Connect Behavioral and Experimental Economics to the Real World
SciEcon AMA Interview with Prof. Gary Charness
About Prof. Gary Charness
Introduction to Prof. Gary Charness:
“Gary is not built in one day”
Research: Gary Charness (born February 3, 1950) is a Professor of Economics and the Director of the Experimental and Behavioral Economics Laboratory in the Department of Economics at the University of California, Santa Barbara. Charness is an economist and social scientist, specializing in experimental and behavioral work; he is currently ranked 3rd in the world by RePEc in the field of experimental economics and has published over 100 academic articles, beginning at the age of 50
Innovations: Charness is a contributor to several areas of economic research, including social preferences, identity and group membership, communication and beliefs, behavioral interventions, group decision-making, social networks, gender, and individual decision-making. A centerpiece of his research has been to effect beneficial social outcomes in difficult economic environments. Charness’s work has been discussed and published in The New York Times and Science, as well as in other media.
Leadership: He is currently a main editor at Games and Economic Behavior (the leading journal in game theory) on the board of six journals (Quantitative Economics, , Experimental Economics, Journal of the Economic Science Association ), Journal of Behavioral Economics for Policy, and Review of Behavioral Economics.
AMA Interview Video
Watch our AMA interview for Prof. Gary Charness on our YouTube channel, SciEcon.
Question 1
Xinyu:
Your story is truly inspiring to all global scholars. Even though you started your academic career as a Ph.D. at 41 years old, you have made incredible achievements in game theory and experimental economics, which is far unattainable for others. Moreover, while many people find it very difficult to do interdisciplinary research, you have been doing it for a long time and your papers are published in top journals, receiving over 26,000 citations. Could you please tell us what your motivation is and how you achieved all of these?
Prof. Gary Charness:
I went to graduate school by accident. I’d been looking for something else to do until, at the age of 40, I saw an article in the San Francisco paper about someone at Stanford, who won a Nobel Prize in economics. I didn’t know the guy but I knew one of his colleagues who was mentioned in the article, he was one year ahead of me at the University of Michigan. He’s Paul Milgrom, a smart guy who also won the Nobel Prize (2020). He was my guide, a big guy at Stanford graduate school. Then I thought I could still apply, so I applied to Berkeley with my top scores transcript, but they rejected me. I was very unhappy because I thought it was age discrimination which was unfair, so I went to Berkeley and persuaded them to reconsider it. I told them I only applied here and I will still come even if you let me in at the last minute. Then they said they would get back to me, so I calmed down, and they called me after a couple of days and said they were going to put me on a short waiting list. They called me back a week later and said they found me a space. So my motivation was that my life experience makes me understand more about how people behave than those guys in graduate school, even though they are super smart in academics, they didn’t have much experience in the real world. I have been to a lot of places, so I have a deep understanding of the real world, which is a useful thing. Studying how people behave is interesting. I think I’m more like a social scientist who studies how people behave in economic environments. I think it is very difficult to start an academic career at my age. I think I’m the only person who had no articles published at the age of 50 but can make a statement. I was always inspired to see what I could do and I was always competitive. Now I have 103 articles. I think my motivation comes from I think what I’m doing is useful for people, I’m not doing it on my own. I don’t think many things are going to change in my career, so I don’t have to be strategic anymore. I am able to do something only because I want to do them, that is also why I am doing the things now. I’m interested in figuring out the way to test out how things work and how people behave.
Question 2
Zesen:
We find all your studies excel in practicality and your papers combine story-telling with research and practice. For example, your paper Understanding Social Preferences with Simple Tests introduces a two-person model for testing social preferences, which is much more direct and simpler than previous ones. We global scholars would like to learn from you. Do you have any suggestions?
