Tulane Economics Working Paper Series

New Technologies and the Evolution of Compliance

James Alm Joyce Beebe Michael S. Kirsch Tulane Economics Rice University Notre Dame Law School [email protected] Omri Marian Jay A. Soled University of California Rutgers University Irvine

Working Paper 2009 September 2020

Abstract Improving tax compliance is a common goal of governments worldwide. The United States is no exception. The size of the nation’s “tax gap” — or the difference between what taxpayers pay in in a timely manner and what they should pay if they fully complied with the tax laws — is hundreds of billions of dollars annually, significantly depriving the nation of much-needed revenue. This paper explores the mixed effects of technological advancements on tax compliance — and, thus, its counterpart, . On the one hand, technological advances have largely eradicated many of the commonplace tax-noncompliance techniques that once reigned during the twentieth century. On the other hand, many of these very same technological advances threaten to usher in new modes of . Which of these emergent trends will dominate is unclear. The outcome will largely depend upon whether Congress updates the tax laws to address technological advances and grants sufficient funding to the Internal to maintain robust enforcement efforts in an ever-changing technological landscape. Failure to take these steps will destine the size of the tax gap to expand. Many prior studies have addressed discreet effects of specific technologies on tax compliance. This paper contributes to this developing literature by offering a cohesive framework to address technological advancement and tax compliance.

Keywords: Technology, Tax compliance, Tax gap. JEL codes: H24, H26.

NEW TECHNOLOGIES AND THE EVOLUTION OF TAX COMPLIANCE

James Alm, Joyce Beebe, Michael S. Kirsch, Omri Marian, and Jay A. Soled*

Improving tax compliance is a common goal of governments worldwide. The United States is no exception. The size of the nation’s “tax gap” — or the difference between what taxpayers pay in taxes in a timely manner and what they should pay if they fully complied with the tax laws — is hundreds of billions of dollars annually, significantly depriving the nation of much-needed revenue. This paper explores the mixed effects of technological advancements on tax compliance — and, thus, its counterpart, tax noncompliance. On the one hand, technological advances have largely eradicated many of the commonplace tax-noncompliance techniques that once reigned during the twentieth century. On the other hand, many of these very same technological advances threaten to usher in new modes of tax evasion. Which of these emergent trends will dominate is unclear. The outcome will largely depend upon whether Congress updates the tax laws to address technological advances and grants sufficient funding to the Internal Revenue Service to maintain robust enforcement efforts in an ever-changing technological landscape. Failure to take these steps will destine the size of the tax gap to expand. Many prior studies have addressed discreet effects of specific technologies on tax compliance. This paper contributes to this developing literature by offering a cohesive framework to address technological advancement and tax compliance.

Keywords: Technology, tax compliance, tax gap.

JEL-Classification: H24, H26.

I. INTRODUCTION

Since the introduction of the nation’s modern in 1913,1 tax noncompliance has been a persistent concern.2 Evidence for this proposition abounds. Court adjudications in which taxpayers have sought to shortchange the government could fill tomes;3 the Internal Revenue Service (Service) customarily dedicates a large proportion of its annual budget to help ensure taxpayer compliance;4 and Congress

* James Alm is an economics professor in the Department of Economics at Tulane University; Joyce Beebe is a fellow in public finance at Rice University’s Baker Institute for Public Policy; Michael S. Kirsch is a law professor at Notre Dame Law School; Omri Marian is a law professor and the academic director of the Graduate Tax Program at the University of California, Irvine School of Law; and Jay A. Soled is a tax professor in the Rutgers Business School at Rutgers University. All five authors have written and lectured about ways to close the tax gap. 1 See Revenue Act of 1913, ch. 16, 38 Stat. 114. 2 Tax noncompliance (or tax evasion) should be distinguished from . Avoidance represents activities by which taxpayers legitimately and legally reduce their tax burden. Evasion occurs when taxpayers purposefully do not pay what the Code instructs is their appropriate and legally due tax burden. See PAUL WEBLEY, HENRY ROBBEN, HENK ELFFERS & DICK HESSING, TAX EVASION: AN EXPERIMENTAL APPROACH 2 (J. Richard Eiser & Klaus R. Scherer eds., 1991); see also Bullen v. Wisconsin, 240 U.S. 625, 630–31 (1916) (“[W]hen the law draws a line, a case is on one side of it or the other, and if on the safe side is none the worse legally that a party has availed himself to the full of what the law permits [referring to tax avoidance]. When an act is condemned as an evasion, what is meant is that it is on the wrong side of the line indicated by the policy if not by the mere letter of the law.”). 3 For a sampling of such adjudications, see Steven R. Toscher, Dennis L. Perez, Charles P. Rettig & Edward M. Robbins Jr., Tax Crimes, U.S. INCOME PORTFOLIOS (BNA) No. 636 (2019). We do not claim that this list is exhaustive. 4 See INTERNAL REVENUE SERV., PUB. 55-B, DATA BOOK, 2018, at 65 tbl.28 (2019), https://www.irs.gov/pub/irs- pdf/p55b.pdf (indicating that the IRS dedicates around 40 percent of its budget to enforcement activities).

1

regularly enacts reform measures designed to curtail derelict taxpayer behavior, such as expanded third- party information returns,5 disclosure statements,6 and the imposition of penalties.7 Numbers confirm this narrative. The Service periodically estimates the size of the “tax gap” — the difference between what taxpayers pay to the U.S. government and what they would owe if they were fully compliant with the tax laws — and these estimates indicate a large, persistent, and growing problem, one whose magnitude is a major contributor to the country’s revenue shortfall. Over time, and in absolute dollar amounts, the tax gap has grown. The first Service estimate of the tax gap occurred in 1973 and indicated a tax gap of about $28 billion with an associated “voluntary compliance rate” (VCR), defined as taxes paid relative to total taxes legally due, of about 84 percent.8 Subsequent estimates throughout the 1970s and 1980s showed a steady growth in the tax gap with a fairly constant VCR of roughly 81–83 percent.9 By the time the Service released updated estimates for 1992 for the individual income tax, the gross tax gap had increased to over $93 billion.10 Most recently, the Service estimated the gross tax gap to be $441 billion annually (for tax years 2011–2013) with an associated VCR of 83.6 percent.11 Eliminating this gap would have come close to abolishing the nation’s annual budget deficit, at least before the recent explosion in the budget deficit following the enactment of the Tax Cuts and Jobs Act of 2017.12 Over the course of the past half-century, the nature of tax compliance and its counterpart tax noncompliance (or tax evasion) has changed. In this paper, we examine the past, present, and potential future trajectory of tax compliance. While many researchers have addressed specific technological advancements and their effects on tax compliance, this paper is the first to offer a cohesive framework to discuss this important issue. Part II examines the underpinnings of tax noncompliance and illustrates several of the twentieth century’s most prominent tax-noncompliance techniques. Parts III and IV discuss, respectively, the likely effects that technological changes will have on the government’s efforts to reduce tax noncompliance and taxpayers’ ability to circumvent those efforts. Part V concludes.

II. FUNDAMENTALS OF TAX NONCOMPLIANCE

Tax-noncompliance techniques often reflect the cultural and societal landscapes of the times. By way of example, consider the plight of taxpayers in ancient Egypt, which was then a largely agrarian society.

5 See, e.g., I.R.C. § 6045(g) (requiring brokers to track and report the tax basis that taxpayers have in their marketable securities). 6 See, e.g., id. § 6707A(c)(1) (delineating what constitutes a so-called reportable transaction). 7 See, e.g., id. § 6662(i) (applying a higher penalty for those transactions lacking economic substance). 8 See Internal Revenue Serv., Pub. 7285, Income Tax Compliance Research: Gross Tax Gap Estimates and Projections for 1973–1992 (1988), https://www.irs.gov/ pub/irs-soi/p7285388.pdf. 9 See Internal Revenue Serv., Pub. 1415, Federal Tax Compliance Research: Individual Income Tax Gap Estimates for 1985, 1988, and 1992, at 5, 6 (rev. Apr. 1996), https://www.irs.gov/pub/irs-soi/p141596.pdf; Jonathan Skinner & , An Economic Perspective on Tax Evasion, 38 Nat’l Tax J. 345, 345 (1985) (estimating a $90 billion tax gap for 1981). 10 See INTERNAL REVENUE SERV., supra note 9, at v. 11 See INTERNAL REVENUE SERV., PUB. 5364, TAX GAP ESTIMATES FOR TAX YEARS 2011–2013, at 1 (Sept. 2019), https://www.irs.gov/pub/irs-pdf/p5364.pdf. For earlier years’ estimates, see The Tax Gap, IRS (last updated Dec. 5, 2019), https://www.irs.gov/uac/the-tax-gap; I.R.S. News Release IR-2012-4 (Jan. 6, 2012). 12 See CONG. BUDGET OFFICE, THE BUDGET AND ECONOMIC OUTLOOK: 2016 TO 2026, at 144, summary tbl.1 (Jan. 2016), https://www.cbo.gov/publication/51129 (indicating that the fiscal year (FY) 2015 deficit for the federal government was $439 billion and rose to $544 billion in FY 2016 and projecting that the deficit will increase over the next decade). For more recent estimates, see CONG. BUDGET OFFICE, THE BUDGET AND ECONOMIC OUTLOOK: 2019 TO 2029, at 7, tbl.1-1 (Jan. 2019), https://www.cbo.gov/system/files/2019-03/54918-Outlook-3.pdf (indicating that the FY deficit for the federal government is likely to exceed $1 trillion in most years of the next decade).

2

When asked to pay taxes on their harvests and herds,13 there is little doubt that some taxpayers hid their produce and livestock, and, with limited resources, tax collectors were likely stymied in their attempts to uncover these taxpayer derelictions. However, tax noncompliance did not end in ancient Egypt. It merely morphed into other shapes and forms.14 Subpart A summarizes the central insights gleaned from academic research on tax noncompliance conducted over the last half-century. Subpart B details the conditions that allow tax noncompliance to flourish. Subpart C then summarizes prevalent twentieth-century tax-noncompliance techniques.

A. Insights from Academic Research on Tax Noncompliance

Comprehending the many ways in which changing technology affects tax compliance/noncompliance requires an understanding of why individuals and firms pay (compliance) — or do not pay (noncompliance) — their legally due taxes. In this subpart, we discuss the research that informs much of these phenomena, and we summarize its main results.15 Almost all of this research starts with a standard theoretical model based on the work of Gary S. Becker and his economics-of-crime model, first applied to tax noncompliance by Michael G. Allingham and Agnar Sandmo.16 Here, a rational individual is viewed as maximizing the expected utility of the tax evasion gamble (or lottery), weighing the benefits of successful cheating against the risk of detection and punishment. The standard conclusion from this approach is that an individual pays taxes because he is afraid of getting caught and penalized if he does not accurately report his income and bear the concomitant tax liability. This approach therefore leads to the plausible and rational conclusion that compliance depends upon audit and fine rates. Indeed, the central point of this approach is that an individual pays taxes because — and only because — of the economic consequences of detection and punishment. This is an important insight, with the obvious implication that the government can encourage greater tax compliance by increasing audit and penalty rates. However, there is a problem with the standard approach to tax compliance/noncompliance: some of its predictions are not strongly supported by the evidence. While its central premise assumes that compliance is driven entirely by financial considerations, it is clear that individuals are not motivated strictly by conducting cost-benefit analyses. The percentage of individual income tax returns that are subject to a thorough tax audit is generally quite small in most countries, almost always well below 1 percent of all returns. Even in a country like the United States, the Service reported an overall audit rate in 2017 of only 0.5 percent.17 Similarly, civil accuracy-related and fraud penalties are infrequently imposed.18 Because it is

13 See Joshua J. Mark, Ancient Egyptian Taxes & the Cattle Count, ANCIENT HIST. ENCYCLOPEDIA (2017), https://www.ancient.eu/article/1012/ancient-egyptian-taxes--the-cattle-count/. 14 See, e.g., CAROLYN WEBBER & AARON B. WILDAVSKY, HISTORY OF TAXATION AND EXPENDITURE IN THE WESTERN WORLD 141 (1986) (explaining that to avoid the imposition of a luxury tax, some ancient Romans secretly buried their jewelry and gold coins); Stephen J. Dubner & Steven D. Levitt, Filling in the Tax Gap, N.Y. TIMES (Apr. 2, 2006), https://www.nytimes.com/2006/04/02/magazine/filling-in-the-tax-gap.html (noting that when Congress mandated that the Social Security numbers of dependents be supplied, “seven million [such] dependents . . . suddenly vanished from the tax rolls, . . . [generating] nearly $3 billion in revenues in a single year”). 15 For recent surveys of this research, see, e.g., James Alm, What Motivates Tax Compliance?, 33 J. ECON. SURVEYS 353 (2019); and Joel Slemrod, Tax Compliance and Enforcement, 57 J. ECON. LITERATURE 904 (2019). 16 Alm, supra note 15, at 354; see Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. POL. ECON. 169 (1968); Michael G. Allingham & Agnar Sandmo, Income Tax Evasion: A Theoretical Analysis, 1 J. PUB. ECON. 323 (1972). 17 INTERNAL REVENUE SERV., supra note 4, at 21; see also Alain Sherter, Fear a Tax Audit by IRS? Don’t — The Odds Are with You, CBS NEWS (Mar. 8, 2019), https://www .cbsnews.com/news/2019-taxes-irs-audit-the-odds-are-with-you/ (“The main reason the IRS is doing fewer audits: years of funding cuts. The agency has reduced its staffing from 100,000 employees in 2010 to just over 79,000.”). 18 INTERNAL REVENUE SERV., supra note 4, at 42–43 tbl.17.

3

unlikely that such cheating will be caught or penalized even if brought to justice, a purely economic analysis of the evasion gamble suggests that most rational individuals should either underreport income not subject to third-party information reporting or overclaim deductions not subject to independent verification. However, as James Alm has written, “even in the least compliant countries, evasion seldom rises to levels predicted by a purely economic analysis, and, in fact, there are often substantial numbers of individuals in most countries who apparently pay all (or most) of their taxes all (or most) of the time,” regardless of the financial upside that evasion offers.19 In large part because of concerns associated with the economics-of-crime model and its shortcomings, academics have sought other explanations for tax compliance/noncompliance.20 One strand of research simply adds a range of considerations (e.g., employer withholding, labor supply decisions, complexity) that make the standard economics-of-crime model more realistic.21 Another strand of research expands the scope of this model by introducing some aspects of behavior considered explicitly by other social sciences, especially psychology, which change the ways in which an individual makes decisions (e.g., misperceived probabilities of audit, guilt and shame, and “rules of thumb” for decisions) and/or which introduce group considerations (e.g., fairness, altruism, and social norms).22 These extensions often improve the predictions of the theory, at the expense of simplicity. Overall, then, the theory of tax compliance/noncompliance suggests that enforcement matters. However, the very same theory also suggests that there are limits to the effects of greater enforcement on compliance and that there are other factors that motivate tax compliance. Regardless, all theories are, by definition, speculative. It is therefore necessary to examine available empirical evidence. At the start, it should be noted that empirical research on tax noncompliance is notoriously difficult to conduct for obvious reasons. After all, tax evasion is illegal, and, in light of potential financial and other penalties, such as prison sentences, taxpayers have strong incentives to be duplicitous. Even so, researchers have been increasingly creative in finding data to examine evasion, using naturally occurring field data, controlled field experiments, and laboratory experiments.23 Further, researchers have also been creative in

19 Alm, supra note 15, at 355. Problems with the standard economic approach are somewhat overstated, however. Alm writes thus: Consider the United States as an example. First, a standard feature of many individual income tax systems is that a third party (e.g., the individual’s employer) reports the relevant part of an individual’s to the tax authority (and often also withholds income taxes on this reported taxable income). This third-party information increases significantly the chances that an individual who underreports income on a filed return will be detected, especially in its combination with employer source-withholding. Second, the “official” IRS audit rate somewhat understates “actual” IRS audit policy. In fact, the IRS conducts a range of audits, and for many types of audits, the actual rates are quite high. Third and relatedly, while overall audit rates are quite low, among certain income and occupation classes, they are more frequent. Fourth, the IRS conducts a wide range of audit-type activities, including line matching and requests for information, and these activities are much more frequent [than full audits]. Finally, it is individual perceptions of audit rates that influence behavior, rather than actual audit rates, and individuals typically misperceive actual audit rates; indeed, it is common for individuals to believe that the audit rates that they face are substantially higher than the audit rates that actually apply to them. Id. at 356. For information on audit rates and policies, see INTERNAL REVENUE SERV., supra note 4, at 21–34. For information on perceptions of audit rates and policies, see Omri Marian & Joshua Blank, Reexamining Tax Compliance, in THE OXFORD HANDBOOK ON COMPLIANCE (forthcoming 2020); and ERICH KIRCHLER, THE ECONOMIC PSYCHOLOGY OF TAX BEHAVIOR (2007). 20 For a recent summary of this literature, see Joshua D. Blank, Collateral Compliance, 162 U. PENN L. REV. 719, 747–48 (2014). 21 See, e.g., James Alm, Jeremy Clark & Kara Leibel, Enforcement, Socio-Economic Diversity, and Tax Filing Compliance in the United States, 82 S. ECON. J. 725 (2016). 22 See, e.g., Nigar Hashimzade, Gareth D. Myles & Binh Tran-Nam, Applications of Behavioral Economics to Tax Evasion, 27 J. ECON. SURVEYS 941 (2013). 23 See, e.g., Alm, supra note 15; Slemrod, supra note 15; Joel Slemrod & Caroline Weber, Evidence of the Invisible: Toward a Credibility Revolution in the Empirical Analysis of Tax Evasion and the Informal Economy, 19 INT’L TAX & PUB. FIN. 25 (2012).

4

finding new methods (e.g., utilizing withholding records) to analyze tax noncompliance.24 These data and methods have provided many important insights. A distillation of empirical tax research reveals twelve basic findings, as detailed below:25

COMPENDIUM OF EMPIRICAL RESEARCH FINDINGS REGARDING TAX COMPLIANCE (1) Both the level and the type of audit matter — a lot. There is strong evidence that more audits increase compliance.26 Audits also typically have a spillover effect of augmenting general deterrence, which enhances overall compliance independent of revenues generated directly from the audits themselves.27 (2) Perceptions of audit rates affect behavior; that is, cognitive considerations matter. Taxpayers appear to substantially misperceive audit rates, typically overweighing a (low) probability of audit.28 (3) Fines, whether financial or nonfinancial, affect compliance, but their deterrent effects are often small. The limited amount of work on fines typically finds that a higher fine rate leads to marginally more compliance, with an estimated reported income-fine rate elasticity of less than 0.1.29 Of some note, there is evidence that nonfinancial penalties (e.g., public disclosure and “shame”) may also function as a deterrent. (4) Tax rates affect compliance, but the effects are nuanced. The level of tax rates matters in a taxpayer’s compliance decision, with an increase in tax rates generally leading to greater noncompliance. In addition, one’s relative to others’ matters; that is, if a taxpayer believes that his or her tax rate is too high relative to others, being less compliant is a common response. (5) Positive inducements, whether to individuals or to groups, improve compliance. There is evidence that when a government grants rewards for tax accuracy (e.g., social insurance benefits) or honesty (e.g., entry into a lottery), these efforts decrease noncompliance. Similarly, if a group of taxpayers receive a benefit (e.g., an increase in public goods), tax compliance generally improves. (6) The social and institutional environment in which individuals live affects compliance. The overall setting in which a taxpayer lives, works, and functions has important effects on tax compliance, consistently demonstrated by empirical findings of differences in behavior among countries with similar fiscal systems (e.g., tax rates, audit rates, and fine rates) but different social and institutional environments. One compelling explanation for these differences in compliance behavior is that there seems to be a “social norm” of compliance, in which one’s compliance depends upon various factors that reflect and capture the many aspects of one’s environment. Consistent with this perspective, when

24 See, e.g., Slemrod & Weber, supra note 23. 25 This summary draws heavily upon Alm, supra note 15. 26 The reported income-audit rate elasticity generally falls between 0.2 and 0.4, meaning that a 10 percent increase in the audit rate would yield an increase of reported income by 2–4 percent. 27 The increased compliance varies from 4 to 12 in magnitude, such that for every $1 of additional direct revenue generated from an audited taxpayer, another $4 to $12 in additional revenue is generated from other, nonaudited individuals. 28 However, some recent work emphasizes that one must distinguish between the effects of audits on taxpayers found to be compliant and those found to be noncompliant: compliant taxpayers who are found to have additional tax liabilities after an audit tend to become more compliant, while noncompliant taxpayers with no additional assessment tend to become less compliant after an audit. One explanation for this phenomenon is that deterrence may diminish a taxpayer’s “intrinsic motivation” to pay taxes; in other words, an individual who is otherwise inclined to pay voluntarily his or her own taxes may, as an act of defiance, respond to the government’s brute force efforts to enforce compliance with audits and fines by reducing compliance. Another explanation is that taxpayers may update their subjective audit probabilities following an audit. 29 This elasticity rate suggests that a 10 percent increase in the penalty rate would yield an increase of reported income by only 1 percent.

