The (Honest) Truth About Dishonesty

A Semioptimistic Ending

People Don’t Cheat Enough!

Throughout this book, we’ve seen that honesty and dishonesty are based on a mixture of two very different types of motivation. On the one hand, we want to benefit from cheating (this is the rational economic motivation), while on the other, we want to be able to view ourselves as wonderful human beings (this is the psychological motivation). You might think that we can’t achieve both of these objectives at the same time—that we can’t have our cake and eat it too, so to speak—but the fudge factor theory we have developed in these pages suggests that our capacity for flexible reasoning and rationalization allows us to do just that. Basically, as long as we cheat just a little bit, we can have the cake and eat (some of) it too. We can reap some of the benefits of dishonesty while maintaining a positive image of ourselves.

As we’ve seen, certain forces—such as the amount of money we stand to gain and the probability of being caught—influence human beings surprisingly less than one might think. And at the same time other forces influence us more than we might expect: moral reminders, distance from money, conflicts of interest, depletion, counterfeits, reminders of our fabricated achievements, creativity, witnessing others’ dishonest acts, caring about others on our team, and so on.

ALTHOUGH THE FOCUS of the various experiments presented here was on dishonesty, it is also important to remember that most of the participants in our experiments were nice people from good universities who will likely attain positions of some power and influence later on in life. They were not the kind of people one typically associates with cheating. In fact, they were just like you, me, and most of the people on this planet, which means that all of us are perfectly capable of cheating a little bit.

Though that may sound pessimistic, the half-full part of the story is that human beings are, by and large, more moral than standard economic theory predicts. In fact, seen from a purely rational (SMORC) perspective, we humans don’t cheat nearly enough. Consider how many times in the last few days you’ve had the opportunity to cheat without getting caught. Perhaps a colleague left her purse on her desk while she was away for a long meeting. Maybe a stranger in a coffee shop asked you to watch her laptop while she went to the restroom. Maybe a grocery clerk missed an item in your cart or you passed an unlocked bicycle on an empty street. In any of those situations, the SMORC thing to do would be to take the money, laptop, or bike or not mention the missed item. Yet we pass up the vast majority of these opportunities every day without thinking that we should take them. This means that we’re off to a good start in our effort to improve our moral fiber.

What About “Real” Criminals?

Across all of our experiments we’ve tested thousands of people, and from time to time, we did see aggressive cheaters who keep as much money as possible. In the matrix experiment, for example, we have never seen anyone claim to solve eighteen or nineteen out of the twenty matrices. But once in a while, a participant claimed to have solved all twenty matrices correctly. These are the people who, having made a cost-benefit analysis, decided to get away with as much money as possible. Fortunately, we didn’t encounter many of those folks, and because they seemed to be the exception and not the rule, we lost only a few hundred dollars to them. (Not exactly thrilling, but not too bad.) At the same time, we had thousands and thousands of participants who cheated by “just” a few matrices, but because there were so many of them, we lost thousands and thousands of dollars to them—much, much more than we lost to the aggressive cheaters.

I suspect that in terms of my financial losses to the aggressive and to the small cheaters, our experiments are indicative of dishonesty in society at large. Very few people steal to a maximal degree. But many good people cheat just a little here and there by rounding up their billable hours, claiming higher losses on their insurance claims, recommending unnecessary treatments, and so on. Companies also find many ways to cheat a little bit. Think about credit card companies that raise interest rates ever so slightly for no apparent reason and invent all kinds of hidden fees and penalties (which are often referred to, within companies, as “revenue enhancements”). Think about banks that slow down check processing so that they can hold on to our money for an extra day or two or charge exorbitant fees for overdraft protection and for using ATMs. All of this means that although it is obviously important to pay attention to flagrant misbehaviors, it is probably even more important to discourage the small and more ubiquitous forms of dishonesty—the misbehaviors that affect all of us most of the time—both as perpetrators and as victims.

A Word About Cultural Differences

I travel a lot, which means that I get to meet people from all over the world, and when I do, I often ask them about honesty and morality in their countries. As a result, I’m beginning to understand how cultural differences—whether regional, national, or corporate—contribute to dishonesty.

