Illustration by Angelika Manhart
I knew I wanted to be a scientist when…

Cab Psychology

By Giovanni Paci

“We don’t know what students do in the third year. But it works out. By the fourth year, they usually have a paper.”

Professor Davis’s remark kept resonating in my mind as I looked out towards Manhattan, cold and still in the distance. The living room was warm and light, but I felt anguished, unsettled. The spring semester in the fourth year of my Ph.D. had just started, and I didn’t know what I had done until then. Luna, my roommate’s little Havanese dog, was sleeping on the sofa next to me.

I had managed to turn my private life around, broke up with a girl who screamed at me all the time, and moved to a shared apartment on Bedford and North Seventh in Brooklyn. But my research was going nowhere. I had gotten access to a data set of New York City taxi transactions the previous September, and I had spent months trying to test my hypothesis/hunch that the tip amount paid to drivers of different ethnicities would reveal unconscious discriminatory tendencies. I had found no evidence of discrimination. At night, I was a confident young man roaming among the music venues of the Lower East Side. In the morning I was hopeless, a researcher wrestling with dull, repetitive failures.

I commuted to Columbia on the L and 1 trains every day. On the ride I would read Daniel Kahneman’s best-seller Thinking Fast and Slow. The book described the systematic errors we make in decisions. Our inaccurate beliefs, our emotional life, and our misunderstanding of probabilities have a powerful influence on our choices. Psychology had something to say about human behavior that my fellow economists didn’t acknowledge.

One week before my mid-February seminar, the secretary of the Economic department sent me an email inquiring about the title for my presentation. I felt angry at myself, contemplating the prospect of yet another embarrassing deck of slides about the absence of results.

I reconsidered my sadness: those psychological theories I was reading about in Kahneman’s book made a lot of sense to me. Economists wanted to describe a world of rational decision-makers, who didn’t have regrets and complex emotions – people whose choices always reflected their true preferences and couldn’t be nudged by marketers. And then it struck me: there was a way to test one of Kahneman’s ideas using the taxi data. If his ideas were right, I could generate a graph showing that in hundreds of thousands of cases, passengers in New York city cabs had been nudged to tip drivers a higher amount with a simple change in the settings of the passenger screen — a fact that could not be explained by standard economic theory. If I was right, the graph would show a large jump in tips that coincided with the abrupt and arbitrary change in the screen’s settings.

I replied to the e-mail: my presentation was going to be about psychological effects in consumer behavior. I was confident that the data would support the theory.

I coded the graph. It would take about 24 hours for the computer to generate the results. I got worried that my hypothesis was false. In the afternoon, I went to 6th Avenue and bought some food at Sammy’s Noodle shop, then went back home to watch Deconstructing Harry for the fourth time. And then, just a minute after the movie had ended, I got the e-mail from the computer clusters: my results were ready. I felt my heart pumping. I opened the graph and suddenly felt my apprehension turn into elation: it looked just perfect. It had the jump that I had predicted right in the middle — a fact could only be explained by psychological theories. I had tested my first hypothesis in a scientific manner.

I jumped to the floor and hugged the dog, screaming with happiness. Luna got scared and started to pee, swinging her tail frantically. And so I was covered in pee, and happy, and tired from the screaming, and satisfied. A few years later that graph was the central result of my first published paper.


Giovanni is now a portfolio manager in the City of London, applying his quantitative models to financial markets. His Ph.D. papers were later published in the American Economic Journal and received media coverage from outlets like The Guardian, The New York Times, and Science. His thoughts sometimes go back to the Ph.D., and he smiles thinking about Luna, Brooklyn, his attempts to feel better about science, and that afternoon when he finally did.