STACEY VANEK SMITH, HOST:
To err is human, but it is also corporate. Companies do their fair share of erring. Now, if you are a person and you mess up, you apologize. But what do you do if you're a company?
CARDIFF GARCIA, HOST:
Yeah. There are the large-scale apologies for let's call it wrongdoing, criminal or otherwise. But what about the more individual issues, like if a company gets your specific order wrong or sends you specifically a faulty product? Then what should a company do? That is the topic of a really interesting new paper by a group of economists on the economics of apologies.
VANEK SMITH: They partnered with Uber, the ride sharing-app, to try to figure out the most effective way for a company to apologize to a wronged customer. So they focused on customers who were dropped off late in the 95th percentile of late - so really late, like, not three minutes but, like, an hour. They looked at about a million and a half late drop-offs all across the country. And they experimented with the best way to make good with those customers.
GARCIA: This, I say unapologetically, is THE INDICATOR FROM PLANET MONEY. I'm Cardiff Garcia.
VANEK SMITH: Sorry, not sorry, I'm Stacey Vanek Smith. Today on the show, how to say you're sorry if you're a company. It turns out there is a lot of money riding on the right apology.
GARCIA: And on the wrong one, as it turns out.
(SOUNDBITE OF DROP ELECTRIC SONG, "WAKING UP TO THE FIRE")
GARCIA: So the economics of apologies - Uber did this large-scale study with a group of economists. And they looked at how to best deal with customers that had just had a bad experience. In this case, they were dropped off way later than the Uber app had estimated that they would be dropped off. One of those economists was Benjamin Ho, an associate professor of behavioral economics at Vassar College.
BENJAMIN HO: I've been thinking about apologies for a very long time. I was working on a presentation today, and I realized I made that slide in 2002. So the examples were all about, like, Bill Clinton apologizing and Enron apologizing. And I just thought, wow, I'm old.
VANEK SMITH: I remember those. Well, tell me about - like, why are apologies interesting to you?
HO: I think because the economy's based on trust. I think economists underappreciate this fact. And our traditional neoclassical models are all about sort of cold, calculating agents. But actually we trust a lot in the economy. Just, you know, we can't write a contract for everything. And so a lot relies on trust. And so basically, you know, when trust falls apart, an apology is what we use to sort of restore that relationship.
VANEK SMITH: You did this, like, really, really interesting study as a way to study apologies.
HO: So the study was partnered with Uber. So we look at - we focus on the 95 percentile of lateness. So the most extreme - most late rides, there's about a 5 percent chance that you never come back to Uber again. And so...
VANEK SMITH: That's pretty high, actually.
HO: Yes. So that's the - you know, so that's what Uber is worried about - about this chance of losing a customer. And the question is, how can we repair this relationship? How can we repair, say, the trust between Uber and its rider? We tested different kinds of apologies. And we found that the right kind of apology could basically increase Uber's revenues by 1 to 2 percent, but the wrong kind of apology could decrease Uber's revenues by 1 to 2 percent.
VANEK SMITH: So what were the apologies that really worked for Uber? Like, what was the magic?
HO: So what - we found that the magic was money basically.
VANEK SMITH: (Laughter) Really?
HO: Yeah, show me the money basically.
VANEK SMITH: Yeah, yeah, yeah.
HO: That - we tested, you know, apologies with or without a coupon. And we tested apologies that involved sort of commitments to do better or not. And we found basically the most effective apology, the ones that sort of increased revenues, was just a $5 coupon.
VANEK SMITH: Was it just a question of coupon, or did it matter what the message was?
HO: So we tested that as well. What we saw was that a commitment to do better in the future - it doesn't seem to matter much right away. But we tracked these riders for three months after the apology. And we found that basically two or three months later, the commitment to do better in the future actually backfired against the company in part because, you know, you might commit to do better in the future, and then the rider says, oh...
VANEK SMITH: Because they didn't do better.
HO: 'Cause then maybe it's - maybe for some of them, they had another bad experience. And then as a result, those people punished Uber more than the control group that got no apology at all. And three months down the line, coupon plus apology did worse than coupon alone.
VANEK SMITH: Did moneyless apologies have any effect?
HO: Yeah, the moneyless apologies had a chance of backfiring. It depended on the condition.
