378 coffees paid forward, and why generosity shrinks along the way

A Florida coffee chain reached 378 cars. Experiments show strangers pass on fairness faithfully and greed readily, while generosity arrives as mere fairness.

At seven in the morning on a Wednesday in August 2014, a woman at a Starbucks drive-through in St. Petersburg, Florida, paid for her iced coffee and for the order of the car behind her. That driver did the same. So did the next one. The staff started keeping a tally, and eleven hours later it stood at 378.

The chain broke at six in the evening. A customer was told her drink had already been paid for and asked whether she wanted to continue. She paid for her own $2.25 drink and drove off. The barista told the Tampa Bay Times he didn't think she had understood the idea. The next day customers started a new chain, and a local blogger drove in on purpose to end it, arguing that the chain ran on guilt rather than generosity.

The research on paying it forward says something that neither the 378 nor the blogger quite captures.

What travels down a chain of strangers

Kurt Gray, Adrian Ward and Michael Norton ran the cleanest version of the question I know of, published in the Journal of Experimental Psychology: General in 2014. In their first experiment they recruited 100 people at subway stations and tourist spots in a city in the north-eastern United States and sat each of them alone on a bench with an envelope. Inside was what an anonymous previous participant had left them from a $6 pot: nothing, half, or all of it. Then they got a fresh $6 to split with an anonymous stranger who would come after them.

People who had been treated fairly passed fairness on almost exactly: they had received $3 and gave, on average, only 38 cents less. People who had been treated greedily passed on greed, though softened a little. And the people who had received the full $6? They did not pass on $6. They passed on about $3.70, a little over half, and statistically no different from the people who had received an even split. Generosity arrived and left as fairness.

The authors ran four more studies, swapping money for chores. In some, participants received a batch of pleasant tasks (rating cartoons, free-associating words) or unpleasant ones (circling vowels in dense Italian prose) and then decided what to leave for the next person. The pattern held. Their own summary is blunt: equality leads to equality and greed leads to greed, "but true generosity results only in a return to equality."

What drove it was bad feeling. In the fourth study, people who had been handed the dull tasks felt measurably more upset, and that feeling predicted how little they left for the next person. In the fifth, 165 online participants were given a short filler before splitting their own tasks. Some rated abstract art, others rated cartoons. Among those who had just been treated greedily, the ones who looked at abstract art left the next person an average of 0.37 good tasks out of two. The ones who looked at cartoons left 0.78. A small mood repair roughly doubled what they passed on. Among those who had been treated generously, the cartoons made no difference at all.

And yet the coffee chains happen

If generosity shrinks to fairness at each step, how do you get 378 cars?

A second paper, by Minah Jung, Leif Nelson, Ayelet Gneezy and Uri Gneezy, published in the Journal of Personality and Social Psychology the same year, gets closer to what happens at a drive-through. They took over the Cartoon Art Museum in San Francisco on its pay-what-you-wish days and a coffee stand at a farmers' market in Oakland. Some visitors were told simply to pay what they wanted. Others were told that someone earlier had already paid for them and they could now pay what they wanted for the next person. The money went to the same place either way.

People paid more when paying forward. At the museum it was $2.67 per person against $1.82. At the coffee stand, $2.33 a cup against $1.93. Then, in lab follow-ups, the researchers told participants exactly how much the previous person had paid, and the gap disappeared. They argue that when you don't know what others gave, you tend to assume they were kinder than they were, and you match the kindness you imagine.

That is exactly the setting at a drive-through window. You don't know how much the car ahead spent. You know only that they paid, that the barista is waiting, and that there is a car behind you. The matching the Gray study found, fair treatment passed on as fair treatment, is what keeps a coffee chain alive, because a coffee for a coffee is an even trade.

So was the blogger right?

Partly. The Jung paper shows that some of what looks like generosity at the window is people living up to a guess about others. Call that social pressure if you like. But the woman who paid her own $2.25 and left didn't harm anyone, and the blogger's point cuts both ways: a chain in which nobody feels free to stop is not a very generous thing.

What I take from the two papers won't warm anyone's heart, but it is useful. A single grand gesture probably won't echo down a line of strangers at full volume. It arrives at the next person as ordinary fairness. That is still a good outcome, since the same studies show fairness is the thing people pass on most faithfully. The part that deserves more attention is the other direction. Being short-changed makes people pass it on, and a small lift in mood can interrupt that.

Worth noting the limits. These were small experiments with small stakes, six dollars and a handful of chores, and the Gray study is one team's work that I haven't seen directly replicated. The authors themselves wonder whether a truly large act of generosity might behave differently. I haven't found anyone who has tested that with a bench and an envelope.

If you want to start a chain, being fair to whoever comes next may matter more than being spectacular. And if someone has just been rude to you, take the minute before you deal with the next person.

Sources: Gray, K., Ward, A. F., & Norton, M. I. (2014). Paying it forward: Generalized reciprocity and the limits of generosity. Journal of Experimental Psychology: General, 143(1), 247–254. Jung, M. H., Nelson, L. D., Gneezy, A., & Gneezy, U. (2014). Paying more when paying for others. Journal of Personality and Social Psychology, 107(3), 414–431. NBC News, "378 Customers Pay It Forward to Strangers at Florida Starbucks", August 2014, citing the Tampa Bay Times.