Policymakers in the developed and developing world are actively looking to improve tax compliance. This desire becomes more evident during adverse economic times, as witnessed during the fiscal crises following the global downturn of 2008 (Acemoglu and Jackson 2014) and more recently in Greece (Tagkalakis 2014). This demand for tax compliance has sparked a rich literature that studies a variety of policies aimed at reducing tax evasion and tax avoidance. For example, we know that tax evasion can be reduced, with a varying degree of success, through several methods: audits (Slemrod et al. 2001), third-party reporting (Kleven et al. 2011), paper trails (Pomeranz forthcoming), public disclosure of tax records (Bø et al. 2015), and satellite detection of unregistered buildings (Casaburi and Troiano 2015).1
Although the vast majority of the tax compliance literature focuses on tax evasion and tax avoidance, a significant portion of the tax gap includes tax delinquencies (i.e. debts incurred by taxpayers who refuse to pay their tax dues).2 Because these potential tax revenues are arguably the most readily available to the tax agencies, tax agencies invest substantial resources in designing and implementing policies to collect tax debts. However, little is known about the effectiveness of these policies.
Theory and evidence on tax delinquency
In a recent paper (Perez-Truglia and Troiano 2015), we provide some theory and evidence about the enforcement of tax debt. In addition to financial penalties, we study a ‘shaming’ penalty in which the names, addresses, and other identifying information of individuals and businesses with delinquent taxes are published online. These shaming policies have been implemented in 23 of 50 states in the US and in other developed and developing countries. In spite of their widespread use, little is known about whether these shaming penalties effectively reduce tax delinquency or whether they are desirable from a social welfare perspective.
First, we present a model to explain why a tax agency may use a shaming penalty even when financial penalties are available. We show that the shaming penalty can be a good option for a government that cares about the private welfare of citizens in addition to tax revenues. Such a government would like to better target individuals who are less vulnerable to the financial penalties and the shaming penalty provides an additional instrument to achieve that goal. As a result, the shaming penalty may not only increase tax revenues but also social welfare.
Second, we present evidence from a field experiment about the effectiveness of the financial and shaming penalties. It is not obvious that the shaming penalty should reduce tax delinquencies; indeed, it could even backfire. For example, shaming penalties may conflict with the intrinsic desire to honour tax debts, as suggested by the literature on crowding out between intrinsic and extrinsic motivation (Bénabou and Tirole 2003, Dwenger et al. 2014).3 The shaming penalty could also alienate delinquents from future sources of income that could be used to pay off their debts. Delinquents may also be proud that they are not paying taxes, in which case being listed online may be seen as a reward rather than a penalty.
A field experiment comparing financial and shaming penalties
We conducted a field experiment with 34,344 individuals who were publicly listed as tax delinquents in three US states: Kansas, Kentucky, and Wisconsin. Collectively, our subject pool owed nearly half a billion dollars. The median subject owed $5,500 and had been a delinquent for two years, despite numerous attempts and solicitations from the tax agency and high financial penalties. We sent letters to all 34,344 of these delinquents. These letters were identical except for a few key pieces of information that were randomly assigned in a non-deceptive manner. These messages were designed to affect the salience of financial and shaming penalties. We then measured how the information about financial and shaming penalties affected the probability that each recipient paid off her debt in the subsequent months.
To measure the effect of the shaming penalty, we compared across individuals assigned to a lower- and higher-visibility treatments. In the lower-visibility treatment, the letter indicated that the recipient was the only individual in the area randomly chosen to receive information about the online list of delinquents. In the higher-visibility treatment, the letter indicated that other individuals from the recipient’s area were also randomly chosen to receive this information, so recipients would feel that their delinquent behaviour could be monitored by neighbours. To measure the effect of financial penalties, we compared across individuals whose letters included information about financial penalties and individuals whose letters did not include such information.
The evidence suggests that both the financial and shaming penalties increased the probability of paying off the tax debt. However, our evidence also indicates some limits of peer pressure – while the effect of the shaming penalty was significant for individuals who owed less than $2,500, it was not significant for higher debts. The effect of the shaming penalty was both statistically and economically significant – the higher visibility condition increased the probability of leaving the list from 10% to 12% in just five weeks after mail delivery. We argue that this 20% increase in the payment rate provides a conservative lower bound to the potential effect of shaming policies, because the effect on visibility from our intervention was small compared to the potential overall effect from the publication of lists of tax delinquents. For example, while our experimental intervention increased the visibility among neighbours only, the publication of online lists would also increase visibility among relatives, friends, and co-workers (or, for that matter, anyone who may Google the delinquent’s name).