5

individual trust in government is greater, enforcement tends to be more effective at deterring tax noncompliance. (7) Individual participation in the choice of institutions affects compliance; that is, process (versus outcome) is an essential determinant of compliance. Independent of the actual levels of tax, audit, and fine rates, individual taxpayer participation in the choice of institutions — the process as distinct from the outcome — improves tax compliance. For example, taxpayers exhibit a greater propensity to pay their taxes when they choose how their taxes are spent compared to when an identical use is imposed upon them. (8) The information that tax authorities have on income sources is an essential component of a compliance strategy. Noncompliance is far lower on income that is subject to employer withholding and to third-party information sources than on income not subject to these features.30 The reason is obvious: taxpayers know that the government can cross-check their submissions for accuracy. (9) Information availability affects tax compliance in sometimes surprising ways. The information that taxpayers receive about the tax system and the role of other taxpayers in it affects compliance. For example, telling taxpayers (via a message) that they will be closely examined generally reduces their noncompliance rate compared to those who do not receive this message. Relatedly, knowing what their “neighbors” are doing affects taxpayers’ own decisions, and not always in a way that decreases cheating: if taxpayers know that their neighbors are derelict in fulfilling their civic duties, they are more apt to do the same. (10) The knowledge (or understanding) that taxpayers have about the tax system affects compliance, but the overall impact is unresolved. Given a complex and uncertain tax system, taxpayers often do not know what they should pay in taxes. As a result, taxpayers increasingly have come to rely upon paid tax practitioners and tax-preparation software.31 Perhaps surprisingly, the data indicate that noncompliance is generally higher for returns prepared by tax practitioners. Accordingly, a complicated tax system often tends to increase noncompliance; conversely, a simpler tax system with better administrative services that make it easier for taxpayers to pay taxes tends to reduce noncompliance. (11) Demographics matter. There is consistent evidence that compliance may be affected by numerous demographic variables. For example, noncompliance tends to be higher for individuals who are younger, single, self-employed, and male. The effects of most other demographic variables are uncertain. (12) Individuals are motivated in their compliance decisions by many factors beyond narrow financial self-interest. Individuals who are identified as expressing greater capacities for sympathy and empathy tend to be more tax compliant. Other motivations have also been found to affect compliance; for example, individuals who exhibit greater patriotism are often more tax compliant. Relatedly, there is some evidence that cheating on one’s taxes creates emotional distress. ______

In summary, while the evidence indicates that taxpayers are motivated by narrowly defined, and individually based, financial considerations (e.g., audit rates and risk of penalty imposition), it also indicates that they are driven by nonfinancial considerations (e.g., sympathy, empathy, guilt, shame, and morality).

30 See INTERNAL REVENUE SERV., TAX GAP ESTIMATES FOR TAX YEARS 2008–2010, at 2 (Apr. 2016), https://www.irs.gov/pub/newsroom/tax%20gap%20estimates%20for%202008 %20through%202010.pdf. 31 Fifty-six percent of tax returns were prepared with the help of a paid tax preparer, while another 34 percent were completed using tax-preparation software. Protecting Taxpayers from Incompetent and Unethical Return Preparers: Hearing Before the S. Fin. Comm., 113th Cong. 131 (2014) (written testimony of John A. Koskinen, Commissioner of the IRS), https://www.govinfo.gov/content/pkg/CHRG-113shrg90922/html/CHRG-113shrg 90922.htm.

6

Further, there is evidence that taxpayers are spurred by social considerations (e.g., social norms, public good, voting, and neighbor behavior). There is also evidence that tax-related information may cause taxpayers to be compliant, albeit it depends upon how they process such information. Finally, due to taxpayer heterogeneity, it is not easy to predict how they will respond to the institution of different compliance measures. The empirical research findings have important implications for government strategies to address tax noncompliance, suggesting three paradigms for tax administration.32 • Under the first paradigm, the traditional “enforcement paradigm,” the emphasis is based exclusively on repression of illegal behavior through frequent audits and stiff penalties. Throughout the history of civil societies, this has been the conventional paradigm of tax administrations, and it fits with the standard portfolio model of tax evasion based upon the economics-of-crime theory. • The second paradigm acknowledges the role of enforcement but also recognizes the role of tax administration as a facilitator and a provider of services to taxpayer citizens, specifically, assisting taxpayers in every step of filing returns and paying taxes. This new “service paradigm” for tax administration fits squarely with the perspective that emphasizes the role of government-provided services as a consideration in the individual tax-compliance decision. Indeed, the most recent literature on tax administration reform has emphasized this new paradigm for tax administration, and many recent administrative reforms around the world have embraced this new paradigm with measured success. • The third paradigm sees the taxpayer as a member of a larger group — as a social creature whose behavior depends upon his or her own moral values (and those of others) and also upon his or her perception of the quality, credibility, and reliability of the tax administration; this approach is a “trust paradigm.” Consistent with the role of various behavioral economics factors (e.g., social norms) in the compliance decision, it is based on the notion that individuals are more likely to respond either to enforcement or to services if they harbor faith in the government generally and the tax administration specifically as honest and if they believe that other individuals are similarly motivated. In sum, trust in the authorities and in other individuals can have a positive impact on compliance. Strategies to control tax noncompliance should thus encompass the three foregoing paradigms: increasing the likelihood and the threat of punishment, improving the provision of tax services, and changing the tax culture. As discussed next, this empirical research helps lay the foundation for a more practical discussion of the major real-world elements that drive tax noncompliance.

B. Applying Academic Research to Tax-Noncompliance Behavior

For tax noncompliance to flourish, certain conditions must exist. Enumerated below, the following six conditions are most prevalent: (1) lack of effective oversight, (2) high tax rates, (3) taxpayers unmoored to social mores, (4) taxpayers’ proclivity toward greed, (5) presence of skilled advisers, and (6) compliance measures that fail to recognize taxpayers’ diverse natures. These conditions reflect the results of the academic research discussed above regarding actual tax-noncompliance behavior.

32 James Alm & Benno Torgler, Do Ethics Matter? Tax Compliance and Morality, 101 J. BUS. ETHICS 635, 646–47 (2012).

7

1. Lack of Effective Oversight

A condition that contributes to tax noncompliance is a lack of effective oversight (see Findings #1, #2, #3, and #8 above). With the passage of the Sixteenth Amendment to the U.S. Constitution, the modern U.S. income tax came into existence in 1913.33 At the time, tax information returns (e.g., Form 1099s and W- 2s) were nonexistent, and the only electronic quality about the tax return submission process was that taxpayers, due to the advent of the incandescent light bulb, no longer had to make their calculations by candlelight. As a result, the primary method employed by the Service to address tax noncompliance was to retain more tax auditors.34 In order to discover whether taxpayer defalcations had been committed, the Service deployed its agents to conduct audits of taxpayers’ books and records.35 Although the historical record regarding the number of Service audits actually conducted during the infancy of the modern income tax is scant, it is easy to imagine, given the overwhelming strain on the Service, that the agency conducted very few of them.36 In the absence of a meaningful number of tax audits, tax noncompliance was likely commonplace.37 And this lack of oversight trend has continued: for the last decade or so, the Service remains grossly underfunded and unable to conduct an adequate number of audits to ensure tax compliance.38 Penalties are also relevant here. In cases of minor violations, the Code imposes accuracy-related penalties;39 in cases of egregious violations, the Code’s penalties include possible incarceration.40 Even so, these penalties tend to be relatively small (in most instances, 20 percent) and are infrequently imposed.41 A final consideration in oversight is the presence of mechanisms that enable individuals to evade detection.42 Taxpayers who engage in tax noncompliance do not want the government to detect their

33 See Revenue Act of 1913, ch. 16, 38 Stat. 114. 34 See, e.g., Stacy Cowley, Why the I.R.S. Fails to Crack the Small-Business Tax Nut, N.Y. TIMES (June 15, 2016), https://www.nytimes.com/2016/06/16/business/smallbusiness /why-the-irs-fails-to-crack-the-small-business-tax-nut.html (“The dreaded audit is the main way the I.R.S. catches scofflaws and ferrets out unreported income, but it is a time-consuming and imperfect tool.”). 35 For a historical overview of the IRS, see JASON BAROUSSE, THE INTERNAL REVENUE SERVICE: WHY U.S. CITIZENS PAY TAXES (Elizabeth Krajnik ed., 2018). 36 See Tax History Museum: 1901–1932, TAX ANALYSTS, http://www.taxhistory.org/ www/website.nsf/Web/THM1901?OpenDocument (“The number of returns filed in 1918 was five times greater than the number from 1917. Subsequent increases only added to the burden. All told, the number of returns increased more than 1,000 percent between 1916 and 1921, giving the BIR an impossible problem. ‘The enormous increase in the revenue,’ one BIR commissioner complained, ‘the overwhelming increase in the number of returns filed and increase in the work to be performed as a consequence thereof went by leaps and bounds. No one did or could foresee it, or prepare for it.’”). 37 The more audits the IRS conducts, the more likely taxpayers will be held accountable; even if taxpayers are not audited, the mere prospect that they might be audited tends to generate tax compliance. See, e.g., JEFFREY A. DUBIN, THE CAUSES AND CONSEQUENCES OF INCOME TAX NONCOMPLIANCE 81 (2012); Jeffrey A. Dubin, Michael Graetz & Louis Wilde, The Effect of Audit Rates on the Federal Individual Income Tax, 1977–1986, 43 NAT’L TAX J. 395 (1990). 38 See, e.g., Kate Gibson, Auditing Taxpayers Is a Dying Tradition at Depleted IRS, CBS NEWS (Oct. 5, 2018), https://www.cbsnews.com/news/auditing-taxpayers-is-a-dying-tradition-at-depleted-irs/ (“Starting in 2011, the GOP-led Congress repeatedly slashed the IRS budget, which fell from $12.1 billion in 2011 to $11.2 billion by 2017. Over that time the agency thinned the ranks of its enforcement personnel by roughly a third.”). 39 See, e.g., I.R.C. § 6662(a) (applying a 20 percent penalty for inaccurate tax-reporting practices). 40 See, e.g., id. §§ 7201–05 (delineating a series of potential tax crimes–related acts perpetrated to purposefully evade one’s tax obligations). 41 See supra note 18 and accompanying text. 42 Note that in the presence of third-party tax information returns and employer source withholding, see I.R.C. §§ 6051–53 (listing over twenty-five instances that require the issuance of third-party information returns), tax noncompliance is virtually nonexistent. See INTERNAL REVENUE SERV., supra note 30 (finding that for nonfarm individual proprietor income for which there is no third-party reporting, the misreporting rate was 63 percent, whereas for wages, salaries, and tips for which there is third-party

8

cheating, and they take various measures to hide their actions. Such measures vary. Some are simple in nature. For example, as part of their business practice, a number of taxpayers keep two sets of “books”: one reflects all income, and the other depicts only noncash receipts — the latter is used for tax-reporting purposes.43 Other mechanisms are far more complex in nature. For example, in their investment practices, some taxpayers have established grantor trusts as devices to mask their receipt of income.44

2. High Tax Rates

Another factor associated with tax noncompliance is the tax rate level (see Finding #4 above). In general, the higher the tax rate imposed on income, the greater the incentive to be tax noncompliant.45 Put somewhat differently, when tax rates are high, the tax savings associated with tax noncompliance are also high; by the same token, little may be gained from cheating on one’s taxes when tax rates are low. However, it should be noted that the empirical evidence on the impact of tax rates on tax noncompliance is somewhat ambiguous.46

3. Taxpayers Unmoored to Social Mores

A third factor is the relationship between the individual and society, including individual perceptions regarding the Code’s equity (see Findings #5, #6, #7, and #9 above).47 If taxpayers perceive that the Code provides a fair fiscal exchange between parties, they are apt to be compliant; in contrast, if the perception is that the fiscal exchange is stacked against them, individuals tend to be noncompliant.48 Put differently, taxpayers are social creatures. As such, their interactions with other taxpayers affect their behavior, and cultural factors such as religious and political party affiliations can also affect their behavior.49 For example, reporting, the misreporting rate was 1 percent); see also Tax Compliance: Multiple Approaches Are Needed to Reduce the Tax Gap: Testimony Before the S. Comm. on the Budget, 110th Cong. 154 (2007) (statement of Michael Brostek, Director, Tax Issues, Strategic Issues Team) (asserting that withholding and information returns are particularly powerful tools to reduce the tax gap); JON BAKIJA & JOEL SLEMROD, TAXING OURSELVES: A CITIZEN’S GUIDE TO THE DEBATE OVER TAXES 178 (3d ed. 2004) (describing a study showing a 99.1 percent compliance rate for wages and salaries and an 18.6 percent compliance rate for informal suppliers (e.g., sidewalk vendors and housepainters)). 43 See, e.g., Potter v. Commissioner, 107 T.C.M. (CCH) 1101 (2014) (corporate owner prepared two sets of books, one reflecting actual income earned and the other used to report income for tax purposes). 44 See I.R.S. Notice 2000-44, 2000-36 C.B. 255 (the IRS declares that those taxpayers and their advisers who purposefully use a grantor trust to conceal their activities risk criminal liability). 45 See, e.g., Charles T. Clotfelter, Tax Evasion and Tax Rates: An Analysis of Individual Returns, 65 REV. ECON. & STAT. 363 (1983); David Joulfaian & Mark Rider, Tax Evasion in the Presence of Rates, 49 NAT’L TAX J. 553 (1996); Karyl A. Kinsey & Howard G. Grasmick, Did the Act of 1986 Improve Compliance? Three Studies of Pre- and Post-TRA Compliance Attitudes, 15 LAW & POL’Y 293, 318–19 (1993); Michael L. Roberts & Peggy A. Hite, Progressive Taxation, Fairness, and Compliance, 16 LAW & POL’Y 27, 38 (1994). 46 See Alm, supra note 15, at 366. Some studies exist showing that higher marginal tax rates resulted in greater compliance. See, e.g., James Alm, Roy Bahl & Matthew N. Murray, Audit Selection and Income Tax Underreporting in the Tax Compliance Game, 42 J. DEV. ECON. 1 (1993); Jonathan S. Feinstein, An Econometric Analysis of Income Tax Evasion and Its Detection, 22 RAND J. ECON. 14 (1991). 47 See TOM R. TYLER, WHY PEOPLE OBEY THE LAW 178 (1990) (arguing that those laws that are perceived to be equitable are more apt to be obeyed). 48 See, e.g., James Alm, Gary McClelland & William Schulze, Changing the Social Norm of Tax Compliance by Voting, 52 KYKLOS 141, 160 (1999); James Alm, Betty R. Jackson & Michael McKee, Estimating the Determinants of Taxpayer Compliance with Experimental Data, 45 NAT’L TAX J. 107 (1992); James Alm, Betty R. Jackson & Michael McKee, Fiscal Exchange, Collective Decision Institutions, and Tax Compliance, 22 J. ECON. BEHAV. & ORG. 285 (1993); James Alm, Betty R. Jackson & Michael McKee, Institutional Uncertainty and Taxpayer Compliance, 82 AM. ECON. REV. 1018 (1992). 49 See, e.g., Harold G. Grasmick, Robert J. Bursik Jr. & John K. Cochran, “Render Unto Caesar What Is Caesar’s”: Religiosity and Taxpayers’ Inclinations to Cheat, 32 SOC. Q. 251, 263 (1991) (religiosity affects taxpayers’ propensity to cheat).

9

if the perception is that tax compliance is only for chumps and that tax noncompliance is in vogue, this sentiment may become the prevailing norm and dominant social norm.50

4. Taxpayers’ Proclivity Toward Greed

A fourth and related condition is the taxpayer’s motivation or mindset (see Finding #12 above). There are numerous studies on why some taxpayers choose to be tax noncompliant.51 As noted above, many focus on the deterrence approach to tax compliance in which risk detection and penalty application are weighed against the financial benefit of being tax noncompliant; under this theory, rational and wealth-maximizing taxpayers will cheat if the latter benefits outweigh the former costs.52 Repeated empirical studies demonstrate that financial incentives of noncompliance clearly matter.53 Given the relatively low audit and penalty rates, these financial incentives often create a condition that encourages tax noncompliance, occurring more often and in larger amounts, by higher-income taxpayers.54 Despite the presence of other studies that demonstrate the relevance of other, nonfinancial motivations, the financial incentives of noncompliance are crucial factors in tax noncompliance.

50 See, e.g., Richard Lavoie, Flying Above the Law and Below the Radar: Instilling a Taxpaying Ethos in Those Playing by Their Own Rules, 29 PACE L. REV. 637, 638 (2009) (“That is, taxpayers pay because they believe that they receive a benefit from the government for paying their taxes, and they are willing to keep paying as long as the government keeps providing valued services and other citizens are not shirking their duties.”); James Alm & Jorge Martinez-Vazquez, Institutions, Paradigms, and Tax Evasion in Developing and Transition Countries, in PUBLIC FINANCE IN DEVELOPING AND TRANSITIONAL COUNTRIES 151 (2003) (“If the perception becomes widespread that the government is not willing to detect and penalize evaders, then such a perception legitimizes tax evasion. The rejection of sanctions sends a signal to each individual that others do not wish to enforce the tax laws and that tax evasion is in some sense socially acceptable, and the social norm of compliance disappears.”); Michael P. Vandenbergh, Beyond Elegance: A Testable Typology of Social Norms in Corporate Environmental Compliance, 22 STAN. ENVTL. L.J. 55, 113 (2003) (“The effect of perceptions of widespread [tax] noncompliance on intentions to comply in the future may result from the norm of conformity, or may simply be the product of a perceived reduction in the risk of formal or informal sanctions.”). By way of contrast, patriotism and love of one’s nation may drive tax compliance. See generally Richard Lavoie, Patriotism and Taxation: The Tax Compliance Implications of the Tea Party Movement, 45 LOY. L.A. L. REV. 39 (2011). 51 These studies vary tremendously in scope. For a comprehensive, if somewhat dated, literature survey, see James Andreoni, Brian Erard & Jonathan Feinstein, Tax Compliance, 36 J. ECON. LITERATURE 818 (1998) (summarizing various tax-compliance studies). See, e.g., Alm, supra note 15; Mary Margaret Frank, Luann J. Lynch & Sonja O. Rego, Tax Reporting Aggressiveness and Its Relation to Aggressive Financial Reporting, 84 ACCT. REV. 467 (2009); Sonja O. Rego, Tax Avoidance Activities of U.S. Multinational Corporations, 20 CONTEMP. ACCT. RES. 805 (2003); Slemrod, supra note 15; Ryan J. Wilson, An Examination of Shelter Participants, 84 ACCT. REV. 969 (2009). 52 See Becker, supra note 16; Allingham & Sandmo, supra note 16. However, many other theories of tax noncompliance exist, including theories that introduce other individual motivations, such as the failure to consider accurately the risks of cheating and the inability of government to provide public goods. See Werner W. Pommerehne, Albert Hart & Bruno S. Frey, Tax Morale, Tax Evasion and the Choice of Policy Instruments in Different Political Systems, 49 PUB. FIN. 52 (Supp. 1994). By way of contrast, if a government fulfills its obligations to deliver public goods, tax compliance is apt to improve. See, e.g., supra note 50 and accompanying text. 53 See, e.g., Edward J. McCaffery, Cognitive Theory and Tax, 41 UCLA L. REV. 1861, 1878 (1994) (“[T]here may be a phenomenon of ‘tax aversion,’ akin to but distinct from loss aversion, whereby individuals attach disproportionate disutility to government extractions perceived or labeled as ‘taxes.’”); Joshua D. Rosenberg, The Psychology of Taxes: Why They Drive Us Crazy and How We Can Make Them Sane, 16 VA. TAX REV. 155, 179 (1996) (“Of course, one of the reasons that we do not tend to think about what we are buying with our tax payments is that the person who pays taxes gets no more services or protection than the person with equal assets who can avoid paying taxes. Unlike the individual who pays for specific goods, and gets what she pays for, the taxpayer receives benefits that neither correspond to nor depend on the size of her own contribution: government will function, and individuals will receive the benefits of government, whether or not any individual pays her tax liability.”). See generally Christopher C. Fennell & Lee Anne Fennell, Fear and Greed in : A Qualitative Research Agenda, 13 J.L. & POL’Y 75 (2003); KIRCHLER, supra note 19. 54 See generally Andrew Johns & Joel Slemrod, The Distribution of Income Tax Noncompliance, 63 NAT’L TAX J. 397, 413 (2010) (“[W]hen taxpayers are arrayed by their ‘true’ income, the ratio of aggregate misreported income to true income generally increases with income, although it peaks among taxpayers with adjusted gross income between $500,000 to $1,000,000.”).

