BiblioTech Uri

BiblioTech
“The biggest mistake people make is thinking that incentives are just money”

Uri Gneezy, author of “Mixed Signals: How Incentives Really Work” joined CTech as part of CTech’s BiblioTech video series to share insights from his new book


Uri Gneezy, author of “Mixed Signals: How Incentives Really Work” with James Spiro
(Photo: Zoom/Sinay David)
Incentives are something that drives families, economies, companies, and societies. We are told as a child that if we behave well, we can have a candy bar or play outside with our friends. As employees, we may be incentivized with money or status, or find fulfillment from within by knowing we are doing something that adds meaning to our lives and our communities.
But what really drives these motivations in our minds, and how can we perform best when the incentives we receive are sending mixed signals?
Uri Gneezy is a Professor of Economics and Strategy and the Epstein/Atkinson Chair in Behavioral Economics at the University of California San Diego Rady School of Management. Before that, he was a faculty member at the University of Chicago, Technion, and Haifa. Uri’s research focuses on using both lab and field experiments to study how people react to incentives. He is the co-writer of The Why Axis, which discussed how using field experiments can improve our understanding of economic interactions in the real world.
Today, he has joined CTech to discuss his latest release, "Mixed Signals: How Incentives Really Work", which focuses on the signals that incentives send and how understanding and framing these signals can increase their effectiveness.
“Every signal [employers] give sends a signal and very often the signals are going to be mixed,” he told CTech. “For example, telling your employees that you really care about quality but then you incentivize in quantity. What you're going to get is that people are going to maximize quantity and the quality is going to suffer.”
While he admits that businesses do it because quantity is usually easier to measure since it's an objective KPI, quality is often subjective and can lead to confusion. “You should know that if you tell your employees they should maximize quality but then you incentivize quantity, then you will get lower quality. If that's what you want that's good, but that's a mixed signal.”
I would like to start by asking - what was my incentive to invite you here today and what was your incentive to write this book?
“My incentives are easy, I want to sell as many books as I can. Why did I write the book? That's a lot of intrinsic motivation because if you want to make money there are other ways that may be more efficient and it is a lot of work to write a book. It is some sort of intrinsic motivation to make me feel good about myself, it was definitely not just money but the feeling I got out of this. What are your incentives?”
My incentive is always to find people like you to talk to about interesting things like this. You mention financial incentives, but what are some of the other forms of incentives that you explore and what can we examine today?
“Kids can get incentives whereby if they don't do their homework they can’t go out and play - that’s also an incentive. Punishment is also a form of incentive. If you do a good job I can tell you that, which is an incentive and something that could motivate you. It’s definitely not just money, money is just the easiest in many aspects because it's easy to measure. I cannot measure if I tell you you did a great job, versus doing a good or fine job. It's harder to measure but with money, it is very easy. You know how to deal with it. Everything to me that can affect your behavior is an incentive.”
I understand why people would respond to a financial incentive itself since everyone wants a few more pennies in their pocket but what is the human reaction to incentives?
“If you offer me money or offer help to do something for money then it is clear why I am doing it. You can take a class in business school about incentives and you can hear from an economist who will tell you that you hate going to work and the only reason you go is to pay the rent and to survive.
“But take a class with the same type of psychologists that talk about fulfillment and feeling good about yourself going to solve things. To them, you're going to work not to pay the rent but because you want some kind of meaning. Those two are really strong. Personally, I go to work because they pay me and I need the money, but also I enjoy doing it and I get some satisfaction from it. If I’m successful, and if at the end of this interview you tell me I did a good job, I would feel good about it.
“There is a combination here, but what I am trying to do with the book is to look at these ways and to actually have them help each other instead of two conflicting views. Monitoring incentives can actually make you feel better about your job and make you feel good about it. Because the starting point is that every incentive contains some sort of signal.
“I don't know what you think about me or what my employer thinks about me, but if my Dean comes and tells me to do something and she’ll be happy with this, then that is a type of incentive. That is a signal. The incentive she will send will send me a signal about what she expects me to do and if I will fulfill what she expects me to do I will feel good about myself. That will make a signal. The biggest mistake people make is thinking that incentives are just the money they give you, and that’s it. It also contains information about the interaction.”
One of the things you touch on is the differences between social signaling and self-signaling. Can you explain what you meant by those terms?
“Imagine you see on a cold morning a neighbor going to the recycling center with a large bag of soda cans. She's walking with 100 soda cans to recycle them. You would think she is a good person, and that she is signaling that she is a good person. I would think that, too. She might also feel good about herself because she can say she could have trashed these cans but she chose to recycle them. The social signaling is what you think, and the self-signaling is her feeling good about what she's doing.
“Now, imagine the city decides to give you 10 cents for every soda can you recycle. It is exactly the same story: you see your neighbor and for 10 cents she went and collected all those cans. This time, you might think she's really cheap. You changed your perception about the social signal, what she sends is very different once there are incentives involved.
“It’s also self-signaling because she is going to feel different about herself since there is money. She will say she's not doing it just because she is a good person but because of the money as well. 10 cents per can completely change the meaning.”
It touches on an interesting idea of self-signaling and people doing things for themselves. At what point is that a virtue and at what point is that ego? If it's not just a financial reward, it might be only because of what others may think. Is that still an authentic form of incentive?
“You're touching on a deep philosophical question. Imagine I give $100 to a charity. I do it because it makes me feel good about myself. Is this altruism or not? That's a deep philosophical discussion that we will not resolve. As an economist, as someone who cares as long as you give the money to charity, that is a good outcome in my world. The charity will enjoy it and if you also enjoy it that's good for you.
