Cambridge International Symposium on Economic Crime - PowerPoint Presentation

Slide1

Cambridge International Symposium on Economic CrimeSeptember 6, 2017

Leandra Lederman William W. Oliver Professor of Tax LawIndiana University Maurer School of Law

“Voluntary” Tax Compliance:A View From the U.S.Slide2

The IRS Versus the Tax Evader2Slide3

A Frequent Starting Point: The Deterrence Model3The standard economic approach to tax noncompliance involves detection and punishment (audits and penalties). E = p(F), where E is expected cost of evasion; p is probability of detection, and F (

fine) is the amount that will be due (tax + penalty). Limited resources mean a low probability of audit. In 2016, the IRS audited just 0.7% of individual income tax returns filed and just 1.1%

of corporate returns.Penalties in the US federal income tax are generally 20% of the underpayment.75% rate for fraud, but the IRS has a high burden of proof.

Source: Internal Revenue Serv., 2016 Data Book 23

tbl. 9a (2017).Slide4

Basic Economics Would Seem to Predict Zero Compliance4A taxpayer who evades $1,000 and is caught will owe $1,200 at a 20% penalty. At a 1% audit rate,

the expected payment if cheating is only $12.At a 1% audit rate, unless the penalty is more than 99 times the tax (9900%), there appears to be no economic incentive to comply.Slide5

But Overall U.S. Voluntary Complianceis Estimated at ~82%5Source: https://www.irs.gov/PUP/newsroom/tax%20gap%20estimates%20for%202008%20through%202010.pdfSlide6

The False Puzzle6Some scholars have called high voluntary compliance in a regime of low audits and penalties a “puzzle.” E.g.:Jack Manhire (2015)Bruno Frey (2011)Dan Kahan

(2002)Eric Posner (2000)Michele Bernasconi (1997)James Alm et al. (1995)They generally say that deterrence does not explain tax compliance at realistic levels of risk aversion.Slide7

7What Does The Basic

Deterrence Model Omit?Slide8

Possible Deterrents Omitted From the Most Basic Economic Model81) Some small ones:Overestimation of likelihood of auditFear of audit itself: time investment, expense, stressPossible stigma of tax investigation

2) And don’t forget:Possible criminal penaltiesSlide9

IRS Activity on Legal-SourceTax Crimes, 20169Source: Internal Revenue Serv., 2016 Data Book 44 tbl.18 (2017).Slide10

The Fear of Criminal Penalties10“It could just be a coincidence, I suppose—but sometimes I wonder if it’s the government’s way of rubbing it in.”Slide11

The US Government Strategically Publicizes Tax Indictments11

Source: Joshua D. Blank & Daniel Z. Levin, When Is Tax Enforcement Publicized?, 30 Va. Tax Rev. 1, 18 (2010).Slide12

3) Third-Party Reporting: It Explains Much U.S. “Voluntary” Compliance12

Noncompliance increases as information reporting decreases.Source: https://www.irs.gov/PUP/newsroom/tax%20gap%20estimates%20for%202008%20through%202010.pdfSlide13

The Effectiveness of Third-Party (Information) ReportingAmounts subject to substantial information reporting

and withholding: 99% voluntary compliance rate.

Amounts subject to substantial information reporting: 93% voluntary compliance rate.Amounts subject to

some information reporting: 81%

voluntary compliance rate.Amounts subject to little or no

information reporting: 37% voluntary compliance rate.

13Slide14

Third-Party Reportingis Akin to the Threat of AuditThird-party reporting (information reporting) allows the government to obtain information about the taxpayer’s tax situation.The taxpayer knows the government has the information.

• This makes information reporting more than just an enforcement tool; it also deters noncompliance.

See Leandra Lederman, Tax Compliance and the Reformed IRS, 51 Kansas L. Rev. 971(2003)

14Slide15

15Are Firms the

Key Instead? Slide16

Taxation Without Information?16In Taxation Without Information: The Institutional Foundations of Modern Tax Collection, Cui argues that 3rd-party reporting is not key.He argues that the key is that firms are pro-social.It’s true that firms usually are the 3rd-party reporters.