10

5. Presence of Skilled Advisers

A fifth condition is the presence of skilled advisers (see Finding #10 above). For many taxpayers, the Code is an impenetrable morass of words and instructions. Even if they want to evade their tax obligations, they lack the skills to do so. They therefore routinely turn to tax and other professionals for planning advice. The Code is supposed to curtail tax professionals and others from rendering advice that would result in tax evasion;55 however, this deterrence is not always effective. Although both the Code and Circular 230 impose steep penalties,56 including criminal liability,57 on those who abet tax evasion, at least some tax- service providers are willing to assume the risks associated with the imposition of these penalties and entice their clientele to take aggressive tax-defeating measures.58

6. Compliance Measures That Fail to Recognize Taxpayers’ Diverse Natures

Finally, it must always be remembered that individuals are complicated (see Findings #11 and #12 above). There is no typical individual who responds predictably and reliably to all policies. People are motivated by many different factors, and responsive (if at all) in different ways. In this regard, the anthropologist Steven J. Gould emphasized that most systems have incredible variety — a “full house” of individual behaviors — and the proper understanding of any system requires recognition of this basic fact.59 This lesson seems especially apt for tax noncompliance: people exhibit a remarkable diversity in their behaviors, and Congress, in crafting administrative-compliance measures, needs to recognize this fact and account for it when devising legislative mandates. As suggested earlier in the discussion about tax administration paradigms, there should be a “full house” of strategies to address the “full house” of taxpayers’ motivations.

C. Major Twentieth-Century Tax-Noncompliance Strategies

When one or more of the aforementioned six conditions exist, tax noncompliance can thrive. Certainly, this was the case during the twentieth century, when technology played only a minor role in thwarting noncompliance — and, by the same token, a marginal role in promoting compliance. Prior to the modern U.S. income tax coming into existence in 1913,60 the mainstays of federal -raising measures consisted of taxes, tariffs, duties, and public land

55 See, e.g., Heather M. Field, Aggressive Tax Planning & the Ethical Lawyer, 36 VA. TAX REV. 261 (2017) (describing the difficult ethical balance that lawyers must strike when rendering tax advice to those taxpayers who seek to pursue aggressive tax strategies); Charles Fried, The Lawyer as Friend: The Moral Foundations of the Lawyer-Client Relation, 85 YALE L.J. 1060 (1976) (same); Tanina Rostain, Sheltering Lawyers: The Organized Tax Bar and the Tax Shelter Industry, 23 YALE J. ON REG. 77 (2006) (same). 56 I.R.C. § 6701; 31 C.F.R. §§ 10.50–10.53. 57 I.R.C. § 7206(2). 58 There are a number of adjudications in which criminal actions were pursued against tax shelter promoters and those who provided erroneous advice knowing that it would be used for tax return preparation. See, e.g., United States v. Bryan, 896 F.2d 68 (5th Cir. 1990); United States v. Flomenhoft, 714 F.2d 708 (7th Cir. 1983); United States v. Dahlstrom, 713 F.2d 1423 (9th Cir. 1983); United States v. Crum, 529 F.2d 1380 (9th Cir. 1976). 59 See Stephen J. Gould, Full House: The Spread of Excellence from Plato to Darwin, passim (1996). 60 See Revenue Act of 1913, ch. 16, 38 Stat. 114.

11

sales.61 Little data exists regarding tax noncompliance during this time.62 Even so, there is a long history of individual resistance to taxation, which manifested itself in a variety of ways, including armed uprisings.63 With the introduction of the income tax in 1913, taxpayers who chose noncompliance had to adapt to a new landscape, learning which tax-evasion strategies worked and which did not. Although many strategies eventually emerged (e.g., failing to pay employment taxes with respect to the use of household help64 and reporting the existence of fictitious dependents65), three strategies of tax noncompliance gained particular prominence: (1) the failure to report cash payments and receipts, (2) the use of sophisticated tax shelters to manufacture noneconomic losses, and (3) the establishment of hidden offshore accounts.

1. The Failure to Report Cash Payments and Receipts

The use of cash has a proven track record as a driving force that propels economies. In its absence, people are forced to barter, finding a “double coincidence of wants” in which two trading parties each own exactly what the other party wishes to possess in quantities that are perceived to be of equivalent value.66 The same is not true of cash. It opens trading options across cities, states, and countries, igniting commerce.67 From the government’s perspective, however, the use of cash is a mixed blessing. On the one hand, the government is the benefactor of robust commerce as taxpayers’ earnings climb, standing to yield hefty tax-revenue collections. A government that issues money also generates revenues from money creation, sometimes referred to as “seigniorage.”68 On the other hand, taxpayers can seek to hide cash payments and

61 See THOMAS L. HUNGERFORD, CONG. RESEARCH SERVICE, RL 33665, U.S. FEDERAL GOVERNMENT REVENUES: 1790 TO THE PRESENT (2006), https://www.everycrsreport.com/ reports/RL33665.html. 62 See U.S. DEP’T OF THE TREASURY, IRS HISTORICAL FACT BOOK: A CHRONOLOGY, 1646–1992, at 52 (1993), https://www.governmentattic.org/5docs/IRSHistoricalFactBook _1992.pdf (noting that in 1874, for example, “[t]he number of deputy collectors was decreased from three to two”). 63 See, e.g., Whiskey Rebellion, ENCYCLOPEDIA BRITANNICA, https://www.britannica. com/event/Whiskey-Rebellion (last visited Feb. 25, 2020) (“Enforcement legislation touched off what appeared to be an organized rebellion, and in July of 1794 about 500 armed men attacked and burned the home of the regional tax inspector after a smaller group had been fended off the previous day.”). 64 See, e.g., David Cay Johnston, Despite an Easing of Rules, Millions Evade ‘Nanny Tax,’ N.Y. TIMES (Apr. 5, 1998), https://www.nytimes.com/1998/04/05/business/despite-an-easing-of-rules-millions-evade-nanny-tax.html. 65 See, e.g., Sara LaLulia & James M. Sallee, The Value of Honesty: Empirical Estimates from the Case of the Missing Children, INT’L TAX & PUB. FIN. (Apr. 5, 2012), https://nature.berkeley.edu/~sallee/lalumia-sallee.pdf (“To claim a dependent prior to 1987, a filer needed only to list the dependent’s first name on his tax return. Since the Internal Revenue Service had no easy way to verify that these dependents existed or to ensure that they were not listed on multiple returns, the system may have tempted filers to either invent dependents or to claim ineligible individuals as dependents.”). 66 See Theodoros Karasavvas, From Barter to Bitcoins: The 5,000 Year History of Money, ANCIENT ORIGINS (July 18, 2018), https://www.ancient-origins.net/history-ancient-traditions/barter-bills-bitcoins-history-007109 (“In ancient times, people did not buy or sell with money. Instead they traded one thing for another to get what they wanted or needed. . . . At the beginning [barter] was sufficient, but with the evolution of societies, certain needs made the creation of a more stable and more efficient way of trading imperative.”). 67 See Chapurukha Kusimba, When — and Why — Did People First Start Using Money?, CONVERSATION (June 19, 2017), http://theconversation.com/when-and-why-did-people-first-start-using-money-78887 (“[Cash] facilitates exchange as a measure of value; it brings diverse societies together by enabling gift-giving and reciprocity; it perpetuates social hierarchies; and finally, it is a medium of state power.”). 68 See, e.g., Stanley Fischer, Seigniorage and the Case for a National Money, 90 J. POL. ECON. 295 (1982). Estimates of the amount of money that is generated for the government vary but can sometimes approach 1 percent of GDP in the United States. See, e.g., Robert J. Barro, Measuring the Fed’s Revenue from Money Creation, 10 ECON. LETTERS 327, 332 (1982); Manfred J.M. Neumann, Seigniorage in the United States: How Much Does the U.S. Government Make from Money Production?, REVIEW 29 (Mar./Apr. 1992), https://doi.org/10.20955/r.74.29-40.

12

receipts, potentially subverting tax collections.69 Indeed, cash payments and receipts have proven too much of a temptation for many individuals seeking to evade taxes. During the course of the twentieth century, the Service has brought case after case after case against taxpayers who, in the endeavor to minimize their tax burdens,70 utilized cash payments71 and/or cash receipts72 to hide their transactions.73 Uncovering taxpayers’ failures to report cash properly is not easy. In large part, this is why this mode of tax cheating has flourished. In many instances, the Service must resort to the so-called bank deposit method.74 If the Service can show that a man has a business or calling of a lucrative nature and is constantly, day by day and month by month, receiving moneys and depositing them to his account and checking against them for his own uses, there is most potent testimony that he has income, and, if the amount exceeds exemptions and deductions, that the income is taxable.75 The linchpin of this method compares the bank deposit amounts of the taxpayer to the taxpayer’s income reported for tax purposes; in those instances where there are significant discrepancies, the Service commences an audit.76 Due to the labor-intensive nature of the bank deposit method, this sort of oversight is not commonplace.

2. The Use of Sophisticated Tax Shelters to Manufacture Noneconomic Losses

When it comes to the obligation to pay taxes, there are many adages that are bantered around. One of the more common ones is that “[a]ny one may so arrange his affairs that his taxes shall be as low as possible; he is not bound to choose that pattern which will best pay the Treasury; there is not even a patriotic to increase one’s taxes.”77

69 See, e.g., Robert W. Wood, Paying in Cash? Careful, It Can Mean Jail, FORBES (July 24, 2014), https://www.forbes.com/sites/robertwood/2014/07/24/paying-in-cash-careful-it-can-mean-jail/#457618ba4dc2 (describing how taxpayers sometime use cash payments in an attempt to cheat on their tax burden). 70 See generally Ronald R. Hiner & Darlene Pulliam Smith, Underreported Income, J. ACCT. (Mar. 1, 2002), https://www.journalofaccountancy.com/issues/2002/mar/underreported income.html. 71 See, e.g., John Hilton, Harrisburg Businessman Sentenced to 37 Months in Prison for Defrauding the IRS Out of $1M, CENT. PENN BUS. J. (Apr. 24, 2015), http://www.cpbj.com /article/20150424/CPBJ01/150429834/harrisburg-businessman- sentenced-to-37-months-in-prison-for-defrauding-the-irs-out-of-1m (reporting that the U.S. Attorney’s Office for the Middle District of Pennsylvania brought action against a group of business owners and their various business operations that supplied personnel services for paying “cash wages of more than $7 million to their employees without withholding any employment taxes, such as for Social Security and Medicare. . . .”). 72 See, e.g., Commissioner v. Wu, 83 T.C.M. (CCH) 1363 (2002) (relating to a taxpayer who operated a computer sales and repair corporation and instructed his customers to pay in cash, leave the payee lines of their checks blank, write the taxpayer’s name (rather than the business name) on the check payee line, and write in the check memo section the word “loan”; ruling that the taxpayer was liable for tax on the cash that he received, the court imposed a civil tax fraud penalty). 73 See Susan Cleary Morse, Stewart Karlinsky & Joseph Bankman, Cash Businesses and Tax Evasion, 20 STAN. L. & POL’Y REV. 37 (2009) (cash businesses are prone to evading their taxes); Joseph Bankman, Eight Truths About Collecting Taxes from the Cash Economy, 117 TAX NOTES 506 (Oct. 29, 2007) (describing the proclivity of small-business enterprises to utilize cash receipts as a means to circumvent their tax obligations); Joseph Bankman, Tax Enforcement: Tax Shelters, the Cash Economy, and Compliance Costs, 31 OHIO N.U. L. REV. 1 (2005) (pointing out that even increasing the number of IRS audits may have a lackluster effect in curtailing tax noncompliance). 74 United States v. Procario, 356 F.2d 614 (2d Cir. 1966); Goldfield v. Commissioner, 26 T.C.M. (CCH) 575 (1967); Gemma v. Commissioner, 46 T.C. 821 (1966). 75 Gleckman v. United States, 80 F.2d 394, 399 (8th Cir. 1935). 76 See IRM 9.5.9.7 (Nov. 5, 2004). 77 Helvering v. Gregory, 69 F.2d 809, 810 (2d Cir. 1934), aff’d sub nom. Gregory v. Helvering, 293 U.S. 465 (1935).

13

Of course, taxpayers are always at liberty to reduce their tax burdens in legitimate ways. For example, in order to secure preferential capital gains treatment, a taxpayer can strategically hold stock for more than one year.78 Likewise, a taxpayer can decide to purchase an electric vehicle rather than one that relies upon fossil fuel to avail himself of a generous .79 Indeed, there are countless ways by which taxpayers may capitalize upon the legal tax-avoidance strategies that Congress dangles in front of them.80 However, some taxpayers have sought to minimize their tax burdens in ways that have gone well beyond congressional intent. Common tax shelter techniques that emerged during the 1970s and early 1980s involved the use of accelerated depreciation. Here, a taxpayer would purchase title to a piece of property whose true worth was, say, $10,000, from a tax shelter promoter at a greatly inflated price, say $1 million, using $10,000 of his own money and borrowing the balance from the promoter on a nonrecourse basis.81 Over the next several years, the taxpayer would secure generous depreciation or amortization write-offs to shelter otherwise taxable income.82 Admittedly, if the taxpayer walked away from the property, there would be a significant recognition event.83 However, if the property was held until death, the entire potential gain would simply disappear.84 When Congress introduced legislative measures to close down these tax shelters (i.e., Code sections 465 and 469), other modes of sheltering income emerged. For example, during the 1990s, a technique that gained widespread use was known as COBRA (Currency Options Bring Reward Alternatives). Although technical in nature,85 the basic elements of a COBRA transaction (and others like it) played off a strict reading of the Code. By reading the Code in such a constricted manner, taxpayers were theoretically able to manufacture whatever amount of loss they wished without any economic risk whatsoever.86 The Service did not ignore the emergence of these strategies. Instead, the agency went on the attack, prodding Congress to take legislative measures and bringing numerous tax audits.87 However, in trying to counter the flourishing tax shelter market, it encountered two problems. First, the vast majority of tax

78 I.R.C. § 1222(4). 79 Id. § 30D. 80 See, e.g., id. § 1400Z-2 (capital gains that are reinvested within 180 days into “opportunity zone property” can be deferred and, in some cases, eliminated). 81 Sheid v. Commissioner, 50 T.C.M. (CCH) 663, 672 (1985) (“We conclude here that the nonrecourse $920,000 promissory note was too contingent as of December 31, 1972 and December 31, 1973 to be included in basis.”). 82 See generally Daniel Simmons, Nonrecourse Debt and Basis: Mrs. Crane Where Are You Now?, 53 S. CAL. L. REV. 1 (1979). 83 I.R.C. § 1001(b); Commissioner v. Tufts, 461 U.S. 300 (1983). 84 I.R.C. § 1014(a). 85 See Karen C. Burke & Grayson M.P. McCouch, COBRA Strikes Back: Anatomy of a Tax Shelter, 62 . 59, 64– 65 (2008) (explaining the details of a typical COBRA transaction). 86 See Marvin A. Chirelstein & Lawrence A. Zelenak, Tax Shelters and the Search for a Silver Bullet, 105 COLUM. L. REV. 1939, 1939 (2005) (“The marketing of tax shelters by leading accounting and investment banking firms has developed into a perfect plague over the past decade. The aim in every case is to create a tax benefit in the form of a loss, expense, or exclusion from gross income that has no economic corollary but is simply the consequence, or the hoped-for consequence, of rule manipulation.”). See generally Steven A. Dean & Lawrence M. Solan, Tax Shelters and the Code: Navigating Between Text and Intent, 26 VA. TAX REV. 879 (2007); James S. Eustice, Abusive Corporate Tax Shelters: Old “Brine” in New Bottles, 55 TAX L. REV. 135 (2002); David A. Weisbach, Ten Truths About Tax Shelters, 55 TAX L. REV. 215 (2002); Michael L. Schler, Ten More Truths About Tax Shelters: The Problem, Possible Solutions and a Reply to Professor Weisbach, 55 TAX L. REV. 325 (2002). 87 See, e.g., I.R.S. News Release IR-2004-128 (Oct. 20, 2004) (“Last month, the IRS Appeals Division and the Large and Mid-Size Business Division began sending letters to taxpayers involved in three abusive transactions, advising them that the settlement terms available to resolve these transactions have changed. Certain taxpayers who reported losses and deductions from lease strips, inflated-basis assets derived from lease strips, and in intermediary transactions are being notified that the Appeals position has tightened.”).

14

shelters were “wrapped” as legitimate business transactions entered into for profit,88 making their discovery by Service auditors painstakingly difficult. Second, even if the Service uncovered a particular tax shelter, attacking its legitimacy was met with fierce resistance because these shelters were often formulated by some of the country’s biggest and most prestigious accounting, law, and banking firms.89

3. The Establishment of Hidden Offshore Accounts

For decades, due to opaque Swiss bank secrecy laws, taxpayers found it attractive to park their assets in Swiss and other offshore accounts.90 From these accounts, taxpayers could make investments and secure earnings. On paper, there were cooperation agreements between the United States and foreign countries to combat tax evasion. However, these agreements proved largely ineffective.91 The use of offshore accounts thus flourished. Taxpayers would stash untold amounts of assets beyond the reach of the Service.92 As a result, from these earnings simply vanished. Strikingly, the Swiss government derived only de minimis revenue from foreign taxpayers opening bank accounts on Swiss soil.93 Presumably, the Swiss government, as well as other tax havens, sought to buoy the income of its banking industry and, by doing so, augment its tax revenues. For the tax havens’ governments, these trivial revenue amounts apparently constituted sufficient motivation to turn a blind eye to the financial havoc their actions were causing tax authorities in the United States and throughout the world.94 On rare occasions, some taxpayers would be caught. In United States v. Williams,95 for example, the taxpayer apparently deposited more than $7 million in two Swiss accounts, opening the accounts in the name of a British company. During the time period involved, these accounts earned more than $800,000 of

88 See, e.g., Black & Decker Corp. v. United States, 340 F. Supp. 2d 621 (D. Md. 2004) (describing a taxpayer who formed a subsidiary and then, with a putative high tax basis in its stock, sold it to generate a significant loss). 89 Due to the resources spent in creating these tax shelters, the taxpayer sometimes prevailed in court adjudications with the IRS. United States v. Home Concrete & Supply, LLC, 132 S. Ct. 1836 (2012); IES Indus., Inc. v. United States, 253 F.3d 350 (8th Cir. 2001); Compaq Computer Corp. v. Commissioner, 277 F.3d 778 (5th Cir. 2001); TIFD III-E, Inc. v. United States, 342 F. Supp. 2d 94 (D. Conn. 2004); Black & Decker, 340 F. Supp. 2d; Coltec Indus., Inc. v. United States, 62 Fed. Cl. 716 (2004). 90 See Carolyn Michelle Najera, Note, Combating Offshore Tax Evasion: Why the United States Should Be Able to Prevent American Tax Evaders from Using Swiss Bank Accounts to Hide Their Assets, 17 SW. J. INT’L L. 205 (2011); Jared Seff, Cracking Down on Tax Evaders — Swiss Banking: Secrets, Lies, and Deception, 38 S.U. L. REV. 159 (2010). 91 See Jane G. Song, The End of Secret Swiss Accounts?: The Impact of the U.S. Foreign Account Tax Compliance Act (FATCA) on Switzerland’s Status as a Haven for Offshore Accounts, 35 NW. J. INT’L L. & BUS. 687, 692–95 (2015) (describing U.S.-Swiss policies on secret banking and tax avoidance prior to FATCA). 92 See Staff of Permanent Subcomm. on Investigations of the S. Comm. on Homeland Sec. & Governmental Affairs, 110th Cong., Banks and U.S. Tax Compliance 1 (Comm. Print 2008), http://www.hsgac.senate.gov//imo/media/doc/071708PSIReport.pdf? attempt=2 (estimating that the U.S. loses approximately $100 billion of revenue annually from individuals utilizing unreported offshore accounts); Cynthia Blum, Sharing Bank Deposit Information with Other Countries: Should Tax Compliance or Privacy Claims Prevail?, 6 Fla. Tax Rev. 579, 591 (2004) (quoting Crime and Secrecy: The Use of Offshore Banks and Companies: Hearings Before the Permanent Subcomm. on Investigations of the S. Comm. on U.S. Gov’t Affairs, 98th Cong. (1985)) (noting that taxpayers’ offshore activities have “‘thwart[ed] the collection of massive amounts of tax revenues”’). 93 See Lisa Goetz, Why Is Switzerland Considered a Tax Haven?, INVESTOPEDIA (last updated Aug. 1, 2019), https://www.investopedia.com/ask/answers/060716/why-switzerland-considered-tax-haven.asp (“Switzerland does not allow foreign individuals to live and bank in its borders tax-free. However, wealthy individuals can pay a low, lump-sum option on the money they bank inside the country, and the government considers their taxes paid. To simplify matters, the government bases the amount of tax foreigners owe on seven times their monthly rent.”). 94 For a discussion of the tax-havens business model, see Omri Marian, Blockchain Havens and the Need for Their Internationally Coordinated Regulation, 20 N.C. J.L. & TECH. 529 (2019). 95 106 A.F.T.R.2d (RIA) 6150 (E.D. Va. 2010).