“As an economist, I try to look at the situation in terms of the outcome - is the outcome a desirable one? That's great. If not, what can we do to change it?”
There's also a philosophy that if you give money anonymously that is the most virtuous form of charity because you don’t get any social signaling, you only get the self-signaling.
“Some philosophers think even if it's anonymous and you feel good when you give the money then it's not altruism.”
What other lessons are there from the book that you extract that could teach us from a business point of view? Think of the business or CEO or the folks who work at high-pressure companies: What can they glean from these kinds of insights?
“They should understand that every signal they give sends a signal and very often the signals are going to be mixed. I talk in the book about a few of those. For example, telling your employees that you really care about quality but then you incentivize in quantity. What you're going to get is that people are going to maximize quantity and the quality is going to suffer. Many businesses do it because quantity is usually easier to measure - it's an objective thing whereas quality is often subjective. But you should know that if you tell your employees they should maximize quality but then you incentivize quantity, then you will get lower quality. If that's what you want, that's good, but that's a mixed signal.
“Or perhaps telling your employees they should work in teams but then compensate them based on individual incentives. If I give you personal incentives, why would you work as part of a team? As a boss, you should decide what is more important: Do you want employees to put in a lot of individual effort and work or is this a job where you need a team? In any case, don't send a mixed signal.
“Another one is to say you care about ‘the long run’ but measure them based on quarterly earnings. That’s again a mixed signal. If I can tell you as the board we care about the long run, but in reality you know you will be measured based on very short-term incentives, you will react to this.
“There are two lessons that are very important: The first is to use common sense. Just think about the problem, the incentive, and the message you are sending. And the second is that you should always test what you do. Do some sort of A/B testing, and see whether your intuition about how your incentives will work is actually correct. In many cases you will have strong intuition, you go to run the A/B test and you find out your intuition was wrong.”
What advice would you give to those who are receiving these mixed signals? From an employer or board member who might not be making much linear sense - how can people interpret and overcome some of those challenges?
"In some cases, because I'm an annoying person who cares so much about incentives, I can tell you ‘Look, that doesn't make sense! You're telling us you want us to publish good papers and then you're rewarding us per paper!’ You can tell your employer: ‘Look, I’m confused - what do you want me to do? Do you want me to really care about customers?’
“Imagine you work at a care center and you are paid per call you answer. It's clear you will be very quick, but the quality will go down. People will call again later or you will not solve the problems. As an employee at a place like this, I would ask the boss: 'What do you care about? I'm confused.' In many cases, we don’t understand we are confused. Most people who don't think about incentives as their day job don't understand that they got mixed signals.”
Another aspect of your book you discuss is the 'Pay to Quit' strategy. What is that?
“If you ask people if they like where they work most of the people will say yes. And it is important because if you don't say yes then your value to the company is lower. But if I come to work because I have to and I hate my work, I'm not going to be such a good worker.
“If I can measure output easily, that helps with the problem. But in many cases, with many jobs out there it is hard to measure the output, especially in the short run. You need to have motivated employees and if the employee doesn’t like where they work the value to the company will be lower but they don't want to lose their job.
“Now, instead of asking them, you can use money as a way to figure out what they really care about. Imagine I come and say ‘James, we like your work and want you to stay. But in case you’re not very happy and you want to leave, we can give you $10,000 and you can leave and we can depart as friends and appreciate the time. We hope you won't take it but you can.’
“This way, you can select and weed out the people who are not very happy with their jobs. Another advantage is that if you decide to stay, then you need to convince yourself that you didn't make a mistake - you just gave up $10,000, and you’ll need to justify it to yourself!"
When you test human reaction to incentives, did you notice any difference between males and females, or different age groups?
“In general, there are definitely differences between people. In Israel, people are competitive and aggressive, and then you go to Thailand where being aggressive and competitive is considered bad. Maybe in some places, you want to have more competition, and in others, you don't, depending on the culture. That could be even within a city, within cities the culture is not just between places like San Diego and Tel Aviv, it is also between professors, taxi drivers, doctors, or lawyers. Every group has a different kind of culture and we should adopt incentives for this.
“If I curse during negotiations between unions, maybe that would be ok. But if I did it with teachers that would be really bad. You need to know the culture and how aggressive or competitive your incentives should be. How much do people care about now versus later?
“There are also gender differences. We know that women are more risk-averse, meaning the more risky your incentives are, they might work better on men than women. In general, women also tend to be less competitive so the more competitive the incentives, the more men will have the job. There are individual differences, for sure.”
How can we equalize that, or is it something that even should be equalized?
“That's my new book. Think about gender differences: the workplace today was created by men, for men, sometime after World War II. And many of the things are just not needed but favored to men. Think about the temperature in the room, we have higher metabolisms than women, and we like the temperature to be a bit lower. If you go to offices then very often the temperature is such that men feel fine and women feel freezing. That’s not about incentives but that’s something that could explain other things.
“For example, my daughter is a software engineer and the job interview she had was half an hour of Zoom with three people watching her try to solve a problem with the code. She got the job, but the job itself had nothing to do with time pressure or people staring at her.
“The point is that you should not make the selection process with an added time pressure or more stressful than the job itself. If you're trying to be a fighter pilot, you need to perform really well under pressure. Software engineers usually don't. Make sure the incentives that you give during the interview are not different from the job itself.”