Are Uber Drivers Economically Rational?

The short answer is that those who will succeed in the business learn to be rational.

The longer answer, as the following article explains, is that many drivers who engage with Uber have limited financial expectations and stop driving when their modest goals are achieved.

That there is more of a learning curve than many drivers expect may explain high driver turnover. But it also underscores the advantages of the company's investment in self-driving car technology. JL

Noam Scheiber reports in the New York Times:

A substantial fraction of (drivers) come into the market with income targeting behavior. The behavior is then “rather quickly learned away in favor of more optimal decision making.” The Uber evidence highlights the idea that drivers get better at wringing more money out of the hours they work as they gain experience. It may be that drivers set goals (but) they may ignore the target when their hourly wage lurches beyond a certain amount.

The ride-hailing company Uber is best known for upending the taxi industry. Now it may be making waves in the economics profession as well.

For nearly 20 years, economists have been debating how cabdrivers decide when to call it a day. This may seem like a trivial question, but it is one that cuts to the heart of whether humans are fundamentally rational — in this case, whether they earn their incomes efficiently — as the discipline has traditionally assumed.

In one camp is a group of so-called behavioral economists who have found evidence that many taxi drivers work longer hours on days when business is slow and shorter hours when business is brisk — the opposite of what economic rationality, to say nothing of common sense, would seem to dictate.

In another camp is a group of more orthodox economists who argue that this perverse habit is largely an illusion in the eyes of certain researchers. Once you consult more precise numbers, they argue, you find that drivers typically work longer hours when it is in their financial interest to do so.

The question has implications for other workers, like farmers and small business owners. And it has taken on added importance as the so-called gig economy grows, leaving more people in the position of deciding how many tasks to perform each day, from furniture assembly to tagging photos.

So who is right? That’s where Uber comes in. When one of the company’s researchers, using its supremely detailed data on drivers’ work time and rides, waded into the debate with a paper this year, the results were intriguing.

Over all, there was little evidence that drivers were driving less when they could make more per hour than usual. But that was not true for a large portion of new drivers. Many of these drivers appeared to have an income goal in mind and stopped when they were near it, causing them to knock off sooner when their hourly wage was high and to work longer when their wage was low.

(The hourly wage of cabdrivers typically reflects how busy they are; the rate they can charge doesn’t change unexpectedly. For Uber drivers, the hourly wage reflects both busyness and rates, since Uber can increase prices when demand is high.)

“A substantial, although not most, fraction of partners do in fact come into the market with income targeting behavior,” the paper’s author, Michael Sheldon, an Uber data scientist, wrote. The behavior is then “rather quickly learned away in favor of more optimal decision making.”

In effect, Mr. Sheldon was saying, the generally rational beings that most economists presume to exist are made, not born — at least as far as their Uber driving is concerned.

The Uber evidence does highlight some points on which both sides of the debate already agreed, like the idea that drivers get better at wringing more money out of the hours they work as they gain experience.

“The learning effect doesn’t bother any principle of economics,” said Colin Camerer, an economics professor at the California Institute of Technology and the lead author of the pioneering study showing that many cabdrivers violate an economist’s idea of rationality.

Henry Farber, a Princeton economist and author of several studies affirming the traditional view, echoed this sentiment, saying even his papers suggest that beginners frequently do not drive enough when business is brisk. “New drivers who can’t figure it out leave the business,” he said. “The ones who stay tend to learn.”

Professor Camerer and Richard Thaler, an economics professor at the University of Chicago and a co-author of the original paper, also posit a middle-ground position for which there is broader support: It may be that cabdrivers and other workers set goals like an income target and just are not fanatical about hitting them. They may simply ignore the target when their hourly wage lurches beyond a certain amount in one direction or the other.

Still, the question of whether many drivers hew closely enough to an income target that they tend to work longer hours when driving pays less than usual, and shorter hours when it pays more, opened a crucial divide.

The conventional view is that while people may fall short of what economics textbooks recommend, sometimes far short when they are just starting off at farming or repairing refrigerators or driving a cab, they usually do not act in the opposite of their self-interest, which is what sticking with an income target would cause them to do.

The behavioralists, by contrast, have been much more open to this possibility, although they have moderated their views on the extent to which drivers do this.

By finding evidence that many people defer to such an income target, if only until they gain more experience, the Uber study appears to have brought Professor Camerer and his co-authors a small measure of vindication after years of skepticism from fellow economists.

(Other studies have affirmed the original paper’s finding, but many relied on relatively crude data and were thus also vulnerable to criticism.)

Professor Camerer noted that the Uber example is a special case that might not apply to all drivers, but that other drivers might be even more likely to fall back on an income target because they had less information about what they could earn if they kept driving.

Professor Farber, for his part, proclaimed the Uber finding “interesting” and said that “for the new drivers, it might mean that without information and a lot of experience, they use rules of thumb.”

Still, he said he did not find it particularly disturbing intellectually that many new Uber drivers would rely on counterproductive rules, and even less so if a portion of these drivers used the platform to earn extra income that they did not need to support themselves.

“I think I’m coming around to the belief that people tend to leave money on the table when there’s no question of survival, or market forces pushing them toward getting that money,” Professor Farber said. He wanted to withhold final judgment pending more data and analysis. Uber generally does not make its data publicly available for proprietary reasons.

As for Mr. Sheldon, the Uber paper’s author, he attributed his finding to the adventurous nature of many Uber drivers, who were open to running headlong into unfamiliar territory. It’s the sheer unfamiliarity of the Uber driving experience, he speculated, that may explain the initial bout of economically irrational behavior.

Mr. Sheldon was less open to the idea that people who did not depend on Uber for their livelihood helped account for his finding. So far as Uber can tell from other research, he said, those who drive irregularly respond more to fare increases than more regular drivers, at any level of earnings.

Whatever the case, the result seems to have one very obvious implication: Anyone trying to make it in the gig economy should probably pick a favorite platform or two and stick with them rather than constantly jump around from one type of gig to another.

“Greater experience is a win for the partners and the platform itself,” said Mr. Sheldon. “We definitely want to keep these individuals engaged with the platform until they learn how to do it the best way.”