Click for Academy of Management home page

Academy of Management

Does unethical behavior trigger anxiety and guilt? Exhilaration is more like it, new research finds

July 1, 2012

For more information, contact: Benjamin Haimowitz, (212) 233-6170,

When Bernard Madoff's Ponzi scheme came to light three and a half years ago, a cause of wonder rivaling its massive size was the state of mind of the culprit: how could someone, people asked, live for so many years with the psychological burden that such a huge swindle must have entailed?

New research on the psychological effects of cheating suggests it may not have been as hard as people imagined.

The research, to be presented at the annual meeting of the Academy of Management (Boston, Aug. 5-7), finds that, contrary to society's prevailing view that cheating triggers such negative feelings as guilt, shame, and anxiety, the immediate effect is instead something akin to exhilaration.

"In spite of the ubiquity of the assumption that unethical behavior triggers negative affect, remarkably little work has investigated its actual affective consequences," observe at the outset the paper's four authors, Nicole E. Ruedy of the Foster School of Business of the University of Washington, Celia Moore of London Business School, Francesca Gino of Harvard Business School, and Maurice E. Schweitzer of the Wharton School of the University of Pennsylvania.

In a series of experimental studies they find that "voluntary unethical acts not only fail to elicit negative affect but actually evoke positive affect, a phenomenon we term the 'cheater's high.' " This phenomenon, they continue, "is consistent with many anecdotal accounts of dishonesty, theft, and fraud," among which they cite such commonplace activities as "computer piracy, tax evasion, insurance fraud, workplace theft, and academic dishonesty" as well as more flagrant examples such as "wealthy individuals who delight in shoplifting affordable goods, joy-riders who steal cars for the thrill, and fraudsters who revel in their misdeeds."

Two of the six studies described in the paper probe "people's predictions of how they or another person would feel after engaging in unethical behavior," while four others investigate people's feelings when they cheat or when someone cheats on their behalf.

Study 1a and 1b: People predict they will feel upset and guilty if they cheat.
In the first experiment, 146 participants were instructed to read a description of a math test and to imagine themselves cheating on it. To what extent, they were asked, on a scale of 1/very slightly or not at all to 5/extremely, would they feel each of 10 positive emotions (such as "excited" or "alert") and 10 negative emotions (such as "upset" or "guilty"). People predicted they would feel significantly higher levels of negative affect (a mean of 3.0 on a scale of 1 to 5) than positive affect (a mean of 2.33 on a scale of 1 to 5).
In another study, 137 subjects were asked to imagine themselves involved in a situation that posed a choice of lying or not lying about billable hours earned by a consultant, and they predicted significantly higher negative affect than positive affect in scenarios in which lying occurred.

But these predictions turned out to be considerably at variance with the results of four different experiments in which participants had actual opportunities to lie and cheat or benefit from lying and cheating.

Study 2: Cheating on a test proves exhilarating.
One hundred seventy-nine individuals engaged in a timed anagram task that entailed unscrambling as many of 15 words as they could in four minutes, earning in the process $1 for every unscrambled word. At the outset, subjects completed a brief questionnaire in which they assessed their positive affect at the moment (to what extent they were excited, enthusiastic, interested, strong, and determined) and their negative affect at the moment (to what extent they felt upset, hostile, ashamed, jittery, and scared). Subjects then proceeded to unscramble as many words as they could, writing them on a sheet of paper that was stapled to a packet of sheets and a manila folder, until they were instructed to tear the top sheet from the packet and hand in the rest. Then they were asked, each in their individual cubicles, to check their work against an answer key that was now provided to them as well as to complete the same affect questionnaire they had completed some minutes earlier.

Unbeknownst to the participants, the papers beneath their answer sheets had created an imprint of their writing, so that it could be readily ascertained whether the subjects had added any words to their answer sheets when asked to check their work in privacy. Forty-one percent of the participants cheated by writing in additional answers, at $1 a shot, adding an average of 2.48 words to the 3.84 words they had unscrambled honestly.The cheaters also experienced statistically significant greater boosts over non-cheaters in positive affect (to 3.32 compared to 3.1). In sum, "after the opportunity to overstate performance, cheaters reported higher positive affect (but no higher negative affect) than non-cheaters did."