15

income. Notwithstanding the existence of these accounts and their earnings, the taxpayer failed to disclose their existence on his U.S. tax return. Years later, when the Swiss authorities froze these accounts, the taxpayer subsequently disclosed their existence. As a result of reporting failures, the taxpayer pled guilty to tax fraud and criminal tax evasion. However, insofar as litigating cases is concerned, the Williams decision represented the exception, not the norm. Furthermore, it appears that while the Swiss banking industry initiated the process of seeking to attract clandestine capital to its shores, other nations eventually followed suit; and banks in other countries, such as Liechtenstein and those in the Caribbean, vied for a piece of the action.96 They patterned their laws after those of the Swiss in order to attract flows of capital. With little to deter these actions, the tax-fraud caravan appeared never-ending, in a fiscal race to the bottom. ______

As demonstrated by the size of the tax gap, these three tax-noncompliance strategies — use of cash, aggressive tax shelters, and hidden offshore accounts — and others have come at a steep price tag to U.S. tax revenues. As a practical reality, every year taxpayers fail to report billions of dollars of taxes.97 Admittedly, a part of the tax gap’s size is due to innocent mistakes and the financial inability of some taxpayers to make payments; however, a far larger component of the tax gap is attributable to the fact that many taxpayers purposefully shortchange the government in terms of fulfilling their tax burdens.98

III. TRENDS: TECHNOLOGICAL ADVANCES THAT ENHANCE TAX COMPLIANCE

Due to changing tax policies, economic forces, institutions, norms, and technology, the methods by which governments enforce the tax laws and by which individuals and firms evade their taxes shift over time. This section of the article highlights recent technological advances that affect the Service’s ability to enforce the nation’s tax laws and strengthen tax compliance. Recent years have seen stunning changes in technology, ushering in a new era of how humans interact and conduct their affairs. These changes touch upon virtually every aspect of human existence, ranging from how homes are built,99 how crops are harvested,100 how companionship is provided for the elderly,101 and even how romance is discovered.102 When it comes to crime prevention, technological advances have been employed in virtually every sphere of human existence. Facial-recognition devices can now scan whole stadiums to identify criminals

96 See Oliver Bullough, Nevis: How the World’s Most Secretive Offshore Haven Refuses to Clean Up, GUARDIAN (July 12, 2018), https://www.theguardian.com/news/2018/jul/12/ nevis-how-the-worlds-most-secretive-offshore-haven-refuses-to-clean-up (“Places such as Jersey, Switzerland and the British Virgin Islands made a handsome living from helping their clients break other countries’ laws for decades, without anyone really noticing.”). See generally J.C. SHARMAN, HAVENS IN A STORM: THE STRUGGLE FOR GLOBAL TAX REGULATION (2006). 97 See supra note 11 and accompanying text. 98 For the most recent tax gap study, see INTERNAL REVENUE SERV., PUB. 1415, FEDERAL TAX COMPLIANCE RESEARCH: TAX GAP ESTIMATES FOR TAX YEARS 2008–2010 (2016), https://www.irs.gov/pub/irs-soi/p1415.pdf. For another comprehensive overview, see Mark J. Mazur & Alan H. Plumley, Understanding the Tax Gap, 60 NAT’L TAX J. 569 (2007). 99 See, e.g., Peter Debney, Artificial Intelligence Can Save Time and Money, ARCHITECT (Nov. 7, 2018), https://www.architectmagazine.com/technology/artificial-intelligence-can-save-time-and-money_s. 100 See, e.g., Miriam Jordan, As Immigrant Farmworkers Become More Scarce, Robots Replace Humans, N.Y. TIMES (Nov. 20, 2018), https://www.nytimes.com/2018/11/20/us/ farmworkers-immigrant-labor-robots.html. 101 See, e.g., Imani Moise, For the Elderly Who Are Lonely, Robots Offer Companionship, WALL ST. J. (May 28, 2018), https://www.wsj.com/articles/for-the-elderly-who-are-lonely-robots-offer-companionship-1527559260. 102 See, e.g., Breeanna Hare, How Technology Has Changed Romance, CNN BUS. (Feb. 12, 2013), https://www.cnn.com/2013/02/12/tech/web/tech-romance-evolution/index.html.

16

lurking in a crowd;103 helicopters can use heat-seeking sensors to determine where felons may be seeking refuge;104 security cameras dot virtually every corner in many cities, monitoring and recording the acts of those who may break the law;105 DNA testing can now help track down rapists using saliva submitted for analysis by relatives;106 and powerful X-ray machines are being used at airports to detect sophisticated explosive devices.107 Society is launching these and other tools to make crimes far harder to perpetrate and, even when they are perpetrated, to make it much easier to detect those responsible. One particular set of technological changes that could benefit crime-prevention efforts but might also hold potentially offsetting effects that make it easier for individuals to commit crimes relates in some way to “digitalization.” Digitalization “is of crucial importance to data processing, storage and transmission, because it allows information of all kinds in all formats to be carried with the same efficiency and also intermingled,”108 thereby opening the door to information storage, transmission, and analysis. In what Robert J. Gordon refers to as the “Third Industrial Revolution,”109 digitalization has led to numerous innovations, such as the introduction of electronic cash,110 electronic commerce,111 blockchain technology,112 supply chains,113 peer-to-peer (P2P) networks,114 mobile phone apps,115 biometrics,116 big

103 See, e.g., Amy B. Wang, A Suspect Tried to Blend in with 60,000 Concertgoers. China’s Facial-Recognition Cameras Caught Him, WASH. POST (Apr. 13, 2018), https://www.washingtonpost.com/news/worldviews/wp/2018/04/13/china-crime-facial- recognition-cameras-catch-suspect-at-concert-with-60000-people/?utm_term=. cc022c2c2663. 104 See, e.g., Kelsey D. Atherton, How It Works: The Thermal Camera That Found the Boston Bomber, POPULAR SCI. (Apr. 25, 2013), https://www.popsci.com/technology/article/ 2013-04/how-works-awesome-thermal-camera-found-boston-bomber. 105 See, e.g., Andrea Noble, D.C. Surveillance Cameras Become Top Crime-Fighting Tools for Police, WASH. TIMES (June 30, 2013), https://www.washingtontimes.com/news /2013/jun/30/dc-surveillance-cameras-become-top-crime-fighting-/. 106 See, e.g., Eliott C. McLaughlin, DNA in NorCal Rapist Case Links Suspect to Sexual Assaults in 6 Counties, CNN (Sept. 23, 2018), https://www.cnn.com/2018/09/23/us/norcal-rapist-arrest-arraignment-dna-genetic-genealogy/index.html. 107 See, e.g., Nicole Hattenschwiler, Yanik Sterchi, Marcia Mendes & Adrian Schwaninger, Automation in Airport Security X-Ray Screening of Cabin Baggage: Examining Benefits and Possible Implementations of Automated Explosives Detection, 72 SCIENCEDIRECT 58 (2018), https://www.sciencedirect.com/science/article/pii/S00036870183 0108X. 108 Digitization, WIKIPEDIA, https://en.wikipedia.org/wiki/Digitization#cite_ref-6 (last visited July 19, 2019). 109 Robert J. Gordon, The Rise and Fall of American Growth — The U.S. Standard of Living Since the Civil War 320, passim (2016). 110 See, e.g., Anton N. Didenko & Ross P. Buckley, The Evolution of Currency: Cash to Cryptos to Sovereign Digital Currencies, 42 FORDHAM INT’L L.J. 1041, 1041 (2019) (“In 2009, Bitcoin created a world-first decentralized alternative currency that has spawned over 1,700 imitations by private parties. In 2018, governments finally joined the race, as Venezuela issued a world-first sovereign digital currency. Major economies like Canada, China, Singapore and the United Kingdom are all developing their own versions.”). 111 See, e.g., Rory Van Loo, Digital Market Perfection, 117 MICH. L. REV. 815, 817 (2019) (“The world’s largest companies, including Apple, Google, and Microsoft, are racing to develop artificially intelligent (AI) butlers. They aim to allow consumers to outsource the tasks of opening a new bank account, locating cheaper laundry detergent, and finding the highest quality groceries.”). 112 See, e.g., Trevor I. Kiviat, Note, Beyond Bitcoin: Issues in Regulating Blockchain Transactions, 65 DUKE L.J. 569, 569 (2015) (“Simply, blockchain technology solves an elusive networking problem by enabling ‘trustless’ transactions: value exchanges over computer networks that can be verified, monitored, and enforced without central institutions (for example, banks).”). 113 See, e.g., Galit A. Sarfaty, Shining Light on Global Supply Chains, 56 HARV. INT’L L.J. 419, 425 (2015) (“Global outsourcing — the practice of sub-contracting business to third parties in other countries [a.k.a., supply chain] — has become a trillion-dollar industry, with an annual growth rate between 12% and 26% across functions.”). 114 See, e.g., Peter K. Yu, P2P and the Future of Private Copying, 76 COLO. L. REV. 653, 676 (2005) (“[A]ll of the new P2P technologies, such as Grokster, iMesh, KaZaA, and Morpheus, do not have centralized servers. Instead, they allow users to transfer files from one location to another while accommodating users’ needs to employ different hardware and software.”). 115 See, e.g., Chris Jay Hoofnagle & Jennifer M. Urban, Alan Westin’s Privacy Homo Economicus, 49 WAKE FOREST L. REV. 261, 288 (2014) (“Mobile phone apps can collect user information both directly and indirectly. Examples of direct collection include tracking posts to social networking sites, harvesting data input by users or their reading, viewing, and listening practices, and collecting information stored in other phone applications.”). 116 See, e.g., Ted Claypoole & Cameron Stoll, Developing Laws Address Flourishing Commercial Use of Biometric Information, BUS. L. TODAY, May 2016, at 1 (“Biometric identification systems are rapidly increasing in use, as advances in sensors,

17

data,117 robotics,118 and deep learning.119 In short, digitalization offers the potential for the government to generate additional information that is more timely, more precise, and easier to analyze. However, in the wrong hands, digitalization also offers the potential for abuse. The unanswered question is how these changes in technology will affect tax compliance. In particular, can the Service capitalize upon these many technological changes to curtail tax noncompliance? Broadly, technology creates vastly increased opportunities for the Service to gather and analyze information on any transactions that leave some form of electronic trail, potentially reducing the scope of taxpayer- noncompliance actions. While driven primarily by their potential for improved business efficiencies, innovations such as P2P, blockchain, and supply chains create paths for the Service to track transactions between parties. Further, these technological abilities also create opportunities for prepopulated tax returns, electronic filing, and expanded third-party information returns, all of which may reduce the opportunity for tax noncompliance. In fact, several of these technological changes have already influenced tax administration. Congress has purposefully sought to use technological advances to promote tax compliance120 and to make the tax administration process more efficient and less prone to taxpayers skirting their civic responsibilities.121 Further, it is well-known that the Service utilizes computer-generated Discriminant Inventory Function (DIF) scores122 in order to engage in a more efficient audit process.123 Moreover, in the 1980s, the Service created the Automated Underreporter (AUR) program, which computerized the process of third-party document matching (e.g., matching between employee and employer tax returns).124 The Service also has an in-house data-mining division — known as Research, Applied Analytics, and Statistics (RAAS) — that

readers, and software make physical characteristics easily measurable. Biometrics are simply measurements of a person’s physical being. Fingerprints, retinal or iris scans, hand geometry, facial recognition, gait analysis, voiceprint reading, and even keystroke analysis are all simple biometric ways to identify a person.”). 117 See, e.g., Neil M. Richards & Jonathan H. King, Three Paradoxes of Big Data, 66 STAN. L. REV. ONLINE 41, 42 (2013) (“Big data analytics depend on small data inputs, including information about people, places, and things collected by sensors, cell phones, click patterns, and the like. These small data inputs are aggregated to produce large datasets which analytic techniques mine for insight. This data collection happens invisibly and it is only accelerating. Moving past the Internet of Things to the ‘Internet of Everything,’ Cisco projects that thirty-seven billion intelligent devices will connect to the Internet by 2020.”). 118 See, e.g., M. Ryan Calo, Open Robotics, 70 MD. L. REV. 571, 571 (2011) (“Robotics is poised to be the next transformative technology. Robots are widely used in manufacturing, warfare, and disaster response, and the market for personal robotics is exploding.”). 119 See, e.g., Wendell Wallach, Rise of the Automatons, 5 SAVANNAH L. REV. 1, 2–3 (2018) (“What deep-learning algorithms do is they can look at a massive amount of data (in fact it’s required that they look at a massive amount of data) about a particular subject, and they will find significant relationships within that data. Often those are significant relationships that humans would not discover or recognize without the help of great computing power.”). 120 See, e.g., Emergency Economic Stabilization Act of 2008, Pub. L. No. 110-343, § 403, 122 Stat. 3765, 3854–56 (legislation requires that all third-party brokers track the tax basis that their customers have in their marketable securities). 121 In light of IRS budget cutbacks, however, there has never been a bigger need to do more with fewer resources. See Kimberly A. Houser & Debra Sanders, The Use of Big Data Analytics by the IRS: What Tax Practitioners Need to Know, 128 J. TAX’N 2 (2018). 122 See Robert E. Brown & Mark J. Mazur, The National Research Program: Measuring Taxpayer Compliance Comprehensively, 51 U. KAN. L. REV. 1255, 1260–61 (2003) (providing a general description of the National Research Program and its predecessor, the Taxpayer Compliance Measurement Program (TCMP)). 123 See INTERNAL REVENUE SERV., PUB. 556, EXAMINATION OF RETURNS, APPEAL RIGHTS, AND CLAIMS FOR REFUND 2 (2013); see also How Tax Returns Are Selected for Audit: Explaining DIF Scores and UI DIF Scores, BROTMAN L., http://info.sambrotman.com /blog/how-tax-returns-are-selected-for-audit/ [https://perma.cc/SU89-2RSC] (last visited July 19, 2019). 124 See Stanley Foodman, Understanding How IRS Identifies and Selects Individual Tax Returns for Automated Underreporter Review, JD SUPRA (Dec. 27, 2018), https://www .jdsupra.com/legalnews/understanding-how-irs-identifies-and- 11599/; Dubin, Graetz & Wilde, supra note 37, at 397.

18

develops data-driven compliance initiatives.125 The RAAS division works closely with outside contractors and the Service’s Criminal Investigation (CI) Division to develop data-mining solutions that identify complex patterns of noncompliance such as identity theft, tax fraud, and money laundering.126 Below, Subpart A explores new tools at the Service’s disposal to spur tax compliance, and Subpart B investigates the utility of such technological tools to defeat twentieth-century modes of tax noncompliance.

A. Tools Available to the Internal Revenue Service to Spur Tax Compliance

Consider three new tools that likely will foster more robust tax compliance, namely, (1) data mining, (2) artificial intelligence, and (3) public-key cryptography.

1. Data Mining

Data mining involves extracting useful information from “big data,” or data sets that are too large for traditional data-processing methods.127 While data mining has many uses (e.g., market analysis and customer retention), one particularly applicable to tax administration is its ability to assist with fraud detection.128 How does data mining accomplish its goals? Its users first propose a function to describe how particular data may relate. For example, a supermarket may want to know what percentage of shoppers who purchase peanut butter also buy jelly; in a somewhat similar vein, the Service may want to learn if there is a correlation between partnerships that make a Code section 754 election and the partners of such electing partnerships subsequently contributing their partnership interests into newly formed partnerships (thereby forming tiered partnership arrangements). Based upon this descriptive function, data miners then classify this relationship and seek to predict an unknown.129 In the case of partnerships, the unknown may be whether the election-contribution dynamic is a telltale sign that participating partners are involved in an illicit tax shelter arrangement designed to circumvent their tax obligations. Once a hypothesis is made, data must be gathered, arranged, configured, and analyzed. More specifically, the data must be coded into a

125 See Siri Bulusu, ‘Doubling Down’ on Big Data Upgrades IRS Fraud Investigations, BLOOMBERG BNA (Dec. 14, 2018), https://www.bloomberglaw.com/ms/product/tax/ document/XCCD07E0000000; see also TREASURY INSPECTOR GEN. FOR TAX ADMIN., REFERENCE NO. 2018-10-026, THE RESEARCH, APPLIED ANALYTICS AND STATISTICS ORGANIZATION PROJECT MANAGEMENT PRACTICES NEED IMPROVEMENT 1 (May 4, 2018), https://www.treasury.gov/tigta/auditreports/2018reports/201810026fr.pdf (“The RAAS organization encompasses two previous IRS offices — the Office of Research, Analysis, and Statistics (RAS) and the Office of Compliance Analytics (OCA) — that were merged in October 2016. The IRS Commissioner merged the two offices to integrate their organizational structures in a manner that supports and enhances the interaction between the two organizations and assists IRS leadership’s emphasis on analytics, innovation, and data-driven decision making.”). 126 See INTERNAL REVENUE SERV., PIA ID NO. 3179, PIA REPORT 2 (Feb. 14, 2018), https://www.irs.gov/pub/irs-utl/lca- pia.pdf. 127 See Houser & Sanders, supra note 121, at 7 (“Data mining is the analysis of large data sets in order to search for previously unidentified relationships in the data. It can be predictive or descriptive. While descriptive summarizes attributes of the data, predictive is performing analysis to forecast attributes that are not visible.”). 128 See, e.g., Neil Issar, More Data Mining for Medical Misrepresentation? Admissibility of Statistical Proof Derived from Predictive Methods of Detecting Medical Reimbursement Fraud, 42 N. KY. L. REV. 341 (2015); Philip K. Chan, Wei Fan, Andreas L. Prodromidis, & Salvatore J. Stolfo, Distributed Data Mining in Credit Card Fraud Detection, 14 IEEE INTELLIGENT SYS. 67 (1999). 129 Data mining is becoming increasingly prevalent in the area of policing. See, e.g., Ellen Huet, Server and Protect: Predictive Policing Firm PredPol Promises to Map Crime Before It Happens, FORBES (Mar. 2, 2015), https://www.forbes.com/sites/ellenhuet/ 2015/02/11/predpol-predictive-policing/ (“In a 2012 survey of almost 200 police agencies 70 [percent] said they planned to implement or increase use of predictive policing technology in the next two to five years.”).

