for consumers to switch to another platform if Uber can’t deliver good service, or offers prices that are too high.

The upshot is that we now have market mechanisms that will look after driver and vehicle quality. In turn, this means that we can be less worried about enforcing regulations about quality.


But what about safety? To be sure, Uber also has incentives to keep passengers and drivers safe. A reputation for unsafe operation damages business – ask car manufacturers or airlines. Indeed, Uber does claim to do significant driver checks, even for its low cost UberX service which uses non-taxi vehicles.

The key regulatory question is whether a loss of reputation alone would be enough to ensure taxi operators maintain safe drivers and vehicles. Here, there are two relevant considerations.

The first is whether consumers will attribute poor safety performance by one taxi or ride-sharing operator to all operators. A side effect of unsafe operation could be a lack of confidence and a decline in demand for all taxi services.

The second consideration is whether reputational effects would provide sufficient protection for drivers. Taxi driving is notoriously dangerous, with some estimates putting average occupational violence at up to 15 times that of other occupations. It is unclear how strong incentives would be for taxi apps to insist on safety measures to protect drivers of affiliated vehicles.

These potential costs seem hard to quantify. It is therefore difficult to assess whether the costs of regulating (for example, driver checks and vehicle testing) are low compared to the expected benefits. Simple safety checks of drivers and vehic les seem to have low costs compared to benefits, such as the avoidance of major accidents, uninsured vehicles, or major personal crimes against drivers or passengers. However, in most Australian jurisdictions, licensing is also used to limit the numbers of vehicles – meaning that even vehicles that could pass appropriate taxi safety checks could not be used. This increases the costs of vehicle regulation and, particularly where taxi markets have been highly restricted, may tip the balance towards a more liberal approach to vehicle sharing.


Consumers are right to be sceptical of taxi regulation and encouraging of new taxi apps. If these apps are allowed to flourish they will increase competition between taxi suppliers, improve service quality and make the jobs of policy makers and regulators easier. Policy makers can forget about regulating vehicle and driver quality, and instead focus on passenger and driver safety.

The key battleground will be how policy makers weigh up the benefits to consumers from having greater access to non-taxi vehicles with the safety benefits of insisting on licensed drivers and vehicles. Good policy demands that these safety benefits must be delivered while fostering competition. Over regulation would waste the opportunity provided by taxi apps to deal with long-standing market problems.

UPDATE: In October 2015, Frontier Economics published a follow-up piece to this bulletin, Levelling Up, which discusses why it is important to understand the economic value likely to be created by these services.

Start typing and press Enter to search