As ride-sharing apps have decimated the taxi industry over the last few years, Uber has consistently presented itself as being good for drivers. It has released absurdly overinflated estimates of average driver compensation. In 2013, the Wall Street Journal reported that “a typical Uber driver takes in more than $100,000 a year in gross sales” based on figures from the company. The company had to pay a large settlement after it recruited drivers using compensation figures that were flat-out false. In 2015, Princeton economist Alan Kreuger, the former head of Obama’s Council of Economic Advisers, partnered with Uber’s head of research to produce an “analysis of Uber’s driver-partners” with impressive earnings statistics, showing that average drivers earned well above minimum wage, with their compensation significantly exceeding that of taxi drivers. Here’s a chart from the paper:

These statistics were cited as proof Uber improves drivers’ lives. The Economist (who else?) ran the headline “Princeton economist explains why we should all stop worrying and learn to love Uber,” saying that the “ambitious study” had shown that fears of Uber driving down wages were overblown. (The press has credulously parroted Uber’s fabricated statistics since the beginning, with headlines like “Could Uber’s 90k salary disrupt the taxi business?“) The Kreuger and Hall paper’s findings, with a highly accredited economist’s name attached, were a formidable public relations tool for Uber, which has long been fighting allegations that it underpays and exploits its drivers. The numbers proved, though, that if you’re mad at Uber, you should be even madder at the taxi industry, and that Uber is actually improving driver compensation. They also bolstered the pro-“disruption” argument that the taxi industry was little more than a cartel trying to protect the corrupt oligopoly that it had established over time. If Uber drivers earn more than taxi drivers, the arguments against Uber’s erosion of the taxi industry could not possibly be based on legitimate concerns for the financial well-being of those who drive for a living. The more Uber grew, the better it would be for drivers.

But the above chart is almost entirely meaningless. That’s because, as it notes in a small footnote, it doesn’t factor Uber drivers’ expenses (gas, taxes, insurance, repairs, depreciation, or possibly even renting the car itself) into its “earnings” calculations. If one column is “pre-expenses” and the other is “post-expenses,” then what conclusions can anyone possibly draw from this chart? Unless we know what drivers actually ended up with, we don’t know anything of value. I actually think it’s deeply intellectually dishonest for a reputable economist to produce a chart like this. Krueger said he didn’t include expenses in the chart because he didn’t have the data with which to calculate them. But if you don’t have the data, then you can’t make a comparison between Uber compensation and taxi compensation. Instead of declining to answer a question that he didn’t have the answer to, Krueger made a chart that showed an unsupportable claim.

(This is one reason why serious economists shouldn’t accept money from corporations to produce research about those corporations. The Uber-funded Krueger and Hall study was almost ludicrously propagandistic, from its use of “driver-partners” to its constant touting of “flexibility.” Look at this quote: “When asked directly in [Question 38], ‘If both were available to you, at this point in your life, would you rather have a steady 9-to-5 job with some benefits and a set salary or a job where you choose your own schedule and be your own boss?’ 73 percent chose the latter.” Talk about a biased question: do you want to be your own boss or do you want to have things “set” for you with “some” benefits? Shockingly, 73 percent of people would rather not have bosses!)

When you try to produce a real comparison of what Uber drivers make compared to what taxi drivers make (as opposed to a fabricated and nonsensical comparison), the situation becomes much, much murkier. Krueger and Hall were criticized for presenting such a misleading statistic, and two years later they updated their paper to include estimates of the effect of expenses on what Uber drivers actually take home. In the new version, they came to the much less impressive conclusion that “taking expenses into account, the average Uber driver-partner is likely to earn at least as much per hour, and probably more, than the average taxi driver and chauffeur.” (Emphasis added.) Even with this estimate, for part-time drivers they “disregard fixed costs, assuming that drivers are using a car they already owned which would have depreciated regardless of driving on the platform.” The expense estimates vary, but can be almost $6.50 per hour for a full-time driver of a large sedan or SUV. (In the original paper, the authors wrote that unless driver expenses ended up being “more than $6 per hour,” their central conclusion would still hold, implying that this was very unlikely.)

There are other factors that may make it still harder to come up with a useful comparison. Taxi drivers may, thanks to a greater reliance on cash tips, have a lot more income that goes unreported and would make the official estimates of their earnings lower than they actually are. At the same time, ride-share drivers can deduct their car expenses from their taxable income at a high per-mile rate, which would mean some fraction of their expenses are recouped on their tax bills (although many almost certainly don’t take all the deductions they’re entitled to).