19

format that captures relevancy and discards what is irrelevant.130 The data must then be warehoused.131 Once warehoused, the data-mining process can proceed. This involves a series of important steps, including, but not limited to, the following: data cleaning, data integration, data selection, data transformation, data mining, pattern evaluation, and knowledge presentation.132 Once these steps are fulfilled, completion of the process is supposed to yield “knowledge” upon which its users can act.133 In the tax sphere, due to both tax return submissions and third-party tax information reporting,134 the availability of data abounds. The Service thus has many opportunities to use data-mining techniques to produce knowledge that reveals areas of taxpayer compliance and noncompliance.135 In the past, the Service always had access to data, but this information was often crude in nature and cumbersome to organize. Now and in the future, there can be systematic and seamless use of this data, opening up vast opportunities to connect noncompliance dots.136 As computers continue to grow in memory storage, processing ability, and speed, it is easy to anticipate that data mining will increasingly become one of the Service’s foremost tools to monitor taxpayer compliance.137

2. Artificial Intelligence

Artificial intelligence (AI) is the process by which machines attempt to duplicate or surpass the cognitive abilities of living creatures.138 While AI has many dimensions, one of the primary ways it functions is through the use of algorithms, or a set of “unambiguous instructions that a mechanical computer

130 See generally Mehmed Kantardzic, Data Mining: Concepts, Models, Methods, and Algorithms (2d ed. 2011). 131 See generally Paulraj Ponniah, Data Warehousing Fundamentals: A Comprehensive Guide for IT Professionals (2001). 132 An overview of this process is set forth in the following piece: Usama Fayyad, Gregory Piatetsky-Shapiro & Padhraic Smyth, From Data Mining to Knowledge Discovery in Databases, 17 AI MAG. 37, 40–42 (1996). 133 See id. at 37 (“There is an urgent need for a new generation of computational theories and tools to assist humans in extracting useful information (knowledge) from the rapidly growing volumes of digital data. These theories and tools are the subject of the emerging field of knowledge discovery in databases.”). 134 See generally Jay A. Soled, Homage to Information Returns, 27 VA. TAX REV. 371 (2007). 135 Beyond tax return information, the IRS can mine other data, including taxpayers’ social media. Dara Kerr, Tax Dodgers Beware: IRS Could Be Watching Your Social Media, CNET (Apr. 15, 2014), http://www.cnet.com/news/tax-dodgers-beware-irs- could-be-watching-your-social-media/ [https://perma.cc/2UFZ-GJTB]; see also Tim Sampson, FYI, the IRS Is Looking at Your Online Activity for Signs of Tax Evasion, DAILY DOT (Apr. 16, 2014), http://www.dailydot.com/news/irs-social-media-tax- evasion/[https://perma.cc/F33W-M9FL]; Report: IRS Data Mining Facebook, Twitter, Instagram and Other Social Media Sites, CBS DC (Apr. 16, 2014), http://washington.cbslocal.com/2014/04/16/report-irs-data-mining-facebook-twitter-instagram-and- other-social-media-sites/[https://perma.cc/8G4W-GEZ3]; Jaikumar Vijayan, IRS, DOJ Use Social Media Sites to Track Deadbeats, Criminal Activity, COMPUTER WORLD (May 16, 2010), http://www.computerworld.com/article/2516372/web-apps/irs--doj-use- social-media-sites-to-track-deadbeats--criminal-activity.html [https://perma.cc/L9GA-JSKY]. 136 See Richard Satran, IRS High-Tech Tools Track Your Digital Footprints, U.S. NEWS (Apr. 4, 2013), http://money.usnews.com/money/personalfinance/mutualfunds/articles/ 2013/04/04/irs-high-tech-tools-track-your-digital- footprints [https://perma.cc/WUC5-86TJ] (“The IRS last year used a profiling test model to study 1,500 tax preparers with histories of reporting deficiencies and managed to recover $200 million. It cited the experience as proof that its data analysis works.”). 137 See, e.g., Laura Saunders, Inside Swiss Banks’ Tax-Cheating Machinery, WALL ST. J. (Oct. 22, 2015), https://www.wsj.com/articles/inside-swiss-banks-tax-cheating-machinery-1445506381?ns=prod/accounts-wsj (“Meanwhile, U.S. officials said they are mining the extensive data uncovered by the program, including the destination of funds transferred out of Swiss accounts, to pursue leads around the world.”). 138 See Artificial Intelligence, MERRIAM-WEBSTER, https://www.merriam-webster.com /dictionary/artificial%20intelligence (last visited July 21, 2019). As an example of how the world of artificial intelligence is developing, see Gideon Lewis-Kraus, The Great A.I. Awakening, N.Y. TIMES MAG. (Dec. 16, 2016), https://www.nytimes.com/2016/12/14/ magazine/the-great-ai- awakening.html (“A rarefied department within the company, Google Brain, was founded five years ago on this very principle: that artificial ‘neural networks’ that acquaint themselves with the world via trial and error, as toddlers do, might in turn develop something like human flexibility.”).

20

can execute.”139 What differentiates the AI of today from that of yesteryear is that technicians have developed algorithms that can self-enhance, becoming “smarter” — and also far more efficient and more effective — as they receive more data.140 Endowed with human intellectual characteristics, what AI could achieve is theoretically limitless. Able to work 24/7, AI could potentially solve pressing global problems such as climate change, population growth, and diseases like cancer and HIV. To date, AI has not been successful in achieving such lofty goals. However, AI has attained some other salient milestones: due to AI, cars can drive autonomously;141 computers can challenge and beat the world’s best chess, go, and shogi players;142 and battlefield scenarios can be simulated that leave military planners in awe.143 Enter AI into the tax arena. Recent research144 has demonstrated that, when fed appropriate data and using various supplied algorithms and those that it may develop on its own, AI can anticipate particular modes of tax evasion.145 The researchers’ goal was to have AI identify those tax schemes that taxpayers and their advisers may use to engage in “malicious tax noncompliance,” by focusing “on rule mining, in which individual tax code regulations are lined up against one another to ascertain if they can be used collectively to create a sophisticated tax dodge.”146 As evidenced by this research, as long as the Service has more powerful computers at its disposal relative to those of plotting taxpayers, AI will permit the agency to remain ahead of the tax-planning curve.

3. Public-Key Cryptography

Ever since the internet emerged, its users have voiced privacy concerns, particularly as related to email exchanges and document attachments.147 Such concerns have not been ill-founded. Routinely and increasingly, there have been internet privacy violations as hackers and trolls have scoured the internet and

139 Firas Kobeissy, Kevin Wang, Fadi A. Zaraket & Ali Alawieh, Leveraging Biomedical and Healthcare Data: Semantics, Analytics and Knowledge 61 (2019). 140 See id. (“AI algorithms can learn from data; they can enhance themselves by learning new strategies that have worked well in the past (positive reinforcement). They can also write or modify algorithms.”). 141 See Jennifer Kite-Powell, How Artificial Intelligence Can Create a Real World Simulation for Autonomous Driving, FORBES (July 10, 2018), https://www.forbes.com/ sites/jenniferhicks/2018/07/10/how-artificial-intelligence-can-create-a-real- world-simulation-for-autonomous-driving/#31b42b2d2884. 142 See Marie Boran, DeepMind’s AlphaZero Teaches Itself to Beat Humans at Chess, Go and Shogi, IRISH TIMES (Dec. 14, 2018), https://www.irishtimes.com/business/technology/ deepmind-s-alphazero-teaches-itself-to-beat-humans-at-chess-go-and- shogi-1.3728692. 143 See M. L. CUMMINGS, CHATHAM HOUSE: ROYAL INST. OF INT’L AFFAIRS, ARTIFICIAL INTELLIGENCE AND THE FUTURE OF WARFARE (Jan. 2017), https://www.chathamhouse.org /sites/default/files/publications/research/2017-01-26-artificial-intelligence- future-warfare-cummings-final.pdf; Sydney J. Freedburg Jr., AI and Robots Crush Foes in Army Wargame, BREAKING DEFENSE (Dec. 19, 2019), https://breakingdefense.com/2019/12/ai-robots-crush-foes-in-army-wargame/. 144 See Erik Hemberg, Jacob Rosen, Geoff Warner, Sanith Wijesinghe & Una-May O’Reilly, Detecting Tax Evasion: A Co- Evolutionary Approach, 24 ARTIFICIAL INTELLIGENCE & L. 149 (2016). 145 See Ryan Kh, Can Predictive Analytics Prevent Tax Evasion?, SMARTDATACOLLECTIVE (Aug. 15, 2018), https://www.smartdatacollective.com/can-predict ive-analytics-prevent-tax-evasion/. 146 Lynnley Browning, Computer Scientists Wield Artificial Intelligence to Battle Tax Evasion, N.Y. TIMES (Oct. 9, 2015), https://www.nytimes.com/2015/10/10/business/computer -scientists-wield-artificial-intelligence-to-battle-tax-evasion.html; see Hemberg, supra note 144. 147 See generally A. Michael Froomkin, The Death of Privacy?, 52 STAN. L. REV. 1461 (2000) (discussing how modern technology threatens privacy); Paul M. Schwartz, Privacy and Democracy in Cyberspace, 52 VAND. L. REV. 1609 (1999) (same).

21

reached into users’ servers to secure credit card accounts,148 Social Security numbers,149 and other personal information, including email exchanges between and among some of our nation’s most notable political figures and compromising pictures/videos of major motion picture stars.150 The financial, political, and emotional damage associated with such violations has been enormous.151 As a means to address these concerns, computer scientists have, over the last several decades, focused on developing various modes of cryptography to safeguard personal information. Without getting into many of the technical aspects of such cryptography, these computer scientists have developed what is known as “public-key cryptography,” which makes the contents of exchanged information closed to anyone outside of the exchange.152 In the tax world, the potential implications associated with public-key cryptography are vast. Currently, the number of taxpayers who use fake or falsified Social Security numbers to secure fraudulent tax refunds is enormous, costing the government billions of dollars of lost revenue annually.153 These scams are hard to detect, forcing the tax authorities to spend massive amounts in their attempt to reduce this fraud. Suppose instead that the Service issued a “key” to every taxpayer. Instead of utilizing a Social Security number or employer identification number, taxpayers would use this unique and designated key to submit their tax returns. This process would ensure taxpayer privacy and inhibit scammers from securing undeserved tax refunds.

B. Application of Tools to Defeat Twentieth-Century Tax Noncompliance

Consider now the ways in which digitization and technological advances affect commonplace twentieth-century tax-noncompliance strategies, namely, (1) the failure to report cash payments and receipts, (2) the use of tax shelters to manufacture noneconomic losses, and (3) the establishment of hidden offshore accounts.

148 See, e.g., Robert Harrow, Is Your Credit Card Less Secure Than Ever Before?, FORBES (Apr. 20, 2018), https://www.forbes.com/sites/robertharrow/2018/04/20/is-your-credit-card-less-secure-than-ever-before/#1e3b76a45c8d (“News of credit card data breaches hit the headlines almost every week.”). 149 See, e.g., Max Jaeger, 143M Social Security Numbers Exposed After Equifax Hack, N.Y. POST (Sept. 7, 2017), https://nypost.com/2017/09/07/equifax-breach-compromises-143m-social-security-numbers/ (“Hackers may have the names and Social Security numbers of 143 million Americans after a massive breach of credit reporting agency Equifax, the company said Thursday. That’s more than half of the nation’s adult population, according to Census figures.”). 150 See, e.g., Jason Hanna, What You Can Do If Someone Posts an Explicit Image of You Online, CNN (Mar. 12, 2017), https://www.cnn.com/2017/03/12/us/nonconsensual-or-revenge-porn-recourse-trnd/index.html (discussing incidents where nonconsensual nude photographs were released and what victims of this practice can do). 151 See, e.g., Herb Weisbaum, Hackers Scored More Social Security Numbers Than Stolen Credit Card Numbers in 2017, NBC (Feb. 21, 2018), https://www.nbcnews.com/tech/ security/smarter-criminals-find-new-ways-commit-cyber-fraud-n849691 (reporting on a study showing that in 2017 “[a]n estimated 16.7 million Americans were victimized . . . [and] fraudsters successfully stole $16.8 billion”). 152 See Brit Cruise, Public Key Cryptography: What Is It?, KHAN ACAD., https:// www.khanacademy.org/computing/computer-science/cryptography/modern-crypt/v/diffie-hellman-key-exchange-part-1 (last visited July 21, 2019); What Is Public-Key Cryptography?, GLOBALSIGN, https://www.globalsign.com/en/ssl-information- center/what-is-public-key-cryptography/ [https://perma.cc/LB3L-9C9R] (last visited July 21, 2019). 153 See, e.g., Robert W. Wood, IRS Paid $5.8 Billion in Fraudulent Refunds, Identity Theft Efforts Need Work, FORBES (Feb. 19, 2015), https://www.forbes.com/sites /robertwood/2015/02/19/irs-paid-5-8-billion-in-fraudulent-refunds-identity-theft-efforts- need-work/#350f1deb251f (“On the good side, the IRS estimates it prevented $24.2 billion in fraudulent identity theft refunds in 2013. Still, the IRS actually paid out $5.8 billion in fraudulent refunds that it realized were fraudulent only later.”).

22

1. Failure to Report Cash Payments and Receipts

Technological advancements have significantly curtailed the use of cash as a mode of tax evasion, for several reasons. First, electronic commerce is decimating cash usage.154 Credit card use is the foundation upon which electronic commerce was initially built. Invented in the mid-twentieth century,155 credit cards have ever since jockeyed with cash for economic prominence. Credit cards afford taxpayers a luxury not found with cash: in lieu of having to carry pockets full of paper that is vulnerable to theft, taxpayers can instead carry around a piece of plastic to acquire almost anything they need. As a result, credit card use has exploded and now dominates the use of cash, a position that cash comfortably had held for well over a millennium.156 Advancements in electronic commerce did not end with the advent of the credit card, though. Today, smartphone apps are competing to replace credit cards, and their entry into the marketplace is meeting with astounding success.157 From a tax administration perspective, both credit cards and smartphone apps leave an electronic mark that can be readily traced and used to verify the veracity of taxpayers’ reporting practices.158 Second, technology in the form of third-party tax information returns affords the Service an oversight tool to monitor cash and income reporting. When Congress enacted the modern income tax, taxpayers had little choice but to make physical bank deposits to pay their tax liabilities; when making purchases, taxpayers had to return and make physical bank withdrawals.159 Large cash amounts would come and go,

154 See generally Jeffrey H. Kahn & Gregg D. Polsky, The End of Cash, the Income Tax, and the Next 100 Years, 41 FLA. ST. U. L. REV. 159 (2013). 155 See Mary Bellis, Invention of Credit Cards, THOUGHTCO. (Feb. 23, 2019), https://www.thoughtco.com/who-invented- credit-cards-1991484 (“The inventor of the first bank issued credit card was John Biggins of the Flatbush National Bank of Brooklyn in New York. In 1946, Biggins invented the ‘Charge-It’ program between bank customers and local merchants. The way it worked was that merchants could deposit sales slips into the bank and the bank billed the customer who used the card.”). 156 See Derek Thompson, How the Decline of Cash Makes America a Safer Country, ATLANTIC (Mar. 25, 2014), https://www.theatlantic.com/business/archive/2014/03/how-the-decline-of-cash-makes-america-a-safer-country/359561/ (“But Americans are rapidly abandoning cash thanks to credit cards, debit cards, and mobile payments. Half a century ago, cash was used in 80 percent of U.S. payments. Now that figure is about 50 percent, according to researchers.”); Catherine New, Cash Dying as Credit Card Payments Predicted to Grow in Volume, HUFFINGTON POST (June 7, 2012), http://www.huffingtonpost.com/2012/06/07/ credit-card-payments-growth_n_1575417.html [https://perma.cc/Z7BD-JTQH] (“What was once the most secure way to pay for things — hard cash — is increasingly becoming currency non grata in wallets and checkouts across the country.”). See generally John Heggestuen, Cash Is Fading, and Checks Are Dying as Smartphones and Tablets Change the Way We Pay, BUS. INSIDER (Aug. 4, 2014), http://www.businessinsider.com/cash-is-fading-and-checks-are- dying-as-smartphones-and-tablets-change-the-way-we-pay-2014-8 [https://perma.cc/3B3U-P2FW]. 157 See A Cash Call, ECONOMIST (Feb. 15, 2007), http://www.economist.com /node/8697424 [https://perma.cc/63HC- CDVM] (“Mobile phones are becoming an increasingly popular way to make all sorts of payments.”). In some countries, such as Kenya, electronic commerce is essentially replacing cash. See, e.g., Tom Standage, Why Does Kenya Lead the World in Mobile Money?, ECONOMIST (Mar. 2, 2015), http://www.economist .com/blogs/economist-explains/2013/05/economist-explains- 18[https://perma.cc/A294-4GWH] (“Launched in 2007 by Safaricom, the country’s largest mobile-network operator, it is now used by over 17 [million] Kenyans, equivalent to more than two-thirds of the adult population; around 25 [percent] of the country’s gross national product flows through it. M-PESA lets people transfer cash using their phones, and is by far the most successful scheme of its type on earth.”). For an interactive timeline of M-PESA, see Celebrating 10 Years of Changing Lives, SAFARICOM (Mar. 2007), https://www.safaricom.co.ke/mpesa_timeline/ timeline.html[https://perma.cc/B99M-WWFL]. 158 See, e.g., Richard Satran, IRS High-Tech Tools Track Your Digital Footprints, U.S. NEWS (Apr. 4, 2013), https://money.usnews.com/money/personal-finance/mutual-funds/ articles/2013/04/04/irs-high-tech-tools-track-your-digital- footprints (“The Internal Revenue Service is collecting a lot more than taxes this year — it’s also acquiring a huge volume of personal information on taxpayers’ digital activities, from eBay auctions to Facebook posts and, for the first time ever, credit card and e-payment transaction records, as it expands its search for tax cheats to places it’s never gone before.”). 159 Evidence for this proposition is found in the decline in the number of bank tellers. Thomas Heath, Bank Tellers Are the Next Blacksmiths, WASH. POST (Feb. 8, 2017), https://www.washingtonpost.com/business/economy/bank-tellers-are-the-next- blacksmiths/2017/02/08/fdf78618-ee1c-11e6-9662-6eedf1627882_story.html?noredirect=on&utm_term=.b65d11205acd.

23

making those related to tax noncompliance hard to detect. Now, due to technology, most individual taxpayers use direct deposit or receive checks from their employers.160 By the same token, when taxpayers make significant purchases, credit cards, smartphone apps, bank checks, and wire transfers are commonplace.161 Even in those rare instances when taxpayers make significant cash deposits or use cash to make significant purchases, financial institutions and retail establishments have a third-party reporting responsibility: anytime a cash deposit or a purchase is made using $10,000 or more of cash, a tax information return known as Form 8300 must be issued to the Service.162 This third-party reporting requirement makes taxpayer noncompliance associated with cash usage far easier to detect and far riskier than it had been in the past.163 Third, during the course of a Service audit, information availability affords the agency a far greater ability to gather a complete understanding of a taxpayer’s finances. In the past, taxpayers could easily hide their expenditures. However, now that taxpayers routinely use credit cards and the internet to make their purchases, little is hidden from Service auditors. When there is a meaningful discrepancy between reported receipts and expenditures, taxpayers must offer plausible explanations for the difference between these two dollar amounts.164 The days in which taxpayers could readily hide income have long since disappeared, particularly because their expenditures are generally so readily translucent.

2. Use of Tax Shelters to Manufacture Noneconomic Losses

During the 1970s, 1980s, and 1990s, tax shelters that manufactured noneconomic losses were commonplace.165 These shelters were often orchestrated by the biggest and most prominent law firms, accounting firms, and banks.166 Their popularity was a testament to the huge tax savings that these devices commanded.167 However, through a series of legislative reforms, Congress sought to curtail tax shelter usage.168 For example, Congress passed legislative measures that included disallowing passive losses,169 instituting tax shelter registration requirements,170 and codifying the economic substance doctrine.171 This series of legislative measures undoubtedly helped curtail tax shelter use.

160 See New NACHA Survey Shows Adoption and Awareness of Direct Deposit via ACH Continues to Build, NACHA (Apr. 18, 2016), https://www.nacha.org/news/new-nacha-survey-shows-adoption-and-awareness-direct-deposit-ach-continues-build (“A full 82 percent of U.S. workers — crossing age, income and other demographic categories — are paid by Direct Deposit via ACH, up from 74 percent in 2011.”). 161 Even at the local coffee store, consumers are forgoing the use of cash. See, e.g., Jeff Sommer, Why Starbucks Prices Went Up as Coffee Beans Got Cheaper, N.Y. TIMES (Aug. 15, 2015), https://www.nytimes.com/2015/08/16/your-money/why-starbucks- prices-went-up-as-coffee-beans-got-cheaper.html (Starbucks’s chief executive reports that “mobile payments now represent 20 percent of all in-store transactions in our U.S. stores, more than double the figure from only two years ago.”). 162 I.R.C. § 6050I(a). 163 See, e.g., Brittany Yantis, Monica Attia & Georgina Lethouris, Money Laundering, 55 AM. CRIM. L. REV. 1469, 1491–94 (2018) (describing the panoply of penalties endured by taxpayers who participate in money-laundering activities). 164 See supra notes 74–76 and accompanying text. 165 See George K. Yin, Getting Serious About Corporate Tax Shelters: Taking a Lesson from History, 54 SMU L. REV. 209, 210–14 (2001) (presenting, among other things, a historical overview of tax-shelter use). 166 See S. REP. NO. 109-54, at 1, 5 (2005). 167 See 150 CONG. REC. S2778, S2779 (daily ed. Mar. 12, 2004) (statement of Sen. Levin) (annual revenue loss associated with tax shelters in the multibillion-dollar range). 168 See Yin, supra note 165, at 214–15 (detailing the numerous legislative measures that Congress passed to rein in tax-shelter use). 169 I.R.C. § 469(a). 170 Id. § 6707(a). 171 Id. § 7701(o).