If you grew up outside the United States, think about this for a minute: do people from your home country cheat more or less than Americans do? After asking many people from various countries this question, I’ve discovered that people have very strong beliefs about cheating in their own countries, and most believe that people in their home country cheat more than Americans do (with the somewhat predictable exception of people from Canada and the Nordic countries).

Understanding that these are only subjective impressions, I was curious to see whether there really was something to them. So I decided to test some of these cultural perceptions more directly. In order to explore cultural differences, we first had to come up with a way to equate the financial incentives across the various locations. If we always paid, for example, an amount equivalent to $1 for a correctly solved question, this would range from being a very high payment in some places to a rather low one in others. Our first idea of how to equate the size of the incentives was to use a product that would be internationally recognized, such as a McDonald’s hamburger. Following this approach, for each matrix solved correctly, participants could receive one-quarter of the cost of a McDonald’s hamburger in that location. (This approach assumed that the people setting prices at McDonald’s understand the economic buying power in each location and set their prices accordingly.)

In the end we decided on a related approach and used the “beer index.” We set up shop in local bars and paid participants one-quarter of the cost of a pint of beer for every matrix that they claimed to have solved. (To make sure that our participants were sober, we only approached bargoers as they were entering the bar.)

BECAUSE I GREW up in Israel, I especially wanted to see how Israelis measured up (I admit that I suspected that Israelis would cheat more than Americans). But as it turned out, our Israeli participants cheated in the matrix experiments just as much as the Americans. We decided to check other nationalities, too. Shirley Wang, one of my Chinese collaborators, was convinced that Chinese people would cheat more than Americans. But again, the Chinese showed the same levels of dishonesty. Francesca Gino, from Italy, was positive that Italians would cheat the most. “Come to Italy, and we will show you what cheating is all about,” she said in her fantastic accent. But she was proven wrong too. We discovered the same results in Turkey, Canada, and England. In fact, the amount of cheating seems to be equal in every country—at least in those we’ve tested so far.

How can we reconcile the fact that our experiments don’t show any real differences in dishonesty among various countries and cultures with the very strong personal conviction that people from different countries cheat to different degrees? And how can we reconcile the lack of differences we see in our results with the clear differences in corruption levels among countries, cultures, and continents? I think that both perspectives are correct. Our data reflect an important and real aspect of cheating, but so do cultural differences. Here’s why.

Our matrix test exists outside any cultural context. That is, it’s not an ingrained part of any social or cultural environment. Therefore, it tests the basic human capacity to be morally flexible and reframe situations and actions in ways that reflect positively on ourselves. Our daily activities, on the other hand, are entwined in a complex cultural context. This cultural context can influence dishonesty in two main ways: it can take particular activities and transition them into and out of the moral domain, and it can change the magnitude of the fudge factor that is considered acceptable for any particular domain.

Take plagiarism, for example. At American universities, plagiarism is taken very seriously, but in other cultures it is viewed as a kind of poker game between the students and faculty. In those cultures getting caught, rather than the act of cheating itself, is viewed negatively. Similarly, in some societies, different kinds of cheating—not paying taxes, having an affair, downloading software illegally, and running red lights when there is no traffic around—are frowned upon, while in other societies the same activities are viewed as neutral or even confer bragging rights.

Of course, there’s a great deal more to learn about the influence of culture on cheating, both in terms of the societal influences that help curb dishonesty and in terms of the social forces that make dishonesty and corruption more likely.

P.S. I SHOULD point out that throughout all of our cross-cultural experiments, there was one time we did find a difference. At some point Racheli Barkan and I carried out our experiment in a bar in Washington, D.C., where many congressional staffers gather. And we carried out the same experiment in a bar in New York City where many of the customers are Wall Street bankers. That was the one place where we found a cultural difference. Who do you think cheated more, the politicians or the bankers? I was certain that it was going to be the politicians, but our results showed the opposite: the bankers cheated about twice as much. (But before you begin suspecting your banker friends more and your politician friends less, you should take into account that the politicians we tested were junior politicians—mainly congressional staffers. So they had plenty of room for growth and development.)

CHEATING AND INFIDELITY

Of course, no book about cheating would be complete if it didn’t contain something about adultery and the kinds of complex and intricate subterfuges that extramarital relationships inspire. After all, in the popular vernacular, cheating is practically synonymous with infidelity.

