Truth or Consequences: An ExperimentJordi Brandts • Gary CharnessInstituto de Análisis Económico, Consejo Superior de Investigaciones Científicas, Barcelona, SpainUniversity of California, Santa Barbara, California [email protected] • [email protected] paper presents evidence that the willingness to punish an unfair action is sensitiveto whether this action was preceded by a deceptive message. One player first sends amessage indicating an intended play, which is either favorable or unfavorable to the otherplayer in the game. After the message, the sender and the receiver play a simultaneous 2× 2game, in which the sender may or may not play according to his message. Outcome cells may,hence, be reached following true or false messages. In the third stage, the receiver may (at acost) punish or reward, depending on which cell of the simultaneous game has been reached.We test whether receivers’ rates of monetary sacrifice depend on the process by which anoutcome is reached. We study two decision-elicitation methods: the strategy and the directresponse methods. For each method, deception more than doubles the punishment rate as aresponse to an action that is unfavorable to the receiver. We also find evidence that 17–25% ofall participants choose to reward a favorable action choice made by the sender, even thoughdoing so leaves one at a payoff disadvantage. Our results reflect on current economic modelsof utility and have implications for organizational decision-making behavior.(Laboratory Experiments; Reciprocity; Fair Process; Honorable Behavior; Deception; Punishment)1. IntroductionNotions of fair process and honorable behavior havepotentially important implications for social andeconomic interactions, which may be found in negoti-ations or organizations. Negotiators often have incen-tives to mislead others and private information maypresent the opportunity for doing so. Managers maybe tempted to use deception to improve the chancesof a desired response from a group of employees.However, there may be significant limitations to thiskind of behavior: Where the character of interac-tion is highly interpersonal in nature, one must takeinto account the potential impact of social considera-tions on one’s motivation, as this may impose boundson selfish or dishonest behavior. We feel that it isa natural intuition that deception will be consideredinappropriate behavior and may lead to substantialpunishment behavior by the deceived. The experi-ment we present allows us to subject this intuition toa rigorous test.The reaction to deception may also be generallyrelevant in the context of the current process of for-mulating more accurate theoretical models of humanmotivation. The most common assumption in eco-nomics is that people only care about maximizingtheir own income. But a large body of research hasshown that many people choose to sacrifice moneyin laboratory experiments. A number of recent for-mal models presume that people are also motivatedby considerations of altruism, inequity, and aspects ofthe process by which an allocation is reached; whilepeople may, nevertheless, maximize their utility, one’sown money is not the sole determinant of utility.This paper reports experimental results in a gamewith communication and the possibility of costly ret-ribution.1Our design uses a game with three stages.Figure 1 shows the central building block of our1Webster’s Third New International Dictionary defines retribution as“the dispensing or awarding of punishment or reward accordingto the deserts of the individual.”Management Science © 2003 INFORMSVol. 49, No. 1, January 2003 pp. 116–1300025-1909/03/4901/0116$5.001526-5501 electronic ISSNBRANDTS AND CHARNESSTruth or ConsequencesFigure 1 Second Stage GameB1 B2A1 2, 2 6, 9A2 2, 2 12, 3experimental design; the two numbers in each cellbelow refer to the respective material payoffs for therow and column players.In the first stage, player A sends a costless and non-binding message to player B, stating that he intends toplay A1 or intends to play A2. In the second stage, theplayers then simultaneously choose actions, and oneof the cells of the Figure 1 matrix is reached. Note thateach cell can be reached via two different message-action paths—the sender may play in accordance withhis message or he may not. In the third stage, player Bmay have additional options: If the outcome cell (A2,B2) is reached, player B can accept the (12, 3) out-come or change it to (2, 2). If the outcome cell (A1, B2)is reached, player B can accept the (6, 9) outcome orchange it to (8, 7). Our experimental design includespossible misrepresentation, and provides a retributionmechanism with monetary consequences. This allowsa clear expression of one’s objection to deception.Previous studies have demonstrated that people arewilling to punish unfair actions, even at a personalcost. Fehr and Gächter (2000) find that free ridersin a public goods experiment are heavily punished,and that this leads to substantially higher levels ofcooperation. Sefton et al. (2000) also note that sanc-tions in a public goods experiment are quite effec-tive in achieving stable group allocations. Boles et al.(2000) study ultimatum game behavior when the pro-poser and the responder may have private informa-tion concerning the pie size or the outside (rejection)payoff, respectively, and can exchange written mes-sages. In their repeated-game fixed-matching design,they find that revelation of deceptive claims didnot substantially increase deceptive proposers’ sub-sequent offers, but responders who learn they havebeen deceived are more likely to reject these subse-quent offers. However, in this design one’s reaction todeception is mixed in with possible strategic consid-erations, because it is common information that onewill be paired with the same person for the durationof the experiment.With the study of the game presented above, wewish to add to the analysis of the effects of decep-tion in both organizational settings and strategic sit-uations. Previous works by Anton (1990), Lewicki(1983), and Shapiro and Bies (1994) find that decep-tion or lying can lead to moral outrage, and damageongoing organizational relationships. Schweitzer andCroson (1999, p. 1) suggest that “deception in organi-zations represents a significant managerial challengeacross a broad range of functional areas.” Our studyprovides clear and direct evidence that the