Corporate fraud is a major challenge in both developing and advanced economies, and employee whistle-blowers play an important role in uncovering it (e.g. Dyck et al. 2010). Indeed, the issue of protecting employee whistle-blowers looms large on the international anti-corruption agenda of the G20, the Council of Europe and the OECD (Council of Europe 2014, OECD 2011, 2016). In order to encourage employees to come forward, their ‘best-practice recommendations’ advocate easy access to protection for whistle-blowers who report in good faith. However, practitioners have worried that easy access to protection might lead to fraudulent claims by low-productivity employees with the sole aim of gaining employment protection.
Our recent study is the first paper to systematically evaluate different whistle-blower protection schemes in a theory-guided lab experiment (Mechtenberg et al. 2017). In particular, we study the impact of the stringency of requirements for obtaining protection on crucial outcome variables such as deterrence, whistle-blowing (truthful and fraudulent), investigation decisions, and retaliation. Our results are not unequivocally supportive of the policy recommendations discussed above (which are already legal practice in countries such as the US or the UK), as more stringent requirements for obtaining protection lead to better outcomes in terms of reporting behaviour and deterrence.
Our theoretical framework is based on a cheap-talk model in the spirit of Crawford and Sobel (1982), and hence also related to the recent whistle-blowing application by Chassang and Padró i Miquel (2016). We consider the interaction between an employer (who may misbehave), an employee (who may blow the whistle), and a prosecutor (who may act upon the employee’s report). Moreover, the employer might retaliate against a non-protected whistle-blower by dismissing him or her. We focus on whistle-blower protection in the form of employment protection (i.e. a protected employee cannot be dismissed), as is common in legal practice. We allow employees to be heterogeneous with respect to their productivity – the incentive structure is such that, whenever feasible, the employer prefers to dismiss low-productivity employees, while their high-productivity counterparts might face retaliation only if they blow the whistle. This might give low-productivity employees an incentive to file fraudulent claims in order to gain employment protection (i.e. in order to avoid dismissal). We solve the model for different legal regimes.
The predictions of the model are then empirically tested in a lab experiment. Thereby, our paper contributes to a growing experimental literature on whistle-blowing (e.g. Apesteguia et al. 2007, Hinloopen and Soetevent 2008, Bigoni et al. 2012, 2015, Bartuli et al. 2016, Feltovich and Hamaguchi 2016, Butler et al. 2017). In the experiment, the following legal regimes are considered (as different treatments).
- First, in a benchmark treatment ‘no whistle-blowing protection’, whistle-blower protection (i.e. employment protection) is not available.
- Second, in treatment ‘whistle-blowing protection 1’, protection is easily obtained by just filing a report. Hence, this treatment is meant to capture a ‘belief-based regime’ as, for example, embodied in US law, UK law, and the G20’s policy recommendation.
- Third, in treatment ‘whistle-blowing protection 2’, protection is obtained if a report is lodged and if, in addition, it indeed triggers an investigation by the prosecutor.
- Fourth, in treatment ‘whistle-blowing protection 3’, protection is only granted if, in addition to the requirements of ‘whistle-blowing protection 2’, there is indeed misbehaviour. Hence, ‘whistle-blowing protection 3’ is even more stringent than ‘whistle-blowing protection 2’ (and might be labelled a ‘fact-based regime’).
Main experimental findings
First, most of the theoretical predictions (with respect to dismissal, misbehaviour and the reporting behaviour of the different productivity types) are broadly supported by the experimental data.
Second, there are also interesting deviations:
- In treatment ‘whistle-blowing protection 1’, we find that fraudulent claims are indeed an issue, and even more so than predicted by theory.
- Moreover, these fraudulent claims do not only affect ‘productive efficiency’ (in the sense that low-productivity employees cannot be replaced by more productive ones). Fraudulent claims also reduce prosecutors’ responsiveness to reports, as these are now less informative about underlying misbehaviour.
- As a consequence, the predicted reduction of misbehaviour in treatment ‘whistle-blowing protection 1’ relative to ‘no whistle-blowing protection’ does not materialise.
- As predicted, behaviour in ‘whistle-blowing protection 1’ and ‘whistle-blowing protection 2’ is very similar.
- In ‘whistle-blowing protection 3’ , there are substantially fewer fraudulent claims than in ‘whistle-blowing protection 1’.
- Moreover, compared to ‘whistle-blowing protection 1’, prosecutors in ‘whistle-blowing protection 3’ make better decisions in terms of less undetected misbehaviour and less unnecessary investigations. Employer misbehaviour is also lower in ‘whistle-blowing protection 3’.
Our findings point to potential shortcomings of easy access to whistle-blower protection (as recommended by the G20 and implemented in a number of countries). Our results support the view that when protection is relatively easy to obtain, fraudulent claims are indeed a prevalent phenomenon. This reduces the informativeness of reports to which prosecutors respond with a lower propensity to investigate.
All in all, our results suggest that the G20’s recommendation of easy access to whistle-blower protection might lead to unintended side effects. In contrast, we find that these can be mitigated under a regime where the requirements for obtaining protection are more stringent.
Apesteguia, J, M Dufwenberg, and R Selten (2007), “Blowing the whistle”, Economic Theory 31: 143-166.
Bartuli, J, B Djawadi, and R Fahr (2016), “Business ethics in organisations: An experimental examination of whistleblowing and personality”, IZA Discussion Paper No. 10190.
Bigoni, M, S-O Fridolfsson, C Le Coq and G Spagnolo (2012), “Fines, leniency and rewards in antitrust”, RAND Journal of Economics 43: 368-390.
Bigoni, M, S-O Fridolfsson, C Le Coq and G Spagnolo (2015), “Trust, leniency and deterrence”, Journal of Law, Economics, and Organization 31: 663-689.
Butler, J, D Serra and G Spagnolo (2017), “Motivating whistleblowers”, Stockholm School of Economics, mimeo.
Chassang, S and G Padró i Miquel (2016), “Corruption, intimidation and whistleblowing: A theory of inference from unverifiable reports”, New York University, mimeo.
Crawford, V P and J Sobel (1982), “Strategic information transmission”, Econometrica 50(6): 1431-1451.
Council of Europe (2014), “Recommendation CM/Rec(2014)7 of the Committee of Ministers to member States on the protection of whistle-blowers (Adopted by the Committee of Ministers on 30 April 2014)”.
Dyck, A, A Morse and L Zingales (2010), “Who blows the whistle on corporate fraud?”, Journal of Finance 65: 2213-2253.
Feltovich, N and Y Hamaguchi (2016), “The effect of whistle-blowing incentives on collusion: An experimental study of leniency programmes,” Monash University, mimeo.
Hinloopen, J and A Soetevent (2008), “Laboratory evidence on the effectiveness of corporate leniency programs”, RAND Journal of Economics 39: 607-616.
Mechtenberg, L, G Muehlheusser and A Roider (2017), “Whistle-blower protection: Theory and experimental evidence”, CEPR Discussion Paper No. 11898.
Muehlheusser, G and A Roider (2008), “Black sheep and walls of silence”, Journal of Economic Behavior & Organization 65: 387-408.
OECD (2011), “G20 anti-corruption action plan: Protection of whistleblowers”.
OECD (2016), Committing to effective whistleblower protection, Paris: OECD Publishing