24

While these legislative measures proved effective in attacking tax shelter use, what should not be overlooked is the complementary role that technology has played in this reduction process. Consider how various technological advances have chilled tax shelter formation. Tax shelters of the past would thrive in an information void of sorts. By way of background, developing a particular tax shelter would take hundreds and sometimes thousands of hours by lawyers, accountants, and bankers.172 If a particular tax shelter could be tapped by a large number of taxpayers, the professionals involved all stood to make handsome fees; conversely, if only a small number of taxpayers could avail themselves of it, the economics of these labor-intensive arrangements made little sense. Tax professionals demanded that their clients subscribe to a strict secrecy protocol.173 Potential tax shelter candidates thus had to sign nondisclosure agreements, they were precluded from taking meeting notes, and their independent tax professionals were often barred from participation.174 This secrecy was designed in part to prevent duplication by competitors; but the great fear was that if the Service got wind of a particular tax shelter, the agency would issue a prohibitory notice for the specific tax shelter scheme.175 Now, however, technology has largely obliterated secrecy. If someone learns of a supposedly secret but profitable scheme, this person can post it on the internet, even anonymously, with little fear of retribution.176 Similarly, in an age when virtually anyone can tape record anyone, take laser-sharp pictures, and record videos — and immediately post these anywhere they wish on the internet — the veil of secrecy that used to shroud tax shelter planning is now routinely stripped away.177 In fact, it is fair to speculate, going forward, that tax professionals will be much less willing to invest the necessary time, energy, and effort to develop tax shelters that their competitors could readily steal or that the Service could immediately discover and quickly prohibit.178

172 See generally Tanina Rostain, Sheltering Lawyers: The Organized Tax Bar and the Tax Shelter Industry, 23 YALE J. ON REG. 77, 78 (2006) (“Large accounting firms, investment banks, and corporate law firms all became involved, designing and marketing hundreds of highly lucrative shelters that are estimated to have cost the federal government tens of billions in lost tax revenue dollars.”). 173 See, e.g., Janet Novack & Laura Saunders, The Hustling of X Rated Shelters, FORBES (Dec. 14, 1998), https://www.forbes.com/forbes/1998/1214/6213198a.html#3cb3891a71db (“What’s driving all this is money, of course. Lawyers who write opinion letters for corporate shelters can charge $500,000 in some cases, and opinions can be recycled for more fees.”). See generally Susan Beck, Gimme Shelters, AM. LAW., Nov. 1999, at 24; Paul Braverman, Helter Shelter, AM. LAW., Dec. 2003, at 51, 65; David Kay Johnston, Tax Magicians: Sham Shelters for Business Flourish as Scrutiny Fades, N.Y. TIMES (Dec. 19, 2000), https://www.nytimes.com/2000/12/19/business/tax-magicians-special-report-sham-shelters-for-business-flourish-scrutiny- fades.html. 174 See generally Joseph Bankman, The New Market in Corporate Tax Shelters, 83 TAX NOTES 1775, 1788 (1999) (“[A] promoter has no generally enforceable intellectual property rights in the idea around which the tax shelter is built. . . . Promoters attempt to limit . . . expropriation [of their ideas] by requiring confidentiality agreements from prospective purchasers and their advisors.”); Novack & Saunders, supra note 173 (“In addition to mass-market ideas, PricewaterhouseCoopers (PwC) also sells . . . ‘black box’ products. These are ‘complex and unique strategies that we do not publicize broadly,’ he says. Each can save a client from tens of millions to hundreds of millions of dollars in tax.”). 175 See, e.g., Joshua D. Blank, Overcoming Overdisclosure: Toward Tax Shelter Detection, 56 UCLA L. REV. 1629, 1647 (2009) (“After learning that accounting firms had been actively marketing the intermediary corporation tax shelter, the IRS designated this tax strategy as a listed transaction in Notice 2001-16, thus subjecting it to mandatory disclosure.”). 176 See generally Novack & Saunders, supra note 173 (“Recently the IRS got an anonymous letter blowing the lid off a particularly smelly scheme, and it was signed simply ‘A Pressured Practitioner.’”). 177 See, e.g., Reporting Team, Paradise Papers: Apple’s Secret Tax Bolthole Revealed, BBC (Nov. 6, 2017), https://www.bbc.com/news/world-us-canada-41889787 (“The world’s most profitable firm has a secretive new structure that would enable it to continue avoiding billions in taxes, the Paradise Papers show. . . . The papers are a huge batch of leaked documents mostly from offshore law firm Appleby, along with corporate registries in 19 tax jurisdictions, which reveal the financial dealings of politicians, celebrities, corporate giants and business leaders.”). 178 By definition, any time a taxpayer now enters into a so-called confidential transaction (i.e., “offered to a taxpayer under conditions of confidentiality and for which the taxpayer has paid an advisor a minimum fee”), it is automatically classified as a “reportable transaction” and, as such, must be disclosed to the Service. Treas. Reg. § 1.6011-4(b)(3) (2010).

25

Technology also affords the Service far better oversight over those tax return preparers who promote questionable tax return positions. With respect to the original Form 1040, there was no requirement that the tax return preparer identify himself/herself and sign it.179 Many decades later, Congress instituted a rule that tax return preparers identify and sign those tax returns that they prepare.180 While the signature requirement provided marginally better oversight, especially meaningful oversight happened when the Service mandated that every practitioner secure a practitioner tax identification number, or P.T.I.N., and place it on whatever tax returns they prepared.181 With this P.T.I.N. requirement, the Service could use technology to readily examine the tax returns of suspect practitioners and determine if there were systematic noncompliance activities.182 Aside from gaining an information advantage and a tool to track practitioners, technology has provided the Service with another edge: advanced algorithms now brightly illuminate tax returns in ways that are detrimental to taxpayers who embrace aggressive reporting positions. In the past, tax returns with income and offsetting losses may have been able to elude detection. However, this is no longer the case. Now, using advanced computer algorithms,183 the agency can more readily identify and, if need be, audit those tax returns that report large amounts of income and correspondingly large offsetting losses. The mere threat of such an action deters taxpayers who might embrace aggressive tax positions.184

3. Establishment of Hidden Offshore Accounts

In the past, a common method for tax evasion was to park investment funds offshore. Due to what were assumed to be impregnable foreign bank secrecy laws,185 U.S. taxpayers thought that their offshore tax activities would go undetected.186 However, once again, technology has fundamentally changed tax compliance on a global level. Akin to the tax shelter arena, technological changes have vastly reduced the scope of secrecy. Consider the plight of those taxpayers who employed the services of Mossack Fonseca, a Panamanian- based company that established over 200,000 offshore entities.187 Many of these offshore entities were

179 Note that the original 1913 tax return allowed tax returns to be prepared by an “authorized agent” of the taxpayer. See I.R.S. Form 1040 (1913), https://www.irs.gov/pub/irs-utl/1913.pdf. The enabling statute did not reference tax return preparers. See Revenue Act of 1913, ch. 16, 38 Stat. 114. 180 I.R.C. § 6109(a)(4). 181 See Treas. Reg. § 1.6109-2(d) (2010) (“[B]eginning after December 31, 2010, to obtain a preparer tax identification number or other prescribed identifying number, a tax return preparer must be an attorney, certified public accountant, enrolled agent, or registered tax return preparer authorized to practice before the Internal Revenue Service under 31 U.S.C. 330 and the regulations thereunder.”). 182 See INTERNAL REVENUE SERV., PUB. 4832, RETURN PREPARER REVIEW 33 (Dec. 2009), http://www.irs.gov/pub/irs- pdf/p4832.pdf (“Registration of all tax return preparers will enable the IRS to collect more accurate data on return preparers.”). 183 See Browning, supra note 146. 184 See, e.g., Blank, supra note 175, at 1640 (“Government officials and academics have widely praised the disclosure approach as an effective response to the tax shelter problem. They have argued that mandatory disclosure rules fortified by monetary penalties aid the audit process, chill participation in abusive tax strategies, and serve as an early warning system for lawmakers.”). 185 See, e.g., Sébastien Guex, The Origins of the Swiss Banking Secrecy Law and Its Repercussions for Swiss Federal Policy, 74 BUS. HIST. REV. 237 (2000). 186 See Alexia Fernández Campbell, The Cost of Corporate Tax Avoidance, ATLANTIC (Apr. 14, 2016), https://www.theatlantic.com/business/archive/2016/04/corporate-tax-avoidance/478293/ (“Large corporations such as Pfizer, Walmart, IBM, and Apple have stashed billions of dollars via more than 1,500 subsidiaries in tax havens like the British Virgin Islands and the Cayman Islands, according to the report, which analyzed the companies’ Securities and Exchange Commission filings. Though this practice isn’t illegal, keeping profits offshore lowers the taxes owed in the United States, and this ends up costing the U.S. government about $111 billion each year in lost revenue, by Oxfam’s calculations.”). 187 See Tom Morgan & Raziye Akkoc, Database Dump Reveals 200,000 Secret Offshore Account Details, TELEGRAPH (May 10, 2016, 7:13 AM), https:// www.telegraph.co.uk/news/2016/05/09/panama-papers-full-database-revealed/.

26

established for legitimate business reasons; however, many others were established for illegal tax evasion.188 It took the investigative work of just one individual (whose identity is still undisclosed) using technology to uncover and then leak the incriminating documents to a consortium of news agencies, which analyzed them and then published stories that made these illicit accounts known on the global stage.189 The release of this information caused numerous politicians to resign from their posts, to be fired, to be criminally charged, or simply to be shamed.190 In a somewhat similar manner, one person (Bradley Birkenfeld, a UBS banker) revealed the intricacies of the Swiss banking system and the ways in which this system enabled U.S. taxpayers to evade their taxes.191 This information proved so revealing that, for Birkenfeld’s whistle-blowing efforts, the Service awarded him the sum of $104 million.192 Even beyond these random disclosures, technological advances have afforded the Service unprecedented access to systematized information. Consider the fact that the institution of third-party reporting requirements has made tax compliance a routine phenomenon in many spheres of economic existence.193 Throughout the twentieth century, use of third-party tax information returns was limited to tracking income earned in the United States. However, this limitation no longer exists: technological advances have erased logistical barriers that previously precluded foreign banks from being able to report information regarding their U.S. investors to the Service in a cost-efficient manner. Congress has capitalized upon this technology-driven opportunity, instituting a broad set of laws known as the Foreign Account Tax Compliance Act (FATCA), which require foreign financial institutions to disclose the identifies of U.S. investors or endure a punishing withholding tax on their U.S. investments.194 Many foreign financial institutions have chosen to fulfill the disclosure reporting requirements.195

188 See Jesse Drucker, U.S. Prosecutors Bring Their First Charges over the Panama Papers, N.Y. TIMES (Dec. 4, 2018), https://www.nytimes.com/2018/12/04/business/panama-papers-indictment.html?login=email&auth=login-email (“According to the indictment, the men [working for and affiliated with Mossack Fonseca] helped United States taxpayers evade taxes by using undisclosed foreign accounts and shell companies in places like the British Virgin Islands, Hong Kong and Panama. They also instructed them on how to repatriate those offshore funds to the United States while concealing them from the I.R.S.”). 189 See Luke Harding, What Are the Panama Papers? A Guide to History’s Biggest Data Leak, GUARDIAN (Apr. 5, 2016), https://www.theguardian.com/news/2016/apr/03/what-you-need-to-know-about-the-panama-papers (“The Panama Papers are an unprecedented leak of 11.5m files from the database of the world’s fourth biggest offshore law firm, Mossack Fonseca. The records were obtained from an anonymous source by the German newspaper Süddeutsche Zeitung, which shared them with the International Consortium of Investigative Journalists (ICIJ). The ICIJ then shared them with a large network of international partners, including the Guardian and the BBC.”). 190 Luisa Kroll, Panama Papers Fallout: Iceland’s PM Resigns, Ukraine’s President Pressured, Billionaire Responds, FORBES (Apr. 5, 2016, 1:31 PM), https://www.forbes.com /sites/luisakroll/2016/04/05/panama-papers-fallout-icelands-pm-resigns- ukraines-under-pressure-russian-billionaire-responds/#5629f42b4cc8. 191 See Matthew Beddingfield & Colleen Murphy, Bloomberg BNA: The UBS Whistle-Blower Who Won’t Back Down, GOV’T ACCOUNTABILITY PROJECT (Apr. 4, 2017), https://whistleblower.org/in-the-news/bloomberg-bna-the-ubs-whistle-blower-who- wont-back-down (“[Bradley Birkenfeld] is known for blowing open a massive tax evasion scheme at UBS Group AG in 2007, which led to a $780 million fine, a cascade of tax treaties and information about previously secretive banking practices, and reforms in the industry.”). 192 See David Kocieniewski, Whistle-Blower Awarded $104 Million by I.R.S., N.Y. TIMES (Sept. 11, 2012), https://www.nytimes.com/2012/09/12/business/whistle-blower-awarded-104-million-by-irs.html. 193 See generally Soled, supra note 134. 194 See Melissa A. Dizdarevic, Comment, The FATCA Provisions of the Hire Act: Boldly Going Where No Withholding Has Gone Before, 79 FORDHAM L. REV. 2967, 2967 (2011) (“In an effort to crack down on offshore tax evasion, the United States is implementing a new set of information reporting and withholding requirements on foreign banks and other foreign entities. These provisions, known as the Foreign Account Tax Compliance Act (FATCA) . . . require thirty percent withholding of the entity’s U.S.-source income, unless they disclose specific information regarding their customers’ identities and account balances.”). 195 See generally Patrick Temple-West, U.S. Says 77,000 Banks, Firms Sign Up to Fight Tax Evasion, REUTERS (June 2, 2014, 1:47 PM), https://www.reuters.com/article/us-usa-tax-fatca/u-s-says-77000-banks-firms-sign-up-to-fight-tax-evasion- idUSKBN0ED1U620140602.

27

Consider, too, the fact that in an age of information availability due to technology, it is increasingly harder for taxpayers to realistically claim ignorance. A simple Google search of foreign account reporting requirements reveals many websites with information on what a taxpayer must report.196 The days are long past when taxpayers could come before a court and contend that they did not know that they had to make disclosures197 or even claim that they relied upon the advice of their tax professionals not to disclose.198 Technology has eliminated this defense — if it ever truly existed.199 Given these twenty-first-century developments, hidden offshore accounts are subject to an increased risk of discovery.200 Utilizing its subpoena power, the Service can track records that pinpoint taxpayer offshore activities, including taxpayers’ cellular phone records, credit card bills, and passport entries/exits. Access to information of this sort can readily reveal if particular taxpayers traveled abroad, had foreign contacts, and engaged in overseas transactions. Put more floridly, oceans once precluded the Service from gaining information access; however, technology has metaphorically drained these oceans, granting the agency access to information that can no longer hide on the ocean floor.

IV. TRENDS: TECHNOLOGICAL ADVANCES THAT MAY SUBVERT TAX COMPLIANCE

Clearly, technology makes certain forms of tax evasion increasingly difficult. In particular, technology almost certainly decreases noncompliance for taxpayers who have income mainly from wages, interest, dividends, and capital gains, all of which are subject to expansive withholding rules and extensive third- party reporting. Technology has enabled the Service to make great strides in ensuring greater taxpayer compliance on these and other types of income. On the flip side, however, technology also opens up new doors to tax noncompliance. The same technological changes that are available to the Service are, of course, also available to private individuals and firms. These technological changes have made certain forms of tax evasion easier to perpetuate. Particularly in environments where there are global supply chains, intellectual property is readily transferable, decentralized modes of economic commerce are increasingly commonplace, and electronic currency provides easy pathways to user anonymity.

196 Not surprisingly, in a Google search the first item that appears is the IRS website that offers detailed compliance instructions. See Report of Foreign Bank and Financial Accounts (FBAR), IRS (last reviewed or updated Jan. 9, 2020), https://www.irs.gov/businesses/small-businesses-self-employed/report-of-foreign-bank-and-financial-accounts-fbar. 197 See, e.g., Moore v. United States, 115 A.F.T.R.2d (RIA) 1375 (W.D. Wash. 2015) (taxpayer’s “age and ignorance” were no excuse for not disclosing the existence of offshore accounts). 198 See, e.g., United States v. Williams, 489 F. App’x 655, 660 n.6 (4th Cir. 2012) (“[T]o the extent [the taxpayer] asserts he was unaware of the FBAR [Foreign Bank Account Report] requirement because his attorneys or accountants never informed him, his ignorance also resulted from his own recklessness. [The taxpayer] concedes that from 1993–2000 he never informed his accountant of the existence of the foreign accounts — even after retaining counsel and with the knowledge that authorities were aware of the existence of the accounts.”). 199 Taxpayers are cognizant of this fact, and, as a result, many have sought to participate in the Service’s Offshore Voluntary Disclosure Program or risk criminal prosecution. Laura Saunders, The IRS Is Still Coming for You, Offshore Tax Cheats, WALL ST. J. (Sept. 14, 2018, 9:39 AM), https://www.wsj.com/articles/the-irs-is-still-coming-for-you-offshore-tax-cheats- 1536917401?ns=prod/accounts-wsj (pointing out that “[h]iding money from the U.S. government is a lot harder than it used to be” and adding that as a result of the Service’s limited-amnesty program, it has been able to collect $11.1 billion from individuals); Danielle Douglas, It Is Getting Harder to Hide Money from Uncle Sam Offshore, WASH. POST (May 22, 2014), https://www.washingtonpost.com/business/economy/it-is-getting-harder-to-hide-money-from-uncle-sam- offshore/2014/05/22/4e3f28de-dfba-11e3-810f-764fe508b82d_ story.html?utm_term=.61e7638e4889 (“Government efforts to combat offshore tax evasion are making it harder for the well-off to dodge the Internal Revenue Service.”). 200 I.R.C. § 7602; see Presley v. United States, 895 F.3d 1284, 1288 (11th Cir. 2018) (citing United States v. Clarke, 134 S. Ct. 2361, 2364 (2014)) (“To ensure compliance with the tax code, Congress designed a system that gives the IRS ‘broad statutory authority to summon a taxpayer to produce documents or give testimony relevant to determining tax liability.’”).

28

In light of these technological advances, Subpart A details emerging taxpayer tax dodges. Subpart B then considers reform options that seek to curtail taxpayer use of these strategies.

A. Emerging Twenty-First-Century Tax-Noncompliance Strategies

Taxpayers who wish to shirk their civic duties must adjust their methodologies to account for the here and now. Thus, in an era in which technology dominates the economy, it comes as no surprise that tax evaders seek to harness technological advances in ways that further their tax-minimization agendas. Evidence for this proposition manifests itself in three evolving forms of tax noncompliance: (1) the rise of blockchain technology, (2) the growth of transactions in a cross-border setting, and (3) the development of the gig economy. Consider the dynamics of each tax-noncompliance mode.

1. The Rise of Blockchain Technology

In 2008, an author published a white paper describing a P2P electronic cash system known as Bitcoin.201 Since then, the underlying architecture of Bitcoin — i.e., the blockchain — has become much more than an infrastructure for electronic payments. Blockchain is a decentralized ledger that records ownership and value transfers with no need for a trusted intermediary.202 As such, blockchain provides a novel framework for autonomously executed transactions that traditionally required a trusted third party. Contractual arrangements can be programmed onto the blockchain and automatically executed once a triggering event occurs.203 These contracts have earned the moniker smart contracts.204 Since the contractual execution is decentralized, neither a party to the contract nor a third party like a government can prevent execution. For example, inheritance funds can be programmed to be transferred to an heir when that heir reaches a certain age, and business profits can be programmed to be distributed once certain financial targets are met; the blockchain can even facilitate tangible transfers (e.g., the operation of a vehicle ignition can be conditioned upon complete transfer of funds from a buyer’s account to a seller’s account). Blockchain technology essentially creates “private regulatory frameworks.”205 While such a system may offer societal benefits, it also presents significant challenges to the tax system. In order to understand the challenges to tax enforcement, it is helpful to consider some of the unique characteristics of blockchain-based applications, namely, (a) disintermediation, (b) pseudoanonymity, (c) tamper resistance, and (d) autonomy.

a. Disintermediation

Blockchain technology offers the ability to document value transfers without the need for a trusted intermediary, such as banks, credit card companies, and other financial clearinghouses. This presents a unique challenge to any tax system because modern tax enforcement relies heavily on reporting by financial

201 See Satoshi Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System, BITCOIN (2008), https://bitcoin.org/bitcoin.pdf. 202 For an explanation of the operation of blockchain technology, see Primavera De Filippi & Aaron Wright, Blockchain and the Law: The Rule of Code 13–57 (2018). 203 Id. at 43–45. 204 See Craig A. de Ridder, Mercedes K. Tunstall & Nathalie Prescott, Recognition of Smart Contracts in the United States, 29 INTELL. PROP. & TECH. L.J. 17 (2017). 205 DE FILIPPI & WRIGHT, supra note 202, at 5.