In fact, infidelity can be considered one of the main sources of the world’s most dramatic entertainment. If modern-day adulterers such as Liz Taylor, Prince Charles, Tiger Woods, Eliot Spitzer, Arnold Schwarzenegger, and many others hadn’t cheated on their spouses, the tabloid magazine and various entertainment news outlets would probably go belly-up (so to speak).

In terms of the fudge factor theory, infidelity is most likely the prototypical illustration of all the characteristics of dishonesty that we have been talking about. To start with, it is the poster child (or at least one of them) of a behavior that does not stem from a cost-benefit analysis. I also suspect that the tendency toward infidelity depends to a great extent on being able to justify it to ourselves. Starting with one small action (maybe a kiss) is another force that can lead to deeper kinds of involvement over time. Being away from the usual day-to-day routine, for example on a tour or a set, where the social rules are not as clear, can further enhance the ability to self-justify infidelity. And creative people, such as actors, artists, and politicians—all known for a tendency to be unfaithful—are likely to be more adept at spinning stories about why it’s all right or even desirable for them to behave that way. And similar to other types of dishonesty, infidelity is influenced by the actions of those around us. Someone who has a lot of friends and family who have had affairs will likely be influenced by that exposure.

With all of this complexity, nuance, and social importance, you might wonder why there isn’t a chapter in this book about infidelity and why this rather fascinating topic is relegated to one small section. The problem is data. I generally like to stick to conclusions I can draw from experiments and data. Conducting experiments on infidelity would be nearly impossible, and the data by their very nature are difficult to estimate. This means that for now we are left to speculate—and only speculate—about infidelity.

What Should We Do Next?

So here we are, surrounded by dishonesty. As one Apoth E. Cary put it in 1873:

Swindle, swindle, everywhere,

Every shape and size;

Take the swindle out of a man,

And you’ve nothing left but lies.

Philanthropy is made to cover a fraud,

Charity keeps humbugs in tow;

And we’re swindled at home, swindled abroad,

And swindled wherever we go.

For the world is full of humbugs

Managed by dishonest men;

One moves on, another comes,

And we’re swindled again and again.

—APOTH E. CARY, “RECOLLECTIONS OF THE SWINDLE FAMILY”¹

As we have seen, we are all capable of cheating, and we are very adept at telling ourselves stories about why, in doing so, we are not dishonest or immoral. Even worse, we are prone to “catch” the cheating bug from other people, and once we start acting dishonestly, we are likely to continue misbehaving that way.

So what should we do about dishonesty? We recently experienced a tremendous financial crisis, which has provided an excellent opportunity to examine human failure and the role that irrationality plays in our lives and in society at large. In response to this man-made disaster, we’ve taken some steps toward coming to terms with some of our irrational tendencies, and we’ve begun reevaluating our approach to markets accordingly. The temple of rationality has been shaken, and with our improved understanding of irrationality we should be able to rethink and reinvent new kinds of structures that will ultimately help us avoid such crises in the future. If we don’t do this, it will have been a wasted crisis.

MEMENTO MORI

There are a lot of possible connections one can draw between Roman times and modern-day banking, but perhaps the most important of them is memento mori. At the peak of Rome’s power, Roman generals who had won significant victories marched through the middle of the city displaying their spoils. The marching generals wore purple-and-gold ceremonial robes, a crown of laurels, and red paint on their face as they were carried through the city on a throne. They were hailed, celebrated, and admired. But there was one more element to the ceremony: throughout the day a slave walked next to the general, and in order to prevent the victorious general from falling into hubris, the slave whispered repeatedly into his ear, “Memento mori,” which means “Remember your mortality.”

If I were in charge of developing a modern version of the phrase, I would probably pick “Remember your fallibility” or maybe “Remember your irrationality.” Whatever the phrase is, recognizing our shortcomings is a crucial first step on the path to making better decisions, creating better societies, and fixing our institutions.

THAT SAID, OUR next task is to try to figure out more effective and practical ways to combat dishonesty. Business schools include ethics classes in their curricula, companies make employees sit through seminars on the code of conduct, and governments have disclosure policies. Any casual observer of the state of dishonesty in the world will quickly realize that such measures don’t get the job done. And the research presented here suggests that such Band-Aid approaches are doomed to fail for the very simple reason that they don’t take into account the psychology of dishonesty. After all, every time policies or procedures are created to prevent cheating, they target a certain set of behaviors and motivations that need to change. And generally when interventions are set forth, they assume that the SMORC is at play. But as we have seen, this simple model has little to do with the driving forces behind cheating.

