What is dynamic pricing? And why is it “dangerous” for gig workers? On the day of co-ordinated protests of Uber drivers in London and Amsterdam for fair pay, the Gig Economy Project takes a look.
Last week, Uber introduced ‘dynamic pricing’ to London. In response, drivers in the App Drivers & Couriers Union (ADCU) protested outside Transport for London’s offices today [7 February], calling on the regulator to prohibit dynamic pricing in the British capital.
“The Mayor and Transport for London failed in their duty when they gave Uber the green light to set loose such dangerous and predatory algorithms in what is supposed to be a closely regulated market,” James Farrar, General Secretary of ADCU, says. “Passengers and drivers are, directly and indirectly, unfairly targeted for personal auto-exploitation.”
At the same time as the London protest, Dutch Uber drivers in the FNV union mobilised outside the company’s European headquarters in Amsterdam to demand “fair pay”, in a co-ordinated action with their British counterparts.
Els Franssens, an FNV platform economy organiser, is also scathing about Uber’s pricing policy.
“They play with the pricing and use it to manipulate the drivers,” she tells the Gig Economy Project.
But what is dynamic pricing exactly? And why is it “dangerous” for gig workers?
What is dynamic pricing?
Dynamic pricing is highly variable price and pay rates based on algorithmic calculations derived from a variety of data inputs, which can include things like real-time supply and demand conditions, competitors’ prices, and historical data from workers and customers.
Dynamic pricing is not exclusive to the gig economy. American Airlines is thought to have been the first company to introduce dynamic pricing and now it is commonplace in the airlines industry. Hotels, retail, e-commerce and real estate are some of the other industries where dynamic pricing is widespread. The difference in gig economy sectors like food delivery and private hire is that dynamic pricing affects the worker just as much, if not more, than the customer.
For companies using dynamic pricing, the benefits are obvious. By continually adjusting prices to changing circumstances and/or specific individuals, they can optimise their rates. In a 2018 presentation on dynamic pricing to Microsoft’s research team, Uber’s former Senior Data Science Manager Dawn Woodard says that dynamic pricing allows Uber to set a “revenue-maximising price” as often as possible.
For customers, dynamic pricing can lead to confusion and anger. They may wonder why they are being charged a higher price for the same distanced journey from one day to the next. They may also question whether the price they are being set is exploiting their vulnerabilities, like the need for a female passenger to get home late at night. Finally, if they are on more than one private hire app, they may question why the prices of supposedly rival firms appear to be moving in harmony.
The risk of explicit or tacit collusion in dynamic pricing is real. Explicit collusion is when companies actively co-ordinate price setting. Algorithmic tacit collusion is when a company follows the price changes of a competitor in real-time and responds accordingly. But even without explicit or tacit collusion, a Harvard University study has found that prices could be inflated due to the indirect impact which powerful algorithmic price-setters have on less technologically developed competitors, pressuring them to set artificially high prices.
“An observer may naturally think that algorithms—which enhance the ability of firms to react to rivals’ prices—would intensify competition, but the reverse is true,” the study finds. “These theoretical models indicate that the increasing use of pricing algorithms will lead to higher prices for consumers, even when firms are unable to collude.”
But what is the impact of dynamic pricing on gig workers?
“If you joined Uber years ago, you will have joined when prices were quite simple. We set prices based on time and distance and then surge helped increase the price when demand was highest. Uber has come a long way since then, and we now have advanced technology that uses years of data and learning to find a competitive price for the time of day, location and distance of the trip.”
This is a correspondence from Uber to its UK drivers in August 2022 about the roll-out of dynamic pricing. It invites as many questions as it answers: What specific types of historical data is being used to calculate the pay rate, and what weighting do these types of data contribute to the overall algorithmic calculations? If time and distance are not dictating the price, then how can a driver work out how much he or she should be getting paid for each trip to ensure they are not being conned? And what does it mean for a price to be “competitive”; competitive for the customer, for the driver, or for Uber?
Uber insist that “dynamic pricing makes best use of drivers time” by optimising the relationship between supply (the drivers) and demand (the customers), but Farrar tells the Gig Economy Project that this does not compute with their experience of dynamic pricing.
“Drivers would be far better off with higher standard fares, because with dynamic pricing the pay is declining all the time not increasing,” he says.
Farrar believes that Uber and other digital labour platforms are increasingly adopting dynamic pricing in a bid to “improve their margins”, as they come under increasing pressure from investors to deliver profitability. An example of this is Just Eat, Europe’s largest food delivery platform, which in a presentation to investors last year stated they would increase revenue per order “through dynamic pricing and margin optimisation”.
But if dynamic pricing is raising prices for consumers, how can it simultaneously be squeezing the wages of drivers? The answer is that the link between consumers’ prices and drivers’ pay (with Uber taking a commission on each fare) has been broken.
“One of the areas that can lead to a driver being fired for fraudulent activity is monitoring the driver app and monitoring the passenger app at the same time,” Farrar says. “They really don’t want drivers looking at what’s being offered in terms of pricing in real-time.
“For a driver, there is no such thing as dynamic pricing. It is dynamic pay.”
Uber’s algorithm is setting prices which maximise revenue from customers, and pay which minimises expenditure to drivers.
“An absolute need for transparency”
On what basis is Uber calculating dynamic pay? The “years of data” which Uber uses to instruct its algorithms is a black box to all but a select few in the loop. However, there is no regulation stopping the company from using the personal data of individual drivers, such as their history of acceptance rates for rides, to inform what pay they give them, or whether they get allocated a ride or not.
Under questioning, Uber has insisted that it “does not use individual behaviour or performance when matching drivers with riders. It is based on location together with road and traffic conditions, rather than based on who they are, how they behave or perform. This enables a consistent, reliable and fair service for all users. Additionally, suggestions that Uber offers variable pricing based on user-profiling is completely unfounded and factually incorrect.”
While job allocation is different from dynamic pay, it obviously has a major impact on the income of Uber drivers if they are more or less likely to be matched with a user based on “past behaviour or preferences”. Drivers may also be wondering what the unexplained “other factors” are.
“What this shows is there is an absolute need for transparency here,” Farrar says. “The questions that remain unanswered here are significant.”
The ADCU General Secretary believes that dynamic pricing may be illegal under Section 1 of the UK Employment Rights Act, which entitles workers to a clear statement from their employer of the terms and conditions of their work before they start, including the rate of pay.
“A zero-hours contract worker has the stress of not knowing if they are going to get work that day, but if they do get work that day they know how much they are going to be paid,” Farrar says.
“An Uber driver doesn’t know if they are going to get work on any given day, and if they do know they don’t know how much they are going to be paid for it, so it’s double the stress.
“It’s impossible to know how much they will be paid, and that is a Section 1 failure.”
If, as a consumer, you don’t like dynamic pricing, imagine how it must feel to be a gig worker every day, trying to navigate your way around dynamic pay?
Transport for London may not be willing to do anything about it yet, but as dynamic pricing affects ever-more consumers and workers, this is not an issue regulators are going to be able to ignore forever.
The ADCU and Worker Info Exchange are looking for riders and drivers to give them permission to request their data in order to investigate dynamic pricing. If you are interested in participating, visit workerinfoexchange.org/request-data for more.
To sign up to the Gig Economy Project’s weekly newsletter, which provides up-to-date analysis and reports on everything that’s happening in the gig economy in Europe, leave your email here.
Thanks to many generous donors BRAVE NEW EUROPE will be able to continue its work for the rest of 2023 in a reduced form. What we need is a long term solution. So please consider making a monthly recurring donation. It need not be a vast amount as it accumulates in the course of the year. To donate please go to HERE.