29

intermediaries: empirical evidence demonstrates that tax-compliance rates are significantly increased when third-party reporting is involved.206 There are multiple mechanisms by which financial intermediaries strengthen tax compliance. They issue to taxpayers tax information returns (e.g., Form 1099s) pertaining to interest, dividends, and capital gains that assist taxpayers in reporting their taxes. At the same time, they submit the same information to the Service, which enables the Service to match information that the agency receives from financial intermediaries to the information received from taxpayers. These very same financial intermediaries sometimes even serve as tax collection agents, required to withhold taxes on payments that they clear.207 If financial intermediaries fail to meet their reporting and withholding requirements, they themselves may be subject to tax-enforcement actions, liable for their customers’ unpaid tax liabilities.208 However, if financial intermediaries are taken out of the picture by blockchain technology, tax authorities lose a powerful enforcement mechanism. As long as the two parties to a blockchain transaction agree not to voluntarily disclose the transaction to the tax authority, there is nothing that the tax authority can do to collect information on the transaction. Given that the system is fairly anonymous (as discussed later), it is possible that taxable events could proverbially “disappear into the ether,” hindering the Service and other tax authorities in the fulfillment of their oversight mission. Further, even if both parties to a blockchain transaction are compliant taxpayers who voluntarily share information with the tax authorities, enforcement still suffers. It is easy and economical to verify transaction information by using centralized databases held by large financial intermediaries, and to use the intermediary as a central point of enforcement. The alternative — enforcement that is solely dependent on the cooperation of dispersed end users — is much more complex, and it is also likely to be much more expensive.

b. Pseudoanonymity

The problem of disintermediation is exacerbated by the high level of anonymity that blockchain technologies accord their users. Blockchain applications can operate with no need to identify participants. Anyone can start a blockchain “wallet” at virtually no cost, without providing any identifying information. Parties do not even need to know one another for a transaction to be consummated.209 This anonymity makes it hard for tax authorities to detect the identity of tax cheats who use blockchain to facilitate their illicit activity. Blockchain is thus a particularly powerful instrument in the hands of tax evaders, even in the face of recent advances in international tax enforcement. As noted earlier, in 2010 the United States adopted the Foreign Account Tax Compliance Act (FATCA).210 Congress instituted this law as an instrument to target tax evasion by those taxpayers who use offshore bank accounts located in secretive tax haven jurisdictions. FATCA requires foreign financial institutions to deliver the account information of U.S. taxpayers to the Service or face burdensome U.S. taxes on their U.S.-sourced income.211 In the context of FATCA, multiple countries have agreed to share

206 For a discussion of third-party reporting and its efficacy in producing tax compliance, see Leandra Lederman & Joseph C. Dugan, Information Matters in Tax Enforcement, 2020 BYU L. REV. (forthcoming 2020) (manuscript at 18–25) (Ind. Legal Studies, Research Paper No. 396, 2019). 207 See, e.g., I.R.C. §§ 871(a), 881 (imposing 30 percent withholding tax on certain U.S. payments to foreign taxpayers). 208 Id. § 1461 (making withholding agents primarily liable for under-withheld taxes). 209 See DE FILIPPI & WRIGHT, supra note 202, at 38–39 (discussing the pseudoanonymity features of blockchain technology). 210 See Foreign Account Tax Compliance Act, Pub. L. No. 111-147, §§ 501–62, 124 Stat. 71, 97–118 (2010). 211 Id.

30

information with the United States,212 and the Organisation for Economic Co-operation and Development (OECD) has developed a “Common Reporting Standards” (CRS) framework based in large part on the FATCA model.213 Under the CRS framework, financial institutions share information with tax authorities, and the latter share information with one another. More than 100 jurisdictions have adopted binding CRS instruments.214 However, blockchain provides a medium to navigate around this elaborate statutory framework. As noted earlier, as part of the blockchain process, value exchanges do not have to rely on intermediaries, and parties to a transaction need not identify themselves. Even for jurisdictions in which blockchain agents wish to identify themselves, it is not clear that they could do so. The combination of disintermediation and anonymity has led to the characterization of blockchain applications as “super tax havens.”215

c. Tamper Resistance

Another unique aspect of blockchain technology is its tamper resistance. “[O]nce information has been recorded to a blockchain, it becomes exceptionally hard to change or delete.”216 Transactions cannot simply be undone: undoing a transaction requires a majority of the dispersed network participants to agree to undo the transaction, a near-impossible task. When tax evasion is involved, once such a transaction is executed, no financial institution or government can unwind the transaction. This means that even if fraud is identified, no one can remedy it. Thus, this makes the recovery of lost tax payments especially challenging even if the illicit activity is discovered.

d. Autonomy

Blockchain facilitates the autonomous execution of smart contracts. This means that once a malicious code is released using blockchain, there is no stopping it, even if one becomes aware of the code’s existence. The execution is based on a dispersed confirmation process, and governments cannot simply ask what may amount to millions of users to stop using a particular blockchain application. The existence of legal standards aimed at preventing such transactions becomes irrelevant. “Autonomous systems do not need to abide by existing rules and jurisdictional constraints; they can be designed to bypass or simply ignore the laws of a particular jurisdiction. Once deployed on a blockchain, these systems will continue to operate.”217 Thus, even when tax authorities identify a blockchain-based application aimed at tax evasion, there is nothing that a government can do short of shutting down the internet. ______

212 Multiple countries have entered into agreements with the United States to share information under FATCA. See Foreign Account Tax Compliance Act (FATCA), U.S. DEP’T OF TREASURY (last updated Jan. 13, 2020), https://www.treasury.gov/resource- center/tax-policy/treaties/pages/fatca.aspx. 213 See Org. for Econ. Co-operation and Dev. (OECD), Standard for Automatic Exchange of Financial Account Information in Tax Matters (2d ed. 2017), https://doi.org/10.1787/9789264267992-en. 214 See ORG. FOR ECON. CO-OPERATION AND DEV. (OECD), CRS MULTILATERAL COMPETENT AUTHORITY AGREEMENT (2018), https://www.oecd.org/tax/automatic-exchange /international-framework-for-the-crs/multilateral-competent-authority- agreement.pdf. For the jurisdictions that signed on to the agreement, see CRS by Jurisdiction, OECD, http://www .oecd.org/tax/automatic-exchange/crs-implementation-and-assistance/crs-by-jurisdiction/ (last visited Feb. 27, 2020). 215 Omri Marian, Are Cryptocurrencies Super Tax Havens?, 112 MICH. L. REV. FIRST IMPRESSIONS 38, 39 (2013) (“Cryptocurrencies possess the traditional characteristics of tax havens: earnings are not subject to taxation and taxpayers’ anonymity is maintained.”). 216 DE FILIPPI & WRIGHT, supra note 202, at 35. 217 Id. at 44.

31

These blockchain technology traits have led several scholars to note the potential of blockchain technology to be used as an instrument of tax evasion,218 and there is at least some anecdotal evidence that the technology is being used as such. For example, in United States v. Coinbase, Inc.,219 the Service successfully forced Coinbase (a cryptocurrency exchange) to turn over information about its account holders. During the period applicable for the request, Coinbase served millions of customers each year, but fewer than 1,000 U.S. taxpayers reported any transactions in cryptocurrencies on their tax returns during the same period.220 Given that Congress has historically drafted our nation’s tax laws without blockchain technology in mind, this technology presents a tax-enforcement challenge. The application of existing legal standards to blockchain-based transactions is uncertain. To date, the Service has issued only two pieces of administrative guidance on the tax consequences of transactions involving cryptocurrencies.221 While the guidance offers some clarity on a few basic issues, it falls short of answering important questions on timing, amount, and character of income inclusion. The Service has justifiably been harshly criticized for not issuing additional guidance.222 This lack of guidance has resulted in three negative outcomes. First, even the best-intentioned taxpayers may find it hard to comply with the law. There are legitimate questions about how to report certain blockchain transactions, and the lack of guidance may cause taxpayers to take legal positions on their tax returns that seem reasonable but that may eventually turn out to be wrong. Second, a lack of guidance creates legal uncertainty, which may discourage entrepreneurs from engaging in this new technology in ways that may be beneficial to society. Third, the regulatory vacuum allows aggressive taxpayers to use the ambiguity for their benefit, and there is some evidence that this indeed is happening.223

218 See Marian, supra note 215; Eric Engle, Is Bitcoin Rat Poison: Cryptocurrency, Crime, and Counterfeiting, 16 J. HIGH TECH. L. 340, 378–83 (2016) (discussing the potential use of blockchain technology as a tax evasion instrument); Manoj Viswanathan, Tax Compliance in a Decentralizing Economy, 34 GA. ST. U. L. REV. 283, 325–26 (2018) (same). 219 United States v. Coinbase, Inc., No. 17-CV-01431-JSC, 2017 WL 5890052 (N.D. Cal. Nov. 28, 2017). 220 Id. at *4. 221 I.R.S. Notice 2014-21; Rev. Rul. 2019-24. 222 See Nikhilesh De, U.S. Lawmakers ‘Strongly Urge’ IRS to Update Crypto Tax Guidance, COINDESK (Sept. 20, 2018), https://www.coindesk.com/us-lawmakers-strongly-urge-irs-to-update-crypto-tax-guidance. 223 Consider one current such example: the use of cryptocurrencies to avoid the recognition principles found in Code section 1259. Section 1259, also known as the “constructive sale” rule, prevents taxpayers from designing transactions through which they dispose of appreciated financial positions without triggering an income-realization event. For example, instead of selling appreciated stock, which would trigger capital gains (I.R.C. § 1001), a taxpayer may enter into a prepaid forward contract to sell the stock in the future. The taxpayer is in the exact same position as if she sold the stock, even though she actually did not. Section 1259 assures that such a transaction is treated as a disposition of the stock for tax purposes. Section 1259, however, is only applicable to an “appreciated financial position.” Section 1259(b)(1) defines appreciated financial position to include any appreciated position with respect to “any stock, debt instrument, or partnership interest.” This definition can reasonably be read as not including cryptocurrencies. Several reports suggest that appreciated cryptocurrency positions are being used in the exact same way that section 1259 was supposed to prevent. The IRS should, at a minimum, explore whether it has sufficient regulatory authority under section 1259 to shut down such transactions. The IRS seems to have broad regulatory authority in the context of constructive sales under Code section 1259(f). For a discussion of Bitcoin schemes associated with Section 1259, see Ted R. Stotzer, Rebirth of the ‘Estee Lauder Transaction’ Using Virtual Currencies, 60 TAX NOTES 1389 (Sept. 3, 2018) (describing the tax strategy of using cryptocurrencies to avoid the limits of section 1259); Lynnley Browning, Crypto Lenders Push No-Tax Perk of Leveraging Bitcoin for Cash, BLOOMBERG (Apr. 2, 2019), https://www.bloomberg.com/news/articles/2019-04-02/crypto-lenders-push-no-tax-perk-of- leveraging-bitcoin-for-cash (describing the use of similar strategies in practice).

32

2. The Growth of Transactions in a Cross-Border Setting

As discussed above,224 detection-avoidance devices and the retention of skilled abettors have always been essential ingredients for tax noncompliance. Even before recent technological developments, these ingredients were potentially much more potent in a cross-border setting. Given jurisdictional limitations, the Service traditionally has faced significant challenges applying the usual investigative tools to detect evasion in an international setting,225 especially if the foreign income generated is held in a foreign institution, devoid of any U.S. connections. Moreover, even if the Service is able to detect unreported offshore income, the long-standing “revenue rule” generally prevents the United States from utilizing its conventional enforcement mechanisms on foreign assets in order to satisfy U.S. tax debts.226 With respect to the “skilled abettor” ingredient of the noncompliance recipe, in the cross-border context this group traditionally has included not only foreign financial institutions that might welcome business from U.S. taxpayers and harbor little concern about the jurisdictional reach of the Service but also some foreign governments themselves. Due to financial opportunism, as well as historic, cultural, and political reasons, a number of countries have come to be viewed as tax havens that provide not only very low tax rates but bank secrecy as well.227 Technological developments have not only eased the implementation of customary forms of offshore tax noncompliance, but they have also spawned a whole new spectrum of noncompliance techniques. For example, whereas hiding funds offshore fifty years ago might have necessitated the physical transportation of large sums of cash, today electronic movement of money exists, ensuring that technological hurdles are trivial. Moreover, as discussed in Subpart A.1, new forms of storing and transferring value, such as cryptocurrencies, provide wholly new vehicles for shifting money across borders. Once a U.S. resident is able to shift financial assets to offshore accounts, technology also makes it much easier for taxpayers to administer and access those foreign accounts, whether through remote wire transfers, use of debit cards, or other vehicles. Of course, taxpayers must be willing to assume the risk that technological advancements that facilitate the movement of funds offshore and allow easy access to them might also enhance the tax authorities’ ability to trace that money. For example, the digitization of bank records raises the risk that a single rogue employee, such as UBS’s Bradley Birkenfeld, can jeopardize the privacy protection of thousands of people with a single flash drive.228 Furthermore, simultaneously with these technological advances, Congress and the Service have added new legal tools that increase detection risk in a cross-border setting, thereby potentially altering another essential ingredient to tax noncompliance — the taxpayer’s mindset of maximizing expected financial returns. For example, Congress has expanded reporting obligations (and related noncompliance penalties)

224 See supra notes 42 and 55 and accompanying text. 225 See, e.g., TAX DIV. U.S. DEP’T OF JUSTICE, SUMMONS ENFORCEMENT MANUAL: AN INTRODUCTION 70–75 (2014) (describing special circumstances involving cross-border summonses). 226 The common-law “revenue rule” provides that the courts of one country are under no obligation to recognize or enforce another country’s tax judgment. See Samuel D. Brunson, The U.S. as Tax Haven? Aiding Developing Countries by Revoking the Revenue Rule, 5 COLUM. J. TAX L. 170, 182–87 (2014) (describing the history of the revenue rule). Even among the United States’ “sixty-eight income tax treaties currently in force, only five include provisions permitting the enforcement of foreign tax judgments.” Id. at 183–84 (footnote omitted). 227 In this context, it should be noted that many countries have viewed the United States as a tax haven where foreign residents might park money, receiving interest exempt from U.S. tax and outside of the home country’s jurisdictional reach. See generally Brunson, supra note 226. 228 See Beddingfield & Murphy, supra note 191.

33

with respect to transfers to and from foreign trusts229 and, as part of FATCA,230 has expanded obligations (with significant penalties) for taxpayers self-reporting information with respect to foreign financial assets. Also, as discussed previously, Congress has significantly expanded the obligation of foreign financial institutions to report, either directly or indirectly, information regarding accounts held by U.S. citizens.231 The Treasury Department has expanded its network of bilateral income tax treaties and other international agreements (some in connection with FATCA) that, at least in certain circumstances, allows the United States to obtain information about income derived by a U.S. citizen from foreign institutions.232 At the same time, the Service has taken steps to dissuade taxpayers from noncompliance by, in particular, creating incentives for noncompliant taxpayers with overseas accounts to enter the U.S. tax-compliance system through a series of elaborate offshore voluntary disclosure programs.233 While these legislative and administrative developments are not themselves technological developments, the implementation of many of them — in particular, the FATCA requirements of information reporting by foreign financial institutions — relies heavily on increases in computer power and other technological advances that would not have been possible several decades ago. In the international setting, perhaps the most significant and current technology-driven threat to the U.S. tax base does not fit squarely within the type of noncompliance (or tax evasion) discussed throughout most of this analysis. As defined earlier, “tax evasion” occurs when taxpayers purposefully do not pay what the Code instructs is their appropriate tax burden, while “tax avoidance” occurs when taxpayers legitimately and legally reduce their tax burden.234 Bearing this distinction in mind, many large multinational corporations use aggressive tax-avoidance strategies that exploit gaps and mismatches in different countries’ tax rules to shift taxable profits to low- or no-tax jurisdictions. Because these profit-shifting strategies often rely on vague rules or inconsistencies among tax systems, these strategies are generally viewed as avoidance (or “aggressive tax planning”) rather than as tax evasion or tax noncompliance.235 Nonetheless, over the past half-decade, the so-called Base Erosion and Profit Shifting (BEPS) Project, a joint effort coordinated by the OECD at the request of the G20 and involving more than sixty countries, has attempted to address these aggressive tax-planning avoidance strategies.236 The rapid evolution of information and communication technology has been one of the principle accelerants of cross-border base erosion and profit shifting. These technology developments have enabled many types of traditional business activities to be conducted on a much larger scale and over longer

229 See, e.g., I.R.C. §§ 6039F (large gifts received from foreign trusts and other foreign persons), 6048 (transfers to foreign trusts), 6677 (related penalties). 230 See Foreign Account Tax Compliance Act, Pub. L. No. 111-147, §§ 501–62, 124 Stat. 71, 97–118 (2010). 231 I.R.C. § 6038D. 232 See IRM 4.60.1.1.1 (Oct. 15, 2018) (listing international agreements that allow exchange of tax-related information between nations, including bilateral tax treaties, tax information exchange agreements, and FATCA-related intergovernmental agreements); see also U.S. Model Income Tax Convention art. 26 (U.S. Dep’t of Treasury 2016) (model provision providing for exchange of information); Reciprocal Model 1A Intergovernmental Agreement (U.S. Dep’t of Treasury 2014) (model agreement for governmental information exchange under FATCA). 233 I.R.C. §§ 1471–74. 234 See supra note 2 (difference between evasion and avoidance). 235 See ORG. FOR ECON. CO-OPERATION AND DEV. (OECD), OECD/G20 BASE EROSION AND PROFIT SHIFTING PROJECT: 2015 FINAL REPORTS 29 (2015), https://www.oecd.org/ctp /beps-2015-final-reports.htm (characterizing this type of aggressive activity by multinational corporations as “legal tax planning techniques, rather than offshore tax evasion”). 236 See Org. for Econ. Co-operation and Dev. (OECD), Background Brief — Inclusive Framework on BEPS 7 (Jan. 2017).

34

distances than was previously possible; and they have also facilitated the creation of entirely new forms of business conduct, such as large-scale social networks and other user-participation platforms.237 As recognized in the Action 1 Report of the BEPS Project,238 this shift toward a digital economy creates a number of challenges for the United States and other governments.239 Most notably, technological developments, such as the use of websites or mobile device applications, enable a nonresident company to interact with customers in a country without needing a significant physical presence (or “nexus”) in that country. This ability to maintain a significant economic presence, void of a physical presence, undermines one of the fundamental touchstones of the international tax system for the past century — namely, the idea that physical presence through a “” is the critical jurisdictional nexus for allowing source-based taxation.240 Beyond circumventing the physical presence / permanent establishment threshold for source-based taxing, multinational companies have turned to other techniques to minimize their tax burdens. In particular, they have become adept at using technology (and technology-driven business developments) to shift taxable profits out of both the residence country and the countries where their business models allow them to derive economic value. Many of these strategies employ greater computer power to implement and track data- dependent transfer-pricing systems, reflecting more aggressive yet still traditional profit shifting. However, due to the mobility of the digital economy, evidenced by computer servers and other technological devices that can be “planted” virtually anywhere with no sense of permanence, new avenues to tax avoidance/evasion are being opened. As recognized in the BEPS Action 1 Report, this mobility is reflected in “(i) the intangibles on which the digital economy relies heavily, (ii) users, and (iii) business functions as a consequence of the decreased need for local personnel to perform certain functions as well as the flexibility in many cases to choose the location of servers and other resources.”241 For example, multinational corporations have an incentive to undervalue transferred intangibles at the time of transfer to an affiliate in a low-tax jurisdiction, and then to allocate large portions of the corporate group’s profits to that affiliate based on its legal ownership of the previously undervalued intangible.242 Alternatively, the multinational group might contractually allocate functions, assets, and risks to the low-tax affiliate, thereby eroding the tax base in the market jurisdiction “in a way that does not fully reflect the actual conduct of the parties, and that would not be chosen in the absence of tax considerations.”243 Given that these trends rely on features of the digital economy that do not easily square with the assumptions underlying the traditional international tax system, these problems appear to be intractable.

237 See generally Org. for Econ. Co-operation and Dev. (OECD), OECD/G20 Base Erosion and Profit Shifting Project: Addressing the Tax Challenges of the Digital Economy, Action 1: 2015 Final Report 54–64 (2015), https://doi.org/10.1787/23132612 [hereinafter BEPS Action 1 Report]. 238 Id. The BEPS Project issued fifteen separate action reports in 2015, addressing various aspects of the base erosion and profit-shifting problem. See generally ORG. FOR ECON. CO-OPERATION AND DEV. (OECD), OECD/G20 BASE EROSION AND PROFIT SHIFTING PROJECT: EXECUTIVE SUMMARIES, 2015 FINAL REPORTS (2015) [hereinafter EXECUTIVE SUMMARIES] (summarizing each report). The digitalization of the economy was identified as having a significant impact on the issues featured in at least six of the action reports. See ORG. FOR ECON. CO-OPERATION AND DEV. (OECD), OECD/G20 BASE EROSION AND PROFIT SHIFTING PROJECT: TAX CHALLENGES ARISING FROM DIGITALISATION, INTERIM REPORT 2018, at 18 (Mar. 16, 2018), https://doi.org/10.1787/23132612 [hereinafter 2018 DIGITALISATION INTERIM REPORT]. 239 See BEPS ACTION 1 REPORT, supra note 237, at 77–84. As noted in the report, “[i]n many cases, the nature of the strategies used to achieve BEPS in digital business is similar to the nature of strategies used to achieve BEPS in more traditional businesses. Some of the key characteristics of the digital economy may, however, exacerbate risks of BEPS in some circumstances.” Id. at 78. 240 Id. at 79 (acknowledging that “while the ability of a company to earn revenue from customers in a country without having a [permanent establishment] in that country is not unique to digital businesses, it is available at a greater scale in the digital economy than was previously the case”). 241 Id. at 64 (emphasis omitted). 242 Id. at 80. 243 Id.

