Fraud and Economic Lies: Information and Strategies – FELIS

FELIS presents several original methodological dimensions. It combines theoretical modeling based on game theory with data collection by means of experiments and advanced econometrics methods. The laboratory has the advantage of eliciting cheating or lying behavior in a controlled environment instead of inferring cheaters’ intentions from reported behavior in surveys. This allows us to circumvent several difficulties of survey data such as reluctance to report acts of noncompliance. Field data continue to be the dominant source of information for testing theoretical models in economics; however, sometimes the data required for testing theoretical models may simply not be available or they are unreliable. This is the case for example for attitudes towards risk, ambiguity, trust, betrayal-aversion, or concerns for self-image, but it is even more sensible for the attitudes towards (dis)honesty and moral norms. By reproducing in the laboratory a stylized economic environment with real subjects and real monetary incentives, the experimental methodology offers the opportunity to provide original tests of theoretical models. Field experiments lose some control of the environment but increase the external validity of the outcomes. FELIS includes a study of the correlation between cheating in the lab and cheating in real settings at the individual level and shows, reassuringly, a strong correlation. A complementary neuroeconomics approach (fMRI) will help us understand the neural processes involved when people deal with the temptation of lying and their moral values.

- Simple tests of dishonesty in the lab can predict moral firmness in real life, although fraudsters who care about social image cheat less when behavior can be verified ex post by the experimenter.
- When a game allows both cheating and strategic truth-telling, lying behavior is not sensitive to revealing the sender’s identity to the observer. The option for observers to communicate with the sender and to reveal the sender’s lies to the receiver also do not affect lying. Senders whose identity is revealed to their observer do not lie less when their interests are misaligned with those of the observer.
- Crackdowns (intense monitoring in a concentrated amount of time) are a frequently used instrument. They have a positive impact on compliance only in the short run compared to non-concentrated random monitoring. They entail potentially deleterious effects in the long run since after monitoring is withdrawn the fraud rate increases to levels that are even higher than before the introduction of the crackdown.
-In the context of criminality, i) previous experience in the illegal sector reduces the likelihood of entering the legal sector in the future; ii) imprisonment has a negative side effect by enhancing recidivism due to a reduction of the legal return and an increase of the illegal return; iii) stronger punishment for repeat offenders, education in prison to avoid a depreciation of human capital, isolation of the most influent prisoners may help reducing the negative recidivism effect.
- Studying embezzlement, we have shown that the impact of a more transparent information on monetary flows in transfer chains between donors and receivers is conditional on the length of the chain and on the position of the intermediaries in the chain. Its direct effect on image encourages honesty. Its indirect effect via expectations plays in the opposite direction.

The work plan for the next months is the following:
-Work Package 1: Conduct of the experiments on self-deception and selective memory, and of the experiment on the strategic manipulation of the moral costs of deception.
-Work Package 2: Theoretical modelling and design of an experiment on cheating and competition in continuous time.
-Work Package 3: Design of experiments on fraud in public transportations. Design of experiments to test the model of criminality trap and recidivism.
-Work Package 4: Theoretical modelling and design of an experiment on cheating in social networks. Design of a neuroeconomic experiment on unethical behavior in competition.
-Work Package 5: Conduct of the experiment on the restoration of trust after unethical behavior within and between groups.

FELIS workshop “Social and Moral Norms”, Ecully, 16-17 October 2015

Softwares for each experimental protocol

Articles
- Van de Ven, J., Villeval, M.C. (2015). Dishonesty under scrutiny. Journal of the Economic Science Association 1, 86-99.
- Villeval M.C. (2014). Professional identity can increase dishonesty, Nature, News and Views, 516 (7529) (December), 48-49.
- Irlenbusch, B., Villeval, M.C. (2015). Behavioral ethics: how psychology influenced economics and how economics might inform psychology? Current Opinion in Psychology 6, 87–92.