If we are really interested in curbing cheating, what interventions should we try? I hope it is clear by now that if we are to stand a chance of curbing dishonesty, we must start with an understanding of why people behave dishonestly in the first place. With this as a starting point, we can come up with more effective remedies. For example, based on our knowledge that people in general want to be honest but are also tempted to benefit from dishonesty, we could recommend reminders at the moment of temptation, which, as we’ve seen, are surprisingly effective. Similarly, understanding how conflicts of interest work and how deeply they influence us makes it clear that we need to avoid and regulate conflicts of interest to a much higher degree. We also need to understand the effects that the environment, as well as mental and physical depletion, plays in dishonesty. And of course, once we understand the social infectiousness of dishonesty, we could take a cue from the Broken Windows Theory to combat the social contagion of cheating.

INTERESTINGLY, WE ALREADY have many social mechanisms in place that seem to be designed specifically for resetting our moral compass and overcoming the “what-the-hell” effect. Such resetting rituals—ranging from the Catholic confession to Yom Kippur, and Ramadan to the weekly Sabbath—all present us with opportunities to collect ourselves, stop the deterioration, and turn a new page. (For the nonreligious, think of New Year’s resolutions, birthdays, changes of job, and romantic breakups as “resetting” opportunities.) We have recently started carrying out basic experiments on the effectiveness of these types of resetting approaches (using a nonreligious version of the Catholic confession), and so far it seems that they can rather successfully reverse the what-the-hell effect.

From the social science perspective, religion has evolved in ways that can help society counteract potentially destructive tendencies, including the tendency to be dishonest. Religion and religious rituals remind people of their obligations to be moral in various ways; recall, for example, the Jewish man with the tzitzit from chapter 2 (“Fun with the Fudge Factor”). Muslims use beads called tasbih or misbaha on which they recount the ninety-nine names of God several times a day. There’s also daily prayer and the confessional prayer (“Forgive me, Father, for I have sinned”), the practice of prayaschitta in Hinduism, and countless other religious reminders that work very much as the Ten Commandments did in our experiments.

To the extent that such approaches are useful, we might think about creating related (albeit nonreligious) mechanisms in business and politics. Maybe we should get our public servants and businesspeople to take an oath, use a code of ethics, or even ask for forgiveness from time to time. Perhaps such secular versions of repentance and appeal for forgiveness would help potential cheaters pay attention to their own actions, turn a new page, and by doing so increase their moral adherence.

ONE OF THE more intriguing forms of resetting ceremonies is the purification rituals that certain religious sects practice. One such group is Opus Dei, a secretive Catholic society, in which members flagellate themselves with cattail whips. I don’t remember exactly how we started discussing Opus Dei, but, at some point Yoel Inbar (a professor at Tilburg University), David Pizarro and Tom Gilovich (both from Cornell University), and I wondered if self-flagellation and similar behaviors capture a basic human desire for self-cleansing. Can the feeling of having done something wrong be erased by self-punishment? Can self-inflicted pain help us ask for forgiveness and start anew?

Following the physically painful approach of Opus Dei, we decided to conduct an experiment using a more modern and less bloody version of cattail whips—so we picked mildly painful electric shocks as our experimental material. Once participants came to the lab at Cornell University, we asked some of them to write about a past experience that made them feel guilty; we asked other participants to write about a past experience that made them feel sad (a negative emotion but not related to guilt); and we asked a third group to write about an experience that made them feel neither good nor bad. After they reflected on one of these three types of experiences, we asked the participants to take part in “another” experiment involving self-administered electrical shocks.

In this next phase of the experiment, we connected the participant’s wrist to a shock-generating machine. Once the connection was secure, we showed the participants how to set the level of the electrical shock and which button to press to give themselves the painful jolt. We set the machine to the lowest possible level of shock and asked participants to press the switch, increase the level of the shock, press the switch, increase the level of the shock, press the switch, and so on until they could no longer tolerate the intensity of the shock.

