Uber, cyber serfdom and the entrepreneurial union

I like Uber, and find myself using it regularly. With a reported valuation of $68 billion, the San Francisco based company is now widely regarded as the fastest growing company of all time. In just six years, it has become more valuable than centuries old corporate giants like General Motors and Ford.

From the customer perspective, it’s cheaper and more convenient than licensed taxi companies. But what about the worker perspective? Proponents argue that Uber offers a way for drivers to earn a decent living in a way that permits flexibility and independence. On close inspection, I’m not convinced. In fact, it seems that Uber’s success is largely down to a type of confidence trick against its drivers. To see why, it’s worth spending a bit of time breaking down the underlying economics of the Uber business model.

All companies need to invest in new assets and maintain existing assets. The costs associated with this has a direct impact the bottom line. In the case of Uber, because drivers supply their own assets (the cars), the costs and risks of running the cars are externalised onto the drivers. Uber’s model relies on drivers not properly accounting for these costs in assessing their earning potential from Uber, and this is helped by the fact that some of these are less visible than others.

To illustrate, each year the AA publishes figures showing the overall cost of running a car in the UK. These include the basic costs of keeping a vehicle ready for use on the road such as car tax, insurance and financing costs as well as ongoing running costs such as fuel and servicing and repair costs. Crucially, it also incorporates depreciation to account for the loss of value of the vehicle — a cost which is easy for drivers to overlook as a day-to-day expense.

In 2014–15 these costs amounted to anywhere between 25 pence and 211 pence per mile, depending on the purchase price of the car and the number of miles driven each year. Of this, depreciation is the biggest single cost factor. If we assume that an Uber driver has three fares per hour and each journey averages 6 miles, this results in a running cost of between £5 and £38 per hour. This cost is borne by Uber drivers.

The amount that an Uber driver can earn in an hour can vary widely. Uber’s charging formula is notoriously opaque, and prices can increase dramatically during peak times due to the effect of “surge pricing”. Uber claims that on average drivers make around £15 an hour, and that’s after the company takes its 20% ‘service fee’ (though now drivers have to pay 25%).

On this basis, it’s clear that once all the running costs are accounted for being an Uber driver is not quite as lucrative as it may first appear. Depending on the car used and the number of miles driven, it’s entirely possible that some drivers may not even be making the minimum wage, and it’s technically possible (though unlikely) that some could even be making a loss. In addition, Uber insists that its drivers are self-employed ‘partners’, meaning they are not entitled to normal worker’s rights. This means that Uber drivers do not have rights to holiday pay or sick pay, or the right to properly challenge a discipline or grievance notice before being dismissed. These are additional costs which are externalised onto the drivers, further helping to bolster Uber’s bottom line.

It’s here that we find the dark side of the “sharing economy”. Despite the benevolent overtones, the reality is an economic model which sees low paid, self-employed workers provide all the labour and capital, bear all the risk, sacrifice their employment rights and then transfer 25% of all earnings to a corporate behemoth 9,000 miles away. As Douglas Rushkoff says: “Instead of creating truly distributive businesses we are just putting industrial economics on steroids, creating more extreme divisions of wealth and more extreme forms of exploitation.” A recent letter in the FT described Uber drivers as “cyber serfs” and accused Uber’s business model of “returning income distribution to the feudal age”. This isn’t a million miles away from the truth.

As this model continues to be replicated across other parts of the economy, it seems I’m not alone in wondering whether there is a better way of doing things. An obvious step would be to create a new open source platform which replicates the features of the Uber app but which would be owned and governed by the drivers. If this happened, Uber could find itself kicked out of the party. By cutting out the middleman, the drivers would keep the full fruits of their labour, and local economies would no longer see 25% of each fare disappear back to California. Fares could even be cut further so that customers benefit too.

Although the idea of a corporate giant such as Uber being outcompeted by an open source collaborative project may sound fanciful, it’s not unprecedented. You may remember Microsoft Encarta, the original (and very expensive) digital Encyclopaedia. Microsoft discontinued Encarta in 2009 after it was made obsolete by Wikipedia — a free, collaborative, non-commercial platform which is operated and maintained by millions of people all over the world. So why hasn’t this happened to Uber? Chris Dillow thinks that there there are three reasons:

1) Lack of entrepreneurship

2) A collective action problem

3) Credit constraints.

To this I would add another — problems of scale. There are already some examples of so-called platform cooperatives taking on the likes of Airbnb, Uber and eBay. The problem is that these tend to be small and locally focused, meaning that they are unlikely to be able to compete with the economies of scale and corporate might of the likes of Uber. Moreover, people appear to like the convenience of being able to use the same platform across different towns, cities and countries.

If we are to overcome the challenges of cyber serfdom, we need to learn how to do platform cooperativism on a large scale. This is no small task, but I wonder whether one option lies in reimagining the concept of collective bargaining for the technological age. While the role of trade unions has always been to organise workers to protect and further their rights and interests, the organisational forms and methods of trade unions are still stuck in the twentieth century. In the case of networked transportation services, it’s clear that the best way forward is not to engage in dialogue with Uber, but to subvert the company altogether by creating a new platform which is more conducive to worker interests.

It seems that there is a case for a new type of trade union, one that sees its role not as a negotiator, but as an originator. A type of trade union which sees its purpose not in fighting for better worker conditions within existing structures, but in redesigning structures on a cooperative and open source basis. A trade union that sees the key battleground as global cyberspace, not the local shop floor.

The question is whether there is the ambition and capacity within the modern trade union movement to make this leap. Mariana Mazzucato talks about the need for an entrepreneurial state. Maybe it’s time for entrepreneurial unions too.

Unless, of course, the robots get there first.