Working papers

- di Falco, S., Magdalou, B., Masclet, D., Villeval, M.C., Willinger, M. Can transparency of information reduce embezzlement? An experiment in Tanzania. IZA Discussion Paper, Bonn, http://ftp.iza.org/dp.pdf; GATE WP 2016-15; ssrn.com/abstract=2775659
- Dai, Z., Galeotti, F., Villeval, M.C. 2016. The efficiency of crackdowns. An experiment in public transportations. GATE WP 2016-07; ssrn.com/abstract= 2729244 (submitted to Theory and Decision)
- Dai, Z., Galeotti, F., Villeval, M.C. 2016. Cheating in the lab predicts fraud in the field. An experiment in public transportations. IZA Discussion Paper 9702, Bonn, http://ftp.iza.org/dp9702.pdf; GATE WP 2016-05; ssrn.com/abstract=2725168. (R&R in Management Science)
- Chen, Y., Jiang, S., Villeval, M.C. 2016. The tragedy of corruption: Corruption as a social dilemma. IZA Discussion Paper, Bonn, http://ftp.iza.org/dp9384.pdf; GATE WP 2015-32; ssrn.com/abstract=2697983. (submitted to Journal of Public Economics)
- Bravard, C., Charroin, L. 2016. Optimal design and defense of networks under link attacks. GATE WP 2015-1519; papers.ssrn.com/sol3/papers.cfm (R&R in Journal of Mathematical Economics).
- Galeotti, F., Zizzo, D.J. 2015. Competence versus Honesty: What Do Voters Care About? GATE WP 15-20.

Trust is a core component of the freedom and security of citizens, making relationships work more cohesively and efficiently. But it is fragile. Undermining trust raises transaction costs, weakens social cohesion, and ultimately reduces the freedom of citizens and impoverishes society (Arrow, 1972). Dishonesty introduces a major threat on trust and the ubiquity of deception is a major concern of modern society. Big scandals have been revealed in finance and sports but ordinary deception is also widespread (cheating in exams, fare dodging, CV inflation, sabotage at work, etc.). Dishonesty often results from abusing information asymmetries. Governments and organizations spend considerable resources to detect dishonesty and implement coercive measures, which may be detrimental on freedom and social welfare. FELIS’s ambition is to contribute to a better understanding of the determinants of deceptive behavior in order to improve the deterrence of dishonesty and promote the security of citizens in a free world.

FELIS seeks to understand when people are more, or less, likely to act honestly and follow the moral course of action instead of serving their strict self-interest at the others’ expense. It aims at undertaking the standard economics-of-crime approach comparing the expected monetary benefits and costs of fraudulent actions by incorporating social, moral, emotional and psychological factors in economic decision-making. It will explore various mechanisms designed to fight against dishonesty while preserving the freedom of citizens.

The key challenge of FELIS is to provide a characterization of 1) strategies used to deceive oneself and others; 2) the links between deception and asymmetric information; 3) the deterrence effect of both uncertainty of audits and sanctions; 4) how social networks form and convey peer effects leading to norms of honesty or dishonesty; 5) how trust is affected by dishonesty. This defines five scientific Work Packages. Work Package 1 investigates how individuals lie while trying to maintain their self-concept of honesty, by exploring how the distance between the truth and a lie affects decisions and can be manipulated. Work Package 2 studies the link between deception and the exploitation of asymmetric information, by exploring strategic truth-telling as part of a greedy deceptive strategy. Work Package 3 examines both the deterrence effect of information disclosure on the occurrence of crackdowns and whether there exists optimal sanctions discouraging dishonesty while avoiding criminality traps. Work Package 4 focuses on the social dimension of deception, notably the formation of social networks and peer effects in networks. Work Package 5 studies how to restore trust after it has been destroyed by the dishonest behavior.

The ground-breaking nature and potential impact of FELIS is to go substantially beyond the current state of the art by offering a unique combination of game theory, experimental economics and psychology to better understand the causes and consequences of dishonesty in order to better identify efficient modes of deterrence that preserve the freedom of individuals. It combines theoretical modeling based on behavioral economics and experimental methods (laboratory experiments, lab-in-the-field experiments, neuroeconomics) to study the motivation of individuals to behave dis/honestly.

The Consortium involves behavioral economists and experts in the physiological and neural analysis of emotions from GATE (CNRS, University of Lyon), the CORTEX laboratory of excellence, and CREM (CNRS, University of Rennes). It includes international partners: experts in the econometrics of networks and tax evasion at Laval University in Quebec, a psychologist specialized in the analysis of lying behavior at the Ben Gurion University, and behavioral economists from the University of Amsterdam and the National Research University Higher School of Economics in Moscow.