The publication of Matthew Taylor’s employment review today has re‐opened the debate on the “gig” or “sharing” economy. Trade unions, politicians and commentators seem to see these platform technologies as a troublesome challenge, given they do not conform to existing business models on which employment legislation has been predicated.
But, dear readers, that is the whole point of these new forms of business. New regulations or rights granted to gig economy sellers will merely cost those sellers and buyers in other ways, and stifle innovation in these sectors.
Let’s start with the economics. One of the major questions considered by Nobel prize winning economist Ronald Coase was “why do firms exist?” If free markets work so well, then why do we see these islands of planning, with their hierarchies, staff oversight, HR departments, and standardized contracts?
His answer was “transaction costs”. It was very costly, in terms of time and money, to continually search for an individual worker to undertake each task, to research his or her likely quality, to negotiate a price with them, to monitor their activity and to prove a transaction had taken place.
But new technologies have lowered these transactions costs substantially. Platforms have allowed the creation of direct two‐sided buyer and seller markets. Apps and online technology provide us with options and pricing information instantly and clearly, use within‐market ratings and trust mechanisms, and allow transactions to take place in a reliable and immediate way, without the need for negotiation.
Uber, AirBnB, Deliveroo and the like are therefore best thought of as providing a platform, and acting as an intermediary, to allow buyers and sellers to interact. They are not “employers” in the traditional sense.
As my former colleague Diego Zuluaga has outlined, since all markets have different types of transaction costs, the nature of each platform and its features will be different. For Uber, and the market for getting a lift from door‐to‐door, the highest transaction cost comes in terms of price negotiation. The Uber platform therefore uses its technology to set prices centrally, albeit varying them with localized demand. For AirBnB, where every room or house being rented is different, that would make no sense. Where that site acts to lower transaction costs significantly is in facilitating trust between users. The two‐way rating system, comments and generous insurance grease the wheels of activity.
Once one grasps this economics then you can see the difficulty of trying to shoehorn gig‐economy activity into the framework of employment rights. On platforms such as Uber, the activity is the service provided, but drivers have lots of flexibility in being able to turn down jobs and work the hours they want. Many drivers work simultaneously across a number of different ride‐share apps. In that scenario, how could drivers be guaranteed a minimum wage or other new “rights” for the time they spend on the app?
The Taylor answer is said to be a new guarantee for “at least the hourly minimum wage to people who want to earn more, make themselves available to work during times of demand, and work at an average pace”. He recognises that those who work across numerous platforms, turn down work or work in “slow” periods should not have these benefits. In other words, the platforms themselves will have to track and provide information on a number of variables to drivers in advance, with the rights dependent on the seller’s activity and general market conditions.
While not as damaging as a crude blanket imposition of rights, this will be a major bureaucratic headache. Having the platforms track all of this information will reduce their efficient scale; some drivers will not engage in enough activity to make up for the fixed costs involved. Some platforms will react then by scaling down, and providing fewer services, to the detriment of users. Others, and Deliveroo has expressed this explicitly, have talked about trading off new rights by reducing flexibility for its riders, and moving to shift work.
The more important deleterious effect though will be a curbing of future innovation. As Sam Dumitriu of the Adam Smith Institute has outlined, projecting likely earnings and tracking activity won’t be a huge deal for companies like Uber, given their incredible software and the largely homogeneous nature of the service. But these types of requirements will act as a significant barrier to entry for potential new entrants, whether that be competitors or new business models entirely. They will also prove a major headache for more complex business models such as Taskrabbit, where the tasks undertaken and productiveness of workers are more variable.
Maybe this would be a debate worth having if there were significant disputes with platforms by users. Yet, to take Uber as an example: extensive polling of Uber employees found that five out of six drivers say they prefer the flexibility of being self‐employed to the possibility of employee perks. The average Uber driver earns well above the National Living Wage already, even after costs. Workers are voting with their feet.
Given the clear preference of those who work in the gig economy, the desire to impose new rights looks like a paternalistic agenda rather than something necessary to quell the dissatisfaction of sellers. Either way, any new regulations or rights will certainly come with real economic costs.