VANEK SMITH: Oh. Did they ever have a good effect?
HO: Not in our data. So...
VANEK SMITH: (Laughter).
HO: ...I think they temporarily had a good effect, right? But I think there's this idea that once you've apologized, we're going to hold you to a higher standard. And sort of by the end of the three months of observation, it seemed like they either had no effect or a negative effect.
VANEK SMITH: OK, so this is confusing to me because I've had some, like, bad experiences, like everybody else, with online companies. So there was, like, a food delivery service that I used, an online food delivery service. And I ordered dinner, and the dinner never came. And I waited for, like, two hours. And I was calling them. And I was so hangry. And they never apologized. They actually sent me a voucher.
VANEK SMITH: But there was no apology. And it made me so mad.
VANEK SMITH: So in this case, I mean, they did give me money.
VANEK SMITH: They gave me a $25 coupon.
VANEK SMITH: And I used it angrily.
VANEK SMITH: But like, they didn't say they were sorry.
VANEK SMITH: Like, that - am I, like, a different...
HO: No, no.
VANEK SMITH: ...Consumer than most people?
HO: No. I think it's - I think saying you're sorry definitely can help, right? And I think we found that especially in short run it definitely can help. It's just that the risk of saying you're sorry is that you may punish them more in the future, right? So say, you know, they said sorry now. And the next week, you know, you get another late ordered delivery. Then maybe that sorry might backfire against them.
VANEK SMITH: It is interesting that we expect remorse from a company. You know, you're ordering from, like, online. Like, it's, like, minimizing human interaction, yet it still seems like we still want the same level of trust and accountability.
HO: I think part of it is that our brain sort of works on scripts and heuristics. And so, you know, when we enter in a relationship with a company, we sort of - like, the same sort of scripts and heuristics trigger in our brain. And so because we expect remorse in real life, we sort of expect it from a company. And in some sense, like, you know, when we're using an app on the phone, the apology sort of pops up as, like, a notification, which looks a lot like a text message. And so almost - it almost sort of becomes like this personal relationship with that company.
I think also, though, the reason I started apologies was to think about actually the sort of apologies by sort of big, abstract entities like corporations or governments, right? Like, you know, do - you know, should the government of Japan apologize for World War II atrocities, or should - how should South Africa apologize, or how should....
VANEK SMITH: Right. Right.
HO: ...You know, sort of these sort of abstract entities apologize? And it really seems strange to ask, you know, what does it mean for a country to show remorse, or what does it mean for a corporation to show remorse? And, you know, while I was trying to do my research, to say well, you know, this script comes from this, like, mathematical model of signaling where - you know, where if these sort of - if these corporations sort of kept - you know, sort of pay this signal and pay this cost, then we will forgive them and use them more in the future. And it's all very rational. It's all based on costs and benefits. And therefore, it doesn't surprise me at all that corporations are using sort of these tools.
VANEK SMITH: Do you think there's a difference between - I mean, do you think this, like, carries over into personal relationships?
HO: I think...
VANEK SMITH: Be like, I'm so sorry that I, like - whatever - was late meeting you for dinner. Here's 10 bucks.
VANEK SMITH: (Laughter).
HO: Well, you know, here's a box of chocolates - right? - or here's...
VANEK SMITH: Oh, yeah.
HO: ...Here are some flowers.
VANEK SMITH: Oh, right, flowers and candy.
VANEK SMITH: Oh, this takes on a whole new significance. Has it changed the way, like, you apologize?
HO: (Laughter) I think people sort of always wind up asking me this question.
VANEK SMITH: Of course they do. I'm so curious.
HO: I think the main way it's changed is that because all my friends and my family know that I study apologies, then they're sort of, oh, I'm awfully skeptical whenever I apologize. So I have to, like, go an extra mile to sort of sound, like, earnest and sincere and remorseful because people are always just like, oh, you're just doing that because of your paper.
VANEK SMITH: This is game theory, isn't it? Are you just trying this out on me?
HO: That's - yeah, that's exactly what I hear from my wife all the time.
VANEK SMITH: (Laughter) Do you do flowers and candy?
HO: I do (laughter) even though I think she's a little tired of the flowers and candy. So...
(SOUNDBITE OF DROP ELECTRIC SONG, "WAKING UP TO THE FIRE")
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.