We don’t really know, then, what an Uber driver makes compared to a taxi driver. There are reasons to believe Uber drivers might do worse on the whole than taxi drivers. A BuzzFeed News investigation, based on Uber’s own internal data from 2016, found that in states with high auto insurance rates like Michigan, expenses for drivers could eat up almost 1/3 of the “earnings” that Uber touts. Detroit Uber drivers ended up taking home about $8.77 on average, not much above the state’s minimum wage at the time of $8.50. (And below Michigan’s current minimum wage of $9.25.) According to BuzzFeed, Uber is not eager to try to figure out what its drivers actually end up with, and “says it doesn’t know how much drivers on its platform actually earn per hour, after expenses.” The company “explicitly discourages employees from comparing [Uber’s] estimates to the minimum wage.” Here‘s an anecdotal account from a taxi driver who previously drove for Uber and Lyft, reporting that his income from ride-share apps was dismal compared to his taxi income. That doesn’t in itself prove anything, but it does mean we shouldn’t allow Uber to treat this as an obviously settled question, and it would be useful to have a better idea of why 96% of Uber drivers quit after less than a year. (Hall and Krueger suggest that this is simply because Uber is used as a “bridge” between jobs, though one wonders why those people even need to find another job if driving is as lucrative as the company says.)

In fact, we know little about the reality of the gig economy, though at least one set of researchers has developed a calculator to estimate what Uber drivers might expect to earn after expenses. The lack of definitive answers shows why this needs to be looked at seriously, in research that isn’t just company propaganda or back-of-the-envelope calculations from industry-sponsored bloggers. There are important reasons to figure out what Uber drivers really earn. The debate over regulating Uber/Lyft in many cities has been framed as pitting innovation/disruption against “taxi company profits.” Language about monopolies and cartels is ubiquitous. Taxi companies just don’t want competition, they want to keep fares high and squeeze money out of consumers. Framed that way, regardless of what you think of Uber, it’s difficult to see any good argument for the taxi lobby’s desire to put restrictions on ride-share apps and keep local licensing requirements in place. But if Uber is creating a “race to the bottom” effect that is causing considerable economic hardship to people who have spent decades driving for a living, we’re in a different situation. And since Uber fares are obviously far lower than taxi fares, with drivers paying more in expenses, the idea that it’s going to end up being better for workers is… counterintuitive to say the least. The whole success of the gig economy is based on the fact that it’s all cheap, and it’s not just cheap because it’s disrupting cartels, but because it cuts costs that companies would previously have had to pay. (Such as, for example, driver safety training, which ride-share apps don’t offer because they’re worried it will lead the government to conclude that drivers are actually employees who deserve benefits. Being a taxi driver is actually one of the most dangerous jobs in America: you’re more likely to be murdered as a taxi driver than doing any other job, and risks of car accidents and robberies also drive up the workplace injury rate.)

It would be easy, of course, to fix everything. We don’t need to reclassify “driver-partners” as actual employees instead of contractors, or even try to get them unionized. What we really need is a rideshare system owned and operated by drivers themselves. Libertarians have snarkily pointed out that the “sharing economy allows workers to own the means of production,” which is every socialist’s dream. Why doesn’t the left love this? Well, we know exactly why: because it’s not about “owning the tools you use at your job” it’s about not being exploited by your employer, i.e. not having to turn over a substantial portion of your earnings to them merely because you happen to be dependent on them. But there is a point here: ride-sharing can be good for workers, if corporate profit is out of the picture. We need an alternative, not for profit service that actually operates in a way that servers its workers.

At the beginning of last month, Doug Schifter, a New York City livery driver, committed suicide on the steps of City Hall with a shotgun. Schifter left a note, explaining that he was taking his life because conditions for him as a driver had become impossible. With the industry collapsing, he had lost his health insurance, was deep in debt, and said he was working up to 100 hour weeks. The New York Times described Schifter as a “casualty of the gig economy,” a man who had driven five million miles in his career only to see everything evaporate.” (Bill deBlasio apparently didn’t accept Schifter’s stated motivation for his act, saying it must have been due to an “underlying mental health issue.”) This situation is happening to career drivers across the country. Uber “devastated” the taxi industry in cities like Los Angeles. The executive director of the New York Taxi Workers Alliance said that never she had “never seen anything like the despair she was witnessing now — the bankruptcies, foreclosures and eviction notices plaguing drivers who were calling her with questions about how to navigate homelessness and paralyzing depression.”

Before Doug Schifter took his own life, he wrote:

We are not a government of the People, by the People and for the People any more. We are turning into a Government of the People, by the Corporation for the Rich. People are becoming enslaved and destroyed by politicians and companies with the aid of the rich and the corporations they control… Forget about great country….we are not even good people if we do not care for others in need. Most of the civilized world cares for their own except here where republicans who do not need more money to survive seek to get all they can get from companies and the rich and then take away all they can from the people… I hope with the public sacrifice I make now that some attention to the plight of the drivers and the people will be done to save them and it will have not have been in vain… 

In Tunisia, when one fruit vendor committed suicide publicly after conditions had become intolerable, it sparked an uprising that toppled the existing regime. In America, when a livery driver takes his life, the country barely bats an eye. After all, Schifter will be only one of 40,000 people who commit suicide in our country this year. And we’ve decided to accept the ruin of people’s lives as the price of cheap fares.

This article originally mentioned an MIT study on Uber drivers’ pay that has been proven to be flawed and has been significantly revised. In the original reference, I counseled readers to be “skeptical” because the study’s results were so drastically different from prior findings. 

If you appreciate our work, please consider making a donation or purchasing a subscription. Current Affairs is not for profit and carries no outside advertising. We are an independent media institution funded entirely by subscribers and small donors, and we depend on you in order to continue to produce high-quality work.