Study 3: Does a greater amount gained through cheating increase the cheater's high?
In an experiment similar to study 2, the professors probed whether the amount paid for each unscrambled anagram -- 10 cents, $1, or $2 -- would be reflected in the amount of subjects' positive or negative affect. "Positive affect in the $2 condition," they found, "was significantly higher than in the 10-cent condition, but the $1 condition was not significantly different from the $2 condition...Most importantly, the effect of cheating is significant even when controlling for the payment condition...suggesting that cheating produces an affective boost independent of the affective benefit triggered by higher payments."

Study 4: People also experience the cheater's high when they someone cheats on their behalf.
The professors tested whether the results of the two scrambled-anagram experiments had to do with the personality of cheaters themselves rather than with the act of cheating -- whether, in other words, the results reflected the fact that the cheaters were self-selected. Subjects in this study undertook a five-minute math test like the one described in study 1a, but did so in the company of a partner who, unbeknownst to them, was a confederate in the experiment. At the conclusion of the test, subjects were asked to report the number of the confederates' correct answers (always five), but, when confederates were asked to do the same for subjects, in half the cases they added five to the number of correct answers, an overstatement that added to the money the partners earned. None of the subjects with cheating partners corrected their over-reporting, and, on average, individuals in this group experienced a boost in positive affect from 2.91 to 3.21, while subjects who had honest partners actually showed a slight drop. The conclusion: "Even when cheating is not a self-selected behavioral choice, it creates a boost in positive affect."

Study 5: Being suspected of cheating actually increases the cheater's self-satisfaction.
This experiment probed whether explicitly recognizing the possibility of cheating would dampen positive affect. Some 203 participants completed an anagram task that included, as the third item on the list, a scrambled word so rare as to be virtually insoluble. Since subjects were asked to report only the number of words they had consecutively solved, any claim of answering 3 or more would be indicative of cheating. Yet, 52 percent of the subjects reported scores of 3 or more (on average 4.63), and, as in the previous experiments, those who did cheat had higher levels of positive affect than those who did not, controlling for baseline positive affect (and had significantly lower levels of negative affect as well, controlling for baseline).

In addition, half the subjects read a statement after they concluded their unscrambling efforts thanking them for their participation but raising the possibility that since "we cannot check your answers (i.e. whether you actually solved correctly the jumbles you told us you solved) may have cheated on this task by inflating your performance. We hope you reported your answers honestly." Then these subjects were specifically asked: "To what extent can we rely on your answers?" with responses provided on a five-point scale ranging from 1/not at all to 5/completely. This reality check notwithstanding, cheaters who were exposed to it revealed no dampening of the cheater's high in the positive-and-negative-affect test they took immediately afterward. Moreover, asked to estimate on a five-point scale their self-satisfaction (to what extent they felt clever, capable, accomplished, satisfied, superior), they evinced a level greater than subjects not exposed to the reality check. As the study puts it, "For cheaters, being confronted with the fact that the experimenters knew they were cheating increased self-satisfaction as compared to cheaters in the control condition."

What to make of these findings? "In addition to considering the material benefits and risks related to unethical behavior," the professors write, "the psychic costs and benefits should be considered. The cheater's high we document offers a potential explanation for the puzzling finding that unethical behavior is so pervasive and persistent, even when the stakes are low and the benefits are negligible."

The paper's findings, they conclude, "offer cause for concern. Many ethical decisions are made privately and are difficult to monitor. Individuals who recognize, perhaps from experience, that they can derive both material and psychological rewards from engaging in unethical behavior may be powerfully motivated to behave unethically. It is imperative that we develop our understanding of how emotions influence our moral behavior, how our moral behavior influences our emotions, and how people expect these relationships to work."

The paper, entitled “The Cheater's High: The Unexpected Affective Benefits of Unethical Behavior," will be as among several thousand research reports at the Academy of Management annual meeting, to be held in Boston from August 5th through 7th. Founded in 1936, the Academy is the largest organization in the world devoted to management research and teaching. It has some 19,000 members in 102 countries, including about 11,000 in the United States. This year's annual meeting will draw more than 9,000 scholars and practitioners for sessions on a host of subjects relating to business strategy, organizational behavior, corporate governance, careers, human resources, technology development, and other management-related topics.

Academy of Management
Member Services
Academy of Management
Online Opportunities
Academy of Management