Last, even though our evidence shows that shaming penalties can increase tax revenues, it is unclear whether shaming penalties are desirable from a social welfare perspective. We provide some suggestive evidence that, in line with the model, the shaming penalty may be used to target penalties towards delinquents who are less vulnerable to the financial penalties.
Bénabou, R J M and J Tirole (2003), “Intrinsic and Extrinsic Motivation”, Review of Economic Studies 70: 489–520.
Bénabou, R J M and J Tirole (2006), “Incentives and Prosocial Behavior”, The American Economic Review 96(5): 1652–1678.
Besley, T, A Jensen, and T Persson (2014), “Norms, Enforcement, and Tax Evasion”, Working Paper.
Blumenthal, M, C Christian, and J Slemrod (2001), “Do Normative Appeals Affect Tax Compliance? Evidence from a Controlled Experiment in Minnesota”, National Tax Journal 54: 125–138.
Chetty, R, M Mobarak, and M Singhal (2014), “Increasing Tax Compliance Through Social Recognition”, Policy Brief.
Bø, E E, J Slemrod, and T O Thoresen (2015), “Taxes on the Internet: Deterrence Effects of Public Disclosure”, American Economic Journal: Economic Policy 7(1): 36–62.
Casaburi, L and U Troiano (2015), “Ghost-House Busters: The Electoral Response to a Large Anti Tax Evasion Program”, 29 January.
DellaVigna, S, J List, and U Malmendier (2012), “Testing for Altruism and Social Pressure in Charitable Giving”, Quarterly Journal of Economics, 127(1): 1–56.
Dwenger, N, H Kleven, I Rasul, and J Rincke (2014), “Extrinsic and Intrinsic Motivations for Tax Compliance: Evidence from a Field Experiment in Germany”, Mimeo.
Gerber, A S, D P Green, and C W Larimer (2008), “Social Pressure and Voter Turnout: Evidence from a Large-scale Field Experiment”, American Political Science Review 102(1): 33–48.
Hallsworth, M, J A List, R D Metcalfe, and I Vlaev (2014), “The Behavioralist As Tax Collector: Using Natural Field Experiments to Enhance Tax Compliance”, NBER Working Paper 20007.
Kleven, H J, M B Knudsen, T Kreiner, S Pedersen, and E Saez (2011), “Unwilling or Unable to Cheat? Evidence from a Randomized Tax Audit Experiment in Denmark”, Econometrica 79(3): 651–692.
Perez-Truglia, R and U Troiano (2015), “Tax Debt Enforcement: Theory and Evidence from a Field Experiment in the United States”.
Perez-Truglia, R and G Cruces (2015), “Partisan Interactions: Evidence from a Field Experiment in the United States”.
Pomeranz, D (forthcoming), “No Taxation without Information: Deterrence and Self-Enforcement in the Value Added Tax”, The American Economic Review.
Slemrod, J, M Blumenthal, and C Christian (2001), “Taxpayer Response to an Increased Probability of Audit: Evidence from a Controlled Experiment in Minnesota”, Journal of Public Economics 79(3): 455–483.
Tagkalakis, A O (2014), “Tax administration reforms and the fight against tax evasion: Recent evidence from Greece”, VoxEU.org, 2 December.
 The evidence about some other methods, such as appealing to normative or moral appeals, is mixed (Blumenthal et al. 2001, Fellner et al. 2013, Hallsworth et al. 2014, Besley et al. 2014, Dwenger et al. 2014). Chetty et al. (2014) also describe results from an intervention in Bangladesh that social recognition may increase reported income.
 Tax delinquency is a significant problem in the developed world, and plays an even more important role in the developing world. For example, according to the US Department of Treasury (2012), delinquent taxes composed more than 20% of the total US gross tax gap in 2006. The Treasury reported $46 billion in underpayment of declared taxes and $65 billion in enforced and other late payments as of 2006. In addition to the previous items, the tax gap includes nonfiling and underreporting, estimated to be near $450 billion dollars.
 Indeed, most of the evidence on peer pressure is about pro-social behaviour, such as voting and giving (Gerber et al. 2008, DellaVigna et al. 2013, Perez-Truglia and Cruces 2013), but being a delinquent seems more in line with anti-social behaviour. For a formal characterisation of who peer pressure may work on, see Bénabou and Tirole (2006).