[Lederman (2010)]CentralizationAccounting infrastructureBUT, as a first cut, compare IRS estimates:93% voluntary compliance rate by individuals for income subject to substantial information

reporting but no withholding 83% voluntary compliance rate by corporations with the corporate income tax Slide17

What if an Individual Reports on a Firm?17An individual 3rd-party reporter should not increase firms’ compliance if firms are inherently honest.This context is fairly rare but we have some evidence:

Naritomi (2013): The Nota Fiscal Paulista (NFP) program in São Paulo, Brazil, designed to decrease retail firms’ VAT evasion. Consumers received incentives to monitor retailers’ reporting. Retailers

’ reported revenues increased by 22% over 4 years. Kumler et al. (2015): Pension reform in Mexico. Younger workers’ pensions were linked to reported wages, providing these workers info & an incentive to monitor employers’ wage reporting. Firms reduced underreporting of wages (and thus payroll tax evasion) for younger workers

.Slide18

18The Effectiveness of Tax EnforcementSlide19

Effects of Audits and Penalties19Audits have a positive spillover effect in the U.S.: Treasury Dept. (2015): 6 to 1 indirect effectPlumley

(2002): 11.7 to 1 indirect effectBirskyte (2013): positive effect of federal (IRS) audits on U.S. state tax collectionsAudit threats are very effective:Hasseldine et al. (2007): U.K. sole proprietorsKleven

et al. (2011): Denmark employees & self-employedDwenger et al. (2015): German church taxAudits with prior notice are generally very effective:Pomeranz (2015): Chilean VATSlemrod et al. (2001): Minnesota individual income taxPenalty threats are less effective than audit threats:

Chirico et al. (2015): Philadelphia property taxIyer et al. (2010): Washington

state Business & Occupation Tax and use taxesMaciejovsky et al. (2007): lab experiment

See Leandra Lederman, Does Enforcement Crowd Out Voluntary Tax Compliance?, __ B.Y.U.

L. Rev. __ (2017) (forthcoming)Slide20

Concluding Points20The deterrence model explains much tax compliance.Audit threats and audits with prior notice are very effective.This does not address the effect of audit on the small % of taxpayers actually audited (the topic of my current paper).The experience of audit seems to increase tax payments by those whose returns are adjusted to increase tax.

Information reporting does much (but not all) of the work.Slide21

21Questions?

Comments are welcome here or by email to

llederma@indiana.edu.Slide22

22Slide23

Will Enforcement Interfere WithTax Compliance Norms?23

Some argue that enforcement may reduce compliance norms.The opposing argument:Too little enforcement can impoverish compliance if taxpayers who comply feel like “chumps.”Reasonable deterrence can foster and sustain compliance norms.

Source: Leandra Lederman, The Interplay Between Norms and Enforcement in Tax Compliance, 64 Ohio State L.J. 1453, 1510 (2003).Slide24

What Effect Does the Audit ExperienceHave on Those Audited?24Overall, audits have positive effects on the audited taxpayers:DeBacker et al. (2015): IRSNational Taxpayer Advocate (2015):

IRS audits of self-employed individualsKleven et al. (2011): DenmarkAudits have a positive effect on effect on future tax payments of taxpayers whose returns were adjusted to owe more taxNational Taxpayer Advocate (2015

)Gemmell & Ratto (2012): Random audits of UK small business and personal taxpayers.Audits seem to have a deleterious effect on taxpayers whose returns were not adjustedNational Taxpayer Advocate (2015): Decline of 35% three years after audit in one model.

Gemmell & Ratto (2012):

Decline but significant at p < .05 only for “personal taxpayers,” not small or medium businesses. This “bomb crater” effect warrants further investigation. E.g., Satterthwaite (2016).Recall that audits have a

positive spillover effect on those not auditedTreasury Dept. (2015): 6 to 1 indirect effect

Plumley (2002): 11.7 to 1 indirect effectSlide25

Possible Explanations for the Negative Effect of No-Change Audits25Taxpayers with no adjustments were very compliant and reacted negatively to audit experience.Maybe their learning was that they were chumps to comply.Taxpayers with no adjustments were very devious and learned from the audit how best to cheat.

Lab experiments can distinguish devious and compliant.Inferring likelihood of future audit is low.More sensible for endogenous audit than random audit.Slide26

Lab Experiments on Effect of Audits26Several lab experiments find a “bomb crater” effect of audits.Mittone (2006): Univ. of Trento: economics studentsDid not distinguish between compliant and noncompliant audited.Maciejovsky et al. (2007): Univ. of Vienna & Vienna Univ. of Economics and Business

Administration: economics students Found no difference between compliant and noncompliant audited.Kastlunger et al. (2009): Univ. of Trento: economics students 52.7% of compliant audited reduced payments in next round; 36.9% of non-compliant audited did so.

Satterthwaite (2016): U.S. residents on Amazon’s Mechanical TurkRandom audit (control) generally had bomb-crater effect. Endogenous audits eliminate bomb crater after round 10.Slide27

Planned work with Satterthwaite:Imperfect Audits in the Lab27Attempt to replicate real-world audits, which can’t perfectly detect compliance.Takes advantage of the fact that we know who failed to comply, regardless of whether that person was “audited.”

Plan to use Amazon’s Mechanical Turk, which she has used before: e.g., Florida Tax Review (2016). Suggestions on study design (imperfect audits, endogenous audits, etc.) are most welcome.