Prof. Gary Charness:
My suggestion is to look at the issues that interest you most and go with what your heart says. I like to do exploratory research with diversification. I have one paper on mindfulness training that looked for ways to understand social preferences and improve behavior. I think the main contribution of that paper is the methodology that can turn the model of economic environments into a really simple two-player game, however, the understanding of how people behave behind the game is not simple. I like the easy papers but they cannot be published in top journals, and working hard with deep work taught me how to write papers for journals. A decade ago, I was much more popular in the top journals, but now I still like exploratory research which I think is possibly useful. Here is an example I did recently called risk elicitation, it is about the decision-making of investment and lotteries, and how well people understand their decision. I think it is interesting. I found different conflicting results and I also think we’re fooling ourselves to think that we can elicit things meaningfully without giving people more practice with the mechanisms.
I think everything in life is about persuasion. The stories, ideas, and points in my research are really about persuasion. Persuasion happens everywhere. I persuaded my way into Berkeley. Don’t underestimate the power of persuasion, if you want something in life, just ask for it. I respect that a lot of people are reluctant to do these things. For me, persuasion is always difficult, but I always did it successfully. I always make my own life difficult, but that is just the way it is, and I don’t regret it. If I’m not doing something interesting, I should do something else. I think I should retire, but not for lack of energy. I’m never short of energy but have more motivation issues.
Question 3
Prof. Luyao Zhang:
Your research has covered abundant topics in game theory and experimental economics, from agent motivation [1] to incentives [3][4], and from cooperation [6] to trust [2]. What inspires you to come up with seminal ideas arranging from this wide range of topics?
Prof. Gary Charness:
I’m interested in a lot of different things, and I don’t know if I can make my research better if I stay focused on a specific topic. I think it is a good strategy to be focused, but that means you can only produce limited papers. I’m prolific so I can manage to juggle several different veins of research and topics. I’m a fast writer, so I can always be able to process many projects at the same time. My advice would be to dabble in different things but keep one thing specialized. But I do not follow it. I’m too interested in different things.
Question 4
Xinyu:
I’m (Xinyu Tian) currently doing research with Prof. Luyao Zhang about cooperative AI, which integrates game theory, agent cooperation, and computer science. As an expert in game theory, do you think there are any unsolved open questions at the intersection of trust and cooperation and game theory?
Prof. Gary Charness:
We still don’t understand how it works. Traditionally, you can understand people’s behavior if you look at their beliefs, but you just pushed it back one level. Where do those beliefs come from? If you are trying to be less or more persuasive, that means you are changing your beliefs more. It is something unsolved, and that is game theory. It is similar to the Traveler’s Dilemma. There is some lost luggage, and they have a value of between 180 and 300. You and somebody else can name a price value and they will pay the lower one of the two values. If somebody’s is higher than yours, they have to pay a fine to you and the fine in one case is 5 and in the other case, it’s 180. So the fine is used for hunting. Nash equilibrium says you always want to undercut the other person and unravel down. When the fine is 180, you will get 180. Because if you don’t say 180 and the other person says 180 or anything lower than you, you will end up with nothing. But if the fine is only five, you can gain a lot if you can coordinate. There is some trickiness. But how does it work? If people have beliefs about how people are going to behave when they’re considering this situation in game theory, I’m sure you’re going to get into networking, which is bound up with everything including cryptocurrencies and all sorts of trading issues. Because the network is a more realistic representation of how people are in this world. It’s not that everybody interacts with everybody. It’s specialized. In terms of game theory, they had more and more refinements, the last one being “divinity”. But it is not beneficial, because people don’t behave that way. So, I think it’s still trying to understand how people will behave and economic environments. We’ve made progress, but the danger to me is what I see happening now that I don’t like. The top journals and the fancier schools like papers that are sophisticated and rich in details, but they are less concerned about the content, and I think we need to think more about the content than these details. I hope they change their tastes, but tastes come and go. I think where we have to go is not in getting more tight little theories, but in trying to figure out difficult questions. Can you tell if someone is looking at you from behind, and can you tell if someone’s staring at you? I think you can. There are a lot of things we don’t understand. For some of them, we can make inroads but I guess that’s pretty vague. When I look at the game theory itself, I don’t see any great advances that need to be done in terms of game theory itself. But usually, I think game theory is pretty well developed, and I think where we can make progress is to understand how people actually use the principles. At least that’s what I hope because that’s what I can understand and I don’t know currently. If you’re in computers, or engineering, or doing with hardware or developing the algorithms, maybe that is the best thing that can be done, but I wouldn’t know what those would be so I can’t comment on that.
