شواهد تجربی از بازی های فرار مالیاتی
Abstract
1- Introduction
2- Truth-telling oath procedure
3- Experiment 1: One-shot tax declaration
4- Experiment 2: repeated tax declaration
5- Conclusion
References
Abstract
Using two earned income/tax declaration experimental designs we show that only partial liars are affected by a truth-telling oath, a non-price commitment device. Under oath, we see no change in the number of chronic liars and fewer partial liars. Rather than smoothly increasing their compliance, we also observe that partial liars who respond to the oath, respond by becoming fully honest under oath. Based on both response times data and the consistency of subjects when several compliance decisions are made in a row, we show that partial lying arises as the result of weak preferences towards profitable honesty. The oath only transforms people with weak preferences for lying into being committed to the truth.
Introduction
Dishonesty erodes opportunities for economic gains in business and society.1 Scandals like Enron and the 2008 financial crisis have prompted the promotion of truth-telling oaths–a non-price commitment device, as well as similar pledges, or codes of conduct, like the MBA oath or Dutch Bankers oath (see e.g., McCabe et al., 2002; Shu et al., 2012; Cohn et al., 2014).2 Empirical evidence confirms that when people voluntarily commit to honesty through a solemn oath, they tell the truth most of the time, holding the lie constant (see e.g., Jacquemet et al., 2018).3 The rationale is that the oath will help promote greater economic exchange by triggering a person’s intrinsic commitment to telling the truth (i) by reducing the ability to rationalize a lie (i.e., self-justification), and (ii) by coupling the desire for a positive self-image together with the desire for consistency (Mazar et al., 2008). But not all lies are the same—a small lie is easier than a big lie (Lundquist et al., 2009; Gneezy et al., 2018; Abeler et al., 2019). And not all liars are the same—some people lie all the time, some never, and some waver between lying and the truth, depending. The open question we address is who actually responds to an oath with honesty, and why. Using an earned income/tax declaration lab experiment, we find that the oath only affects partial liars, not full liars.4 What is more, partial liars do not react to the oath by smoothly increasing their level of compliance under oath (as a homogeneous change in the cost of lying would predict), but rather jump to full compliance. We argue this is consistent with partial compliance arising from weak preferences for profitable dishonesty-partial liars seem to lack clarity about how to process profitable dishonesty, which suggests they were neither committed to lying nor to being truthful 100%. Their preferences were fungible-and the oath acted as a moral anchor, so that we observe fewer partial liars and more full truth tellers under oath than without the oath.