My Taxes are Too Darn High: Tax Protests as Revealed Preferences for Redistribution

Tax Audits as Scarecrows. Evidence from a Large-Scale Field Experiment

“…The effects of the public-goods message in the second post-treatment year, however, were close to zero (0.4%) and statistically insignificant (p-value=0.906). The evidence on the effects of moral messages is mixed, and they appear to work in some contexts (Bott et al, 2020;Nathan et al, 2020;Hallsworth et al, 2017) but not others (Blumenthal et al, 2001;Fellner et al, 2013;Castro and Scartascini, 2015;Dwenger et al, 2016;Meiselman, 2018;Perez-Truglia and Troiano, 2018). A closely 44 The results from panels (b) and (c) in Table 2 are based on a regression specification equivalent to the one from equation (1) above, which was used to obtain the estimates in panel (a).…”

Your Place in the World: Relative Income and Global Inequality

“…shows that, when forming inflation expectations, the average Argentine respondent assigns a weight of 0.432 to the signal provided to them (coefficient α-statistics reported in Panel B, column (1) of Table1). AndNathan et al (2020) shows that, when forming beliefs about the average tax rate, the average subject a weight of 0.459 to the signal (the difference in slopes from Figure A.5).38 In Appendix A.9 we provide a falsification test of the information intervention, by showing that there are no "effects" on the two survey outcomes measured pre-treatment (the belief in the importance of effort versus luck for individual economic success both at the national the global level). In Appendix A.10, we present results for the average effects of receiving information (i.e., regardless of whether the feedback was above or below the prior belief), and in Appendix A.11, we present the effects on the redistributive preferences and support for globalization and immigration elicited in the follow-up survey.…”

Betting on the House: Subjective Expectations and Market Choices

“…As shown inside the blue box at the bottom of Figure 2.a, the letter includes a URL to an online survey and a unique survey code to verify that the response came from a legitimate subject. 7 The main goal for including the survey link was to provide a proxy for the dates when recipients opened the letters, as in Perez-Truglia and Cruces (2017) and Nathan et al (2020).…”

“…32 Moreover, the type of information that we included in the letter was tailored to each subject (i.e., to their specific ZIP Code and property type), which limits the scope of spillovers even further. There are standard methods for measuring and accounting for spillovers in an experiment like ours (Cullen and Perez-Truglia, 2018;Nathan et al, 2020). However, given how negligible the spillovers are in our context, we did not bother making these adjustments.…”