Question 5
Zesen:
What do you think about the subjects not making decisions according to utility in some experiments? For example, one of the papers, Lies in Disguise — An Experimental Study on Cheating in the book The Arts of Experimental Economics shows that even though people know the reward mechanism in the experiment, some of them still do not behave to pick the highest level of rewards. Would it be because of mistrust in the experimenters?
Prof. Gary Charness:
They are maximizing their utility. It is just their rewards, not their utility. You may think you are going to be suspected if you report the highest numbers. People may not do the things that are maximizing their utility, and people might make mistakes when they try to maximize their utility. I would rather believe they’re maximizing their utility than making mistakes. We see people sacrificing money all the time in different situations, however, sacrificing utility and sacrificing money are not the same thing. They sacrifice money to increase the total utility. That is what trust is, you have to take a risk of loss for everyone to benefit.
Question 6
Zesen:
This question is related to the last one. Blockchain, as a decentralized trust machine, uses consensus protocols to maintain trust among participating parties [5]. Do you plan to do a blockchain experiment in the lab?
Prof. Gary Charness:
It is still in the discussion stages. I have an idea but I’m also curious about it. I’m also happy to try it out. Blockchain seems like something going to be a main force in the future. I’d like to see how it works for people, and how people are going to deal with it.
Question 7
Xinyu:
We have a Metaversity project that aims to develop a platform for experimental economics (for example, an App like Otree) on the blockchain. Internet Computer is currently the fastest and most scalable general-purpose blockchain run by DFINITY Foundation. We plan to build two canisters, an autonomous one that is hardcoded on blockchain and a human-controlled one that can be changed in the experiment. The canisters are used to test if the problem (in questions 5 and 6) is caused by mistrust in the experiment designer. What do you think about this plan? Do you think there are any problems or possible developments?
Prof. Gary Charness:
In the blockchain, you’re still dealing with strangers in the real world. There’s always the possibility that someone could still steal money from you. You will have fewer issues of mistrust in a laboratory environment at university, which is not the same as the whole society. Mistrust always exists in society. It can be minimized by making the platform as transparent as possible. One efficient way to protect everyone is to let the mistrusting person leave the system. People will be willing to expel such a person because the mistrusting person is a potential threat to everyone in the system.
Question 8
Prof. Luyao Zhang:
The study of contract theory, Nobel Prize in Economics 2016, was highly valued as it contributes to many real-life problems, and your work in contract design is of great significance, especially as we now research and innovate on the blockchain. Your paper Promises in Contract Design highlights the application of workers’ promises in designing contracts, which can be very inspiring in smart contract design on the blockchain. The application of smart contracts can be a double-edged sword for being tamper-proof. On the one hand, the hardcoded program is inflexible in cases of an incomplete contract with unforeseen contingencies [9]; on the other hand, smart contracts do not depend on a trusted third party for the same reason of being tamper-proof. Do you have any ideas about applying your findings in contract design to smart contract design on blockchains?
Prof. Gary Charness:
This is not an easy problem unless there is a way to get some sort of a timestamp on the transaction before it happens. Whoever does the registration could always steal the idea, so the only way is that you can directly register yourself. It is not only about blockchain but also about the general idea. This is always the problem when you are buying a market order. There is always someone who can take advantage of you. Maybe there is a way to institute the order, and the order can be used to register the priority. However, it is not that bad to let someone take advantage of me in the market order.
Acknowledgment
Interviewee: Prof. Gary Charness
Interviewers: Prof. Luyao Zhang, Xinyu Tian, Zesen Zhuang
Executive Editors: Ray Zhu, Haoyang Yu
Chief Editor: Prof. Luyao Zhang