We really aren’t as sadistic as it might sound, but we wanted to see how far participants would push themselves on the pain scale and to what extent their level of self-administered pain would depend on the experimental condition they were in. Most important, we wanted to see whether being reminded of a guilt-related past experience would cause our participants to cleanse themselves by seeking more pain. As it turned out, in the neutral and sad conditions, the degree of self-inflicted pain was similar and rather low, which means that negative emotions by themselves do not create a desire for self-inflicted pain. However, those in the guilty condition were far more disposed to self-administering higher levels of shocks.

As difficult as it might be to appreciate this experimental support for the practice of Opus Dei, the results suggest that purification through the pain of self-flagellation might tap into a basic way we deal with feelings of guilt. Perhaps recognizing our mistakes, admitting them, and adding some form of physical punishment is a good recipe for asking forgiveness and opening a new page. Now, I am not recommending that we adopt this approach just yet, but I can think of some politicians and businessmen whom I would not mind trying it out on—just to see if it works.

A MORE SECULAR (and more elegant) example of resetting was told to me by a woman I met at a conference a few years ago. The woman’s sister lived in South America, and one day the sister realized that her maid had been stealing a little bit of meat from the freezer every few days. The sister didn’t mind too much (other than the fact that sometimes she didn’t have enough meat to make dinner, which became rather frustrating), but she clearly needed to do something about it. The first part of her solution was to put a lock on the freezer. Then the sister told her maid that she suspected that some of the people who were working at the house from time to time had been taking some meat from the freezer, so she wanted only the two of them to have keys. She also gave her maid a small financial promotion for the added responsibility. With the new role, the new rules, and the added control, the stealing ceased.

I think this approach worked for a number of reasons. I suspect that the maid’s habit of stealing developed much like the cheating we’ve been discussing. Perhaps it began with a single small action (“I’ll just take a little bit of meat while I’m cleaning up”), but having stolen once, it became much easier to continue doing so. By locking the freezer and giving the maid an additional responsibility, the sister offered the maid a way to reset her honesty level. I also think that trusting the maid with the key was an important element in changing her view on stealing meat and in establishing the social norm of honesty in that household. On top of that, now that a key was needed to open the freezer, any act of stealing would have to be more deliberate, more intentional, and far more difficult to self-justify. That is not unlike what happened when we forced participants to deliberately move the mouse to the bottom of the computer screen to reveal an answer key (as we saw in chapter 6, “Cheating Ourselves”).

The point is that the more we develop and adopt such mechanisms, the more we will be able to curb dishonesty. It is not always going to be simple, but it is possible.

IT’S IMPORTANT TO note that creating an endpoint and the opportunity for a new beginning can take place on a broader social scale. The Truth and Reconciliation Commission in South Africa is an example of this kind of process. The purpose of this courtlike commission was to enable the transition from the apartheid government, which had sharply oppressed the vast majority of South Africans for decades, to a new beginning and to democracy. Similar to other methods of stopping negative behavior, pausing, and starting again, the goal of the commission was reconciliation, not retribution. I’m sure that no one would claim that the commission erased all memories and remnants of the apartheid era or that anything as deeply scarring as apartheid could ever be forgotten or fully healed. But it remains an important example of how acknowledging bad behavior and asking for forgiveness can be an important step in the right direction.

FINALLY, IT IS worth trying to examine what we have learned about dishonesty from a broader perspective and see what it can teach us about rationality and irrationality more generally. Through the different chapters, we have seen that there are rational forces that we think drive our dishonest behavior—but don’t. And there are irrational forces that we think don’t drive our dishonest behavior—but do. This inability to recognize which forces are at work and which are irrelevant is something we consistently see in decision making and behavioral economics research.

Viewed from this perspective, dishonesty is a prime example of our irrational tendencies. It’s pervasive; we don’t instinctively understand how it works its magic on us; and, most important, we don’t see it in ourselves.

The good news in all of this is that we are not helpless in the face of our human foibles (dishonesty included). Once we better understand what really causes our less-than-optimal behavior, we can start to discover ways to control our behavior and improve our outcomes. That is the real goal of social science, and I am sure that the journey will only become more important and interesting in the years to come.

Irrationally yours,

Dan Ariely