Groove Track: Why You Can’t Find a Cab on a Rainy Day!

The Cab Driver Study, officially known by its formal name of the “LABOR SUPPLY OF NEW YORK CITY CAB DRIVERS: ONE DAY AT A TIME,” is a seminal study in behavioral science.  Google shows that it has been cited almost 1,600 times and beyond that, the impact that it has had on the incentive industry is significant, changing the way that contests and incentives were thought of and designed. Published in May of 1997 in the Quarterly Journal of Economics, the study was written by Colin Camerer, Linda Babcock, George Loewenstein, and Richard Thaler.  

[Watch this Groove Track here]

After talking with the authors, their stories vary on how this study came to be. However, one thing they all agree on is that Colin Camerer had tons of data on New York City cab drivers in his office at Cal Tech.  It was a treasure trove that they thought might hold some interesting insights.  

And it did.   

As they sifted through the data they found, in Thaler’s words, “economic anomalies.” Economic anomalies occur the behaviors exhibited by people are different than what is predicted by economic theory. 

For instance, since cab drivers in New York City had a fixed expense that they needed to cover, such as renting their cabs, economic theory predicted that they would work extra long hours when their earnings were going through the roof (e.g., when it was raining or the subway was closed, or there was a big convention in town and tons of people were wanting to take cabs).

However, in shifting through the data, the researchers found that wasn’t the case.  This is how George Loewenstein describes it in our interview with him:  “…on days when cab drivers were making a lot of money and when conventional economic logic would suggest that they should be working long hours, instead, they were quitting early.”

Since economists believe that workers ought to maximize their revenue opportunities and particularly on days when they are making a lot of money, this shouldn’t be happening.  In fact, the reverse should be happening, cab drivers should be working longer hours on those days.  This was the economic anomaly that was the basis of the paper.  “...on days when cab drivers were making a lot of money and when conventional economic logic would suggest that they should be working long hours, instead, they were quitting early.” George Loewenstein Click To Tweet

The Study 

This study is interesting because of the data that the team had to work with which included hand-written trip sheets that documented how many fares each driver had and at what time they picked up and dropped off fares.  From that, they were able to determine the number of fares each cab driver had, the time those fares took, the average fare, and their take-home pay (less tips).  In addition to the thousands of trip sheets from the Taxi and Limousine Commission, they gathered survey responses from more than a thousand drivers, as well as conducted interviews with cab owners and fleet managers

What about the weather?  We almost always talk about this study as being connected to work habits on sunny days versus rainy days. 

There are many articles, and we have done this too, that informally reference that cab drivers drove less on rainy days. However, the study did not actually compare driving habits to weather patterns.

It just happens to be part of the story of how Camerer, Babcock, Thaler and Loewenstein came to ask the question, “Why is it so hard to get a cab on a rainy day, when economists would predict there’d be enough supply to meet the increased demand?” 

The study never even talks about the difference between rainy and sunny days.  Although, that is the question that ultimately led the research being done. 

Study Findings

When analyzing the data, the team found that a large portion of cab drivers appeared to focus on a daily wage target or a narrow bracket for their wages, and once they hit that daily target, they stopped working.  From the trip sheets, the researchers found that cab drivers worked about 9.5 hours per day, averaged between 28 and 30 trips, and collected almost $17 per hour in revenues (excluding tips).

But the researchers also found that they worked fewer hours when they hit their daily earnings target faster.  In other words, their daily target income became their primary measure, not maximizing each day’s or each week’s income which is what classical economics would suggest. 

The findings suggest that the cab drivers are using the simple heuristic of average daily wage to manage their business, and because of that, they end up passing up high-revenue opportunities – like working a full shift on a day when they are raking in the fares. 

So while the weather is not looked at in the study, we know that there are many more customers wanting cabs on rainy days than there are on dry days. But because a cab driver could reach their daily targeted income earlier in the day, they’d stop working early, rather than do what economists believed they ought to do, which was work a full shift and really maximize their earnings.   

Thus, we can’t seem to hail a cab on a rainy day!  

Interestingly, they also found that the more experienced drivers, when compared to less experienced drivers, were less apt to knock off early on days when they had a ton of fares. To quote the study, drivers, “learn over time that driving more on high wage days and less on low wage days provides more income and more leisure.”

Criticism 

From time to time, the cab driver study has been under fire. Henry Farber wrote a scathing review in 2008 that indicated that drivers have both daily wage and daily-hours-worked targets, not just daily wage targets. His concern was that the original paper framed daily wage targets as the primary point of measure and his research suggested that some drivers relied on number-of-hours-worked, and some even had hourly wage targets, as well. There is some merit in this critique, but does not discredit the findings, that cab drivers did not optimize their earnings in those situations where economic theory would have predicted they would.  

Implications

 Many people, looking at it from the outside, judge this type of behavior badly. They think to themselves, “There’s NO WAY I would stop working early when I could really make a ton of money that day!” But that is not always true. 

The idea that motivation is a simple equation where the more I pay you, the more motivated you are, is false.  The opportunity to earn more does not always lead to greater effort or motivation.  

Here is the important lesson we want people to take away from this: we don’t always act in ways we think we should act in a situation. In fact, we often do things that go against our pre-stated intent.  It’s what we refer to as the Say-Do Gap in psychology.

And this concept, this idea of a say-do gap,  is common among people in a whole variety of settings. 

When we connect this to other studies, like in the world of incentives, we find that salespeople exhibit these same habits. They say that maximizing cash is the most important thing in their world, and that tell their bosses they’ll work harder for extra cash, but when presented with the chance to do just that, only a small percentage actually take advantage of it.  

In other words, they too act like cab drivers!  

 Salespeople may say they want to earn a lot of money, but they don’t always take advantage of those opportunities and are often setting their own daily wage or effort targets.

There are exceptions, but they’re typically the minority. So if you’re a sales manager or a leader in general and you’re thinking that you’re people are different, we’d like to caution you to avoid that lens. A few of them might be different, but definitely not all of them.  We all are anchored in on some performance target  – and that might not be the one that you think should be a priority.   

 

References:

Camerer, C., Babcock, L., Loewenstein, G., & Thaler, R. (1997). Labor Supply of New York City Cabdrivers: One Day at a Time*. The Quarterly Journal of Economics, 112(2), 407-441.  Link here

For more on incentives – see here