35

3. The Development of the Gig Economy

A discussion regarding tax noncompliance in the so-called gig economy first requires (a) an understanding of the gig economy and its dynamics and (b) an overview of relevant tax issues. With this background in mind, this analysis then explores (c) why tax noncompliance has become a ubiquitous feature of the gig economy.

a. Fundamentals of the Gig Economy

Technological advancements over the last decade have enabled many innovative business models. The term gig economy stems from the notion that each work assignment is akin to performing an individual “gig,” and gig workers expect to maintain a short-term relationship with their clients.244 Many studies use the term gig economy interchangeably with other terms, such as sharing economy, peer economy, collaborative economy, on-demand economy, matching economy, access economy, and platform economy.245 In general, companies participating in the gig economy operate as online intermediaries that use app- or web-based software platforms to match suppliers and consumers of goods or services. These platforms typically feature mechanisms for accepting payments and settling transactions. Ride-hailing (Uber and Lyft), home-sharing (Airbnb and HomeAway), peer-to-peer sales (eBay and Etsy), and service-based platforms (TaskRabbit and Postmates) are common examples of the gig economy.246 However, the gig economy’s scope is amorphous and expanding.247 To date, the most mature enterprises of the gig economy are the ride-hailing and home-sharing platforms; and, as such, this analysis uses them for illustrative purposes. Although the gig economy lacks a universally recognized definition and comprehensive data regarding its operations,248 studies consistently agree on its central features. First, there are many more consumers than suppliers for a given platform. Second, most workers participate in the gig economy on a part-time basis to earn a secondary income that supplements their primary income. Third, millennials constitute the

244 See Nicole Kobie, What Is the Gig Economy and Why Is It So Controversial?, WIRED (Feb. 7, 2018), http://www.wired.co.uk/article/what-is-the-gig-economy-meaning-definition-why-is-it-called-gig-economy; Working in the Gig Economy, U.S. GOVERNMENT ACCOUNTABILITY OFFICE: WATCHBLOG (Aug. 30, 2018), https://blog.gao.gov/2018/08/30 /working- in-the-gig-economy/. 245 See Sarah Donovan, David Bradley & Jon Shimabukuro, Cong. Research Serv., R44365, What Does the Gig Economy Mean for Workers? (2016), https ://fas.org/sgp/crs/misc/R44365.pdf; Joyce Beebe, Baker Inst. for Pub. Pol’y, How Should We Tax the Sharing Economy? (Oct. 24, 2018), https://www.bakerinstitute.org/media /files/files/53fb91b2/bi-report-102418-cpf- sharingeconomytax.pdf; Niam Yaraghi & Shamika Ravi, Brookings India, The Current and Future State of the Sharing Economy (Mar. 2017), https://www.brookings.edu/wp-content/uploads/2016/12/sharing economy_032017 final.pdf. 246 See, e.g., Diana Farrell & Fiona Greig, JPMorgan Chase & Co. Inst., Paychecks, Paydays, and the Online Platform Economy: Big Data on Income Volatility (Feb. 2016), https://www.jpmorganchase.com/corporate/institute/document /jpmc- institute-volatility-2-report.pdf; The Power of Connection: Peer-to-Peer Businesses: Hearing Before the H. Comm. on Small Bus., 113th Cong. 3–4 (2014) (statement of Arun Sundararajan, Professor and NEC Faculty Fellow, N.Y. Univ. Stern Sch. of Bus.); Caroline Bruckner, Kogod Tax Pol’y Ctr., Shortchanged: The Tax Compliance Challenges of Small Business Operators Driving the On-Demand Platform Economy (May 2016), https://www.american.edu/kogod/research/upload /shortchanged.pdf. A 2017 proposed U.S. Senate bill on worker classifications uses a similar range of examples for a “marketplace platform.” New Economy Works to Guarantee Independence and Growth Act of 2017, S. 1549 / H.R. 4165, 115th Cong. (2017), https://www.congress.gov/bill/115th-congress/senate-bill/1549/all-info. 247 Other studies include a broader or narrower range of platform companies. See BEEBE, supra note 245. 248 See U.S. Gov’t Accountability Office, GAO-19-273R, Contingent Workforce: BLS Is Reassessing Measurement of Nontraditional Workers (2019), https://www.gao.gov /products/GAO-19-273R.

36

largest age group that participates in the gig economy. Finally, the gig economy has experienced and will continue to experience unprecedented growth.249

b. Overview of Tax Issues

Most businesses participating in the gig economy classify workers as independent contractors instead of employees, a position that has been challenged and repeatedly adjudicated.250 Although worker classification is beyond the scope of this analysis, this issue has important tax implications: each employee misclassified as an independent contractor significantly reduces the amount of federal employment taxes collected.251 As independent contractors, gig workers have tax-compliance obligations similar to those of small- business owners or self-employed individuals. Because their incomes are not subject to employee tax withholding, they must budget for self-employment and income taxes themselves and pay quarterly estimated taxes in order to avoid penalty imposition.252 In addition, when filing their annual tax returns (i.e., Form 1040s), gig workers need to report income earned and expenses incurred on Schedule C (Profit or Loss from Business) and Schedule SE (Self-Employment Tax).253

249 See Catherine Sullivan, Forty-Five Million Americans Say They Have Worked in the On-Demand Economy, While 86.5 Million Have Used It, According to New Survey, ASPEN INST. (Jan. 6, 2016), https://www.aspeninstitute.org/news/press- release/forty-five-million-americans-say-they-have-worked-demand-economy-while-865-million/ (noting that there are more consumers than suppliers); TREASURY INSPECTOR GEN. FOR TAX ADMIN., REF. NO. 2019-30-016, EXPANSION OF THE GIG ECONOMY WARRANTS FOCUS ON IMPROVING SELF-EMPLOYMENT TAX COMPLIANCE (2019), https://www.treasury.gov/tigta/auditreports/2019 reports/201930016fr.pdf (since the IRS last estimated the tax gap in 2008–2010, the gig economy has “emerged and grown considerably, with thousands of new taxpayers each year being responsible for self-employment taxes”); BRUCKNER, supra note 246, at 5 (“[I]n 2015, more than 75% of Lyft drivers reported working less than 15 hours per week, and more than half of Uber drivers worked less than 10 hours per week.”); PRICEWATERHOUSECOOPERS, THE SHARING ECONOMY: CONSUMER INTELLIGENCE SERIES 10 (Apr. 2015), https://www.pwc.com /us/en/technology/publications/assets/pwc-consumer-intelligence-series-the- sharing-economy.pdf (large swaths of the U.S. population have participated in the shared economy). 250 For instance, in Razak v. Uber Technologies, Inc., the U.S. District Court for the Eastern District of Pennsylvania held that Uber drivers are not employees. Razak v. Uber Technologies, Inc., No. 2-16-CV-00573, 2018 WL 1744467 (E.D. Pa. Apr. 11, 2018). The plaintiffs appealed to the U.S. Court of Appeals for the Third Circuit. Razak v. Uber Technologies, Inc., No. 18-1944 (3rd Cir. filed Apr. 27, 2018). This decision is pending. Uber recently settled a lawsuit with its drivers involving this issue but maintained their status as independent contractors. See Kate Conger, Uber Settles Drivers’ Lawsuit for $20 Million, N.Y. TIMES (Mar. 12, 2019), https://www.nytimes.com/2019/03/12/technology/uber-drivers-lawsuit-settle.html. 251 See Treasury Inspector Gen. for Tax Admin., Ref. No. 2018-30-077, Improvements to the SS-8 Program Are Needed to Help Workers and Improve Employment Tax Compliance 1 (2018), https://www.treasury.gov/tigta/auditreports /2018reports/201830077fr.pdf (noting that in 2016, “business that misclassifies a worker as an independent contractor reduces the amount of employment taxes it pays by $4,141”); U.S. Gov’t Accountability Office, GAO-17-371, Employment Taxes: Timely Use of National Search Program Results Would Help IRS Improve Compliance and Tax Gap Estimates 13, 46 tbl.11 (2017), https://www.gao.gov/assets/690/684162.pdf (stating that between 2008 and 2010, worker classification issues resulted in $44.3 billion in wage adjustments, the highest in employment tax examinations); see also Treasury Inspector Gen. for Tax Admin., supra note 249, at 1 (IRS studies show that the self-employment tax underreporting portion of the tax gap increased from $39 billion in 2001 to $57 billion in 2006 and to $65 billion in 2008–2010 (average). These dollar figures respectively represent 11 percent, 13 percent, and 14 percent of the total tax gap in those years. Self-employment income is often underreported for income tax purposes: based on the 2008–2010 average, sole proprietors underreported their net income by 64 percent, an increase from 57 percent in 2001). 252 See I.R.C. § 6654(d)(1). To avoid penalty, total estimated tax payments must equal at least 90 percent of the current year’s tax liability or 100 percent of last year’s liability. See id. 253 As self-employed individuals, gig workers are responsible for the full 15.3 percent self-employment tax on their earnings (consisting of 12.4 percent for Social Security and 2.9 percent for Medicare under the Federal Insurance Contributions Act), and they may deduct half of their self-employment tax from their net business income. INTERNAL REVENUE SERV., PUB. 334, TAX GUIDE FOR SMALL BUSINESS 4, 8–9 (2019).

37

c. Ubiquity of Tax Noncompliance

The Code and tax-compliance issues are beset by complexity. This complexity extends to gig workers, who often must endure nettlesome and labor-intensive reporting requirements. The misalignment of costs and benefits, where taxpayers’ perceived risk of detection by the Service is low (because the income amount is relatively small) and compliance costs are high, cultivates a mindset that tends to prioritize the individual over society — an essential condition for taxpayer noncompliance.254 Many surveys show that gig workers lack a basic understanding of how to properly report their taxable income.255 For example, the National Taxpayer Advocate’s (NTA) 2017 Annual Report to Congress lists gig economy noncompliance in its “Most Serious Problem” category, indicating that participants may not fully understand their tax obligations.256 An analysis by the Treasury Inspector General for Tax Administration (TIGTA) lends credence to the NTA’s report, finding evidence that between 2012 and 2016 approximately 25 percent of gig workers per year failed to report their incomes in an accurate manner257 and pointing out that the underreporting of income associated with the gig economy results in billions of dollars of lost tax revenue.258 In the gig economy, three factors drive tax noncompliance: lack of withholding requirements, uncertain information-reporting standards, and overreporting of business expenses.259 The first two factors have received more legislative and administrative attention,260 and, as such, they present a relatively clear path toward reform. The last factor, despite its salience, has not been the focus of much attention. Due to the deductible nature of associated expenses,261 it is obviously beneficial for gig workers to use such expenses as deductions. However, the accuracy of such deductions is often questionable. In fact, as platform

254 See STAFF OF J. COMM. ON TAXATION, 114TH CONG., JCX-49-15, COMPLEXITY IN THE FEDERAL TAX SYSTEM 15 (2015) (noting that “complexity or ambiguity in the tax law can lead to taxpayer cynicism potentially resulting in intentional noncompliance”); INTERNAL REVENUE SERV., supra note 4, at 27 tbl.9b (stating that, generally, the smaller the income, the lower the likelihood of audit). 255 See, e.g., Nat’l Ass’n for the Self-Employed, Small Biz Survey: 69% of Sharing Economy Entrepreneurs Received Zero Tax Guidance, NASE (May 11, 2016), https://www.nase.org/about-us/Nase_News/2016/05/11/small-biz-survey---69-of-sharing- economy-entrepreneurs-received-zero-tax-guidance (finding that 69 percent of survey respondents did not receive any tax guidance from the shared economy platform, and over 90 percent of small-business owners used a professional tax preparer or software); Nat’l Ass’n of Enrolled Agents, Survey Reveals Top Reasons Taxpayers Receive an IRS Notice for Incorrect Payment, NAEA (July 31, 2017), https://www.naea.org/newsroom/press-releases/survey-reveals-top-reasons-taxpayers-receive-irs-notice-incorrect- payment (“Independent contractors participating in the ‘gig economy’ . . . were cited as among those most at risk of failing to accurately report all of their income.”). 256 See TAXPAYER ADVOCATE SERV., NAT’L TAXPAYER ADVOCATE ANNUAL REPORT TO CONGRESS (2017), https://taxpayeradvocate.irs.gov/reports/2017-annual-report-to-congress. 257 See Treasury Inspector Gen. for Tax Admin., supra note 249, at 7. 258 Id. at 13. 259 See BRUCKNER, supra note 246, at 9 (listing the inconsistency, estimated tax payments, self-employment taxes, expensing and record keeping, and cost of compliance on Forms 1099-K and 1099-MISC as major challenges of compliance); Kathleen DeLaney Thomas, Taxing the Gig Economy, 166 U. PA. L. REV. 1415, 1421–22, 1428, 1453 (2018) (listing tax implications of nonemployee status, including the lack of withholding, the failure to accurately report self-employment taxes, and the overstatement of business deductions). 260 See, e.g., TREASURY INSPECTOR GEN. FOR TAX ADMIN., supra note 249, at 31–37 (newly revised I.R.C. § 6050W now requires expanded credit card information reporting); I.R.S. Tax Tip 2017-39 (Mar. 30, 2017), https://www.irs.gov/newsroom/keep-in-mind-these-basic-tax-tips-for-the-sharing-economy (suggesting that gig workers who are employees at another job withhold more taxes from their regular paychecks to circumvent the lack of platform withholding and avoid making estimated tax payments). 261 I.R.C. § 162(a). By way of illustration, this means that Uber drivers may deduct parking expenses, payments for car washes, tolls, car repairs, car depreciation, gas purchases, Uber-related fees, and any other expenses. They can also choose to deduct standard mileage rates ($0.58 per mile in 2019) instead of actual expenses related to driving their car for Uber. I.R.S. News Release IR-2018-251 (Dec. 14, 2018).

38

companies report the gross income that each of their contractors receives,262 the latter tend to inflate the expenses that they incur.263 Consider Uber drivers. The company reports drivers’ gross income to the Service and provides to each driver a tax summary statement, including Uber-related fees, reimbursements, miles traveled, and completed trips.264 As such, these gig workers are unlikely to underreport income or overstate Uber-related fees because inflating these expenses could, upon audit, be easily detected. However, these gig workers have another avenue to defeat their tax obligations: they can recast a portion or all of their personal expenses as being business-related. Several recent court cases confirm this commonplace practice of taxpayers masquerading their personal expenses as gig-related business expenses. In one case,265 a driver failed to provide documents to substantiate expenses for items such as uniforms and safety shoes (his attire when driving for Uber), commercial insurance, car repair and maintenance, and tax return preparation fees. He also reported higher Uber fees, device subscription fees, and miles driven for customers than those listed on his statement from Uber. In the end, the court disallowed all expenses that had no supporting documents, and only allowed expenses related to miles driven as reported by Uber. In two other cases,266 gig workers similarly exaggerated their business-related expenses. With the gig economy growing by leaps and bounds,267 the onslaught of such adjudications is poised to explode.268

B. Reform Options

Addressing these three evolving modes of tax noncompliance will be challenging but feasible. It will require that Congress and the Service utilize twenty-first-century technology to curtail twenty-first-century modes of cheating. To this end, various reform options that address these methods of noncompliance are discussed next, including (1) reforming tax laws applicable to blockchain, (2) reforming tax laws applicable to transactions in a cross-border setting, and (3) reforming tax laws applicable to the gig economy.

262 See Thomas, supra note 259, at 1435 (noting that Form 1099-K reports drivers’ gross income, inclusive of commissions and fees charged by Uber, which are deductible for tax purposes). 263 See Joel Slemrod, Brett Collins, Jeffrey L. Hoopes, Daniel Reck & Michael Sebastiani, Does Credit-Card Information Reporting Improve Small-Business Tax Compliance?, 149 J. PUB. ECON. 1 (2017) (finding that taxpayers subject to information- reporting requirements demonstrate “bunching” behaviors: when they report higher income on Form 1099-K, they also tend to report a similar increase in expenses that offsets the increased income, leading to limited change in taxable income); see also Bibek Adhikari, Timothy F. Harris & James Alm, Taxpayer Responses to Third-Party Income Reporting: Evidence from Spatial Variation Across the U.S., Paper Presented at the Ninth Annual IRS/TPC Joint Research Conference on Tax Administration in Washington, D.C. (June 2019) (finding that information requirements increase taxpayer reporting on items subject to the information requirements while creating incentives for taxpayers to decrease their reporting on items not subject to the information requirements). 264 An example of tax documents provided by Uber and Lyft can be found at Harry Campbell, Ultimate Tax Guide for Uber & Lyft Drivers (Updated for 2020), RIDE SHARE GUY (Jan. 30, 2020), https://therideshareguy.com/all-of-your-2014-rideshare-tax- questions-answered-a-turbotax-giveaway/. 265 See Hagos v. Commissioner, 116 T.C.M. (CCH) 357 (2018). 266 See Justine v. Commissioner, 114 T.C.M. (CCH) 413 (2017); Pokawa v. Commissioner, 114 T.C.M. (CCH) 355 (2017). 267 As tech-savvy millennials, many gig workers have turned to the internet to find tax solutions. There, they can secure quick and free responses; however, those responses are not always accurate, and the IRS has been regrettably slow to be responsive. Taxpayer Advocate Serv., Participants in the Sharing Economy Lack Adequate Guidance from the IRS, NTA BLOG (June 20, 2018), https://taxpayeradvocate.irs.gov/news/nta-blog-participants-in-the-sharing-economy-lack-adequate-guidance-from-the-irs. 268 In gig-related cases adjudicated to date, the courts’ position is clear and consistent. Under Code section 162, all “ordinary and necessary expenses paid or incurred in carrying on any or business” are deductible. Conversely, under Code section 262, personal, living, and family expenses are not deductible. The initial burden of proof is on taxpayers to show that their claimed deductions satisfy the Code’s requirements. I.R.C. § 7491(a). Among other methods, this means that taxpayers should maintain supporting documentation to substantiate their deductions. Id. § 274(d).

39

1. Reforming Tax Laws Applicable to Blockchain

The challenges presented by blockchain technology are formidable but not insurmountable. The first and most obvious action that tax authorities around the world can take is to clarify the application of existing tax law to blockchain technology. Tax authorities have indeed issued some guidance,269 but there is much room for improvement. Especially in the United States, where the Service has issued limited guidance, administrative action would be welcome. Exploring how current law applies to blockchain technology is also necessary in order to identify particular areas of vulnerability that may require dedicated legislative action. In some instances, it is plausible to expect that traditional tax-enforcement tools, slightly adjusted, may work equally well in the blockchain context. This is particularly true in the context of disintermediation. The libertarian-utopian description of the blockchain markets as an intermediary-less economy is highly exaggerated, implicitly assuming that financial intermediaries are just rent seekers that perform no useful function in the market. However, financial intermediaries play important roles in the economy;270 and, as such, they may play an integral role in the blockchain environment as well.271 In particular, most taxpayers are not tech-savvy enough to transact in cryptocurrencies on their own, and instead they use a third-party exchange or a custodian service. Tax authorities can target such intermediaries, as the Service did in the Coinbase case.272 The pseudoanonymity trait of the technology can also present challenges where parties transact with no intermediary. Governments have two viable courses of action to address this issue, namely, to de- anonymize users or to exert a cost on anonymity.273 De-anonymization is theoretically possible because all blockchain transactions leave a digital trace on a public ledger. Unlike hard cash, blockchain transactions are generally traceable to individual blockchain wallets. It is true that most owners of the wallets are anonymous, but not all of them are. This allows for the application of statistical methods to effectively perform a “triangulation”: identifying anonymous wallet owners based on known wallet owners and their transaction history. Such methods have indeed proven to be successful in identifying blockchain users or, at a minimum, in narrowing significantly the scope of the possible identities of users.274 In order to internalize the risks created by anonymity, a second measure that governments can consider is exerting costs upon users.275 This can be achieved in a variety of ways. For example, governments can impose a tax on any blockchain transaction as it interacts with the tangible economy. Akin to a , a tax can be levied on any transaction when a digital token is converted to a product or service, and the seller can be

269 For a recent comparative analysis of various countries’ approaches to the taxation of cryptocurrencies, see Why There’s Global Tax Confusion About Digital Currencies, EY (June 12, 2019), https://www.ey.com/en_gl/tax/why-there-is-global-tax- confusion-about-digital-currencies. 270 See generally Franklin Allen & Anthony M. Santomero, What Do Financial Intermediaries Do?, 25 J. BANK. & FIN. 271, 287, 289 (2001). For the role of intermediaries in the context of online markets, see generally George M. Giaglis, Stefan Klein & Robert M. O’Keefe, The Role of Intermediaries in Electronic Marketplaces: Developing a Contingency Model, 12 INFO. SYS. J. 231 (2002). 271 See Omri Marian, A Conceptual Framework for the Regulation of Cryptocurrencies, 82 U. CHI. L. REV. DIALOGUE 53, 67 (2015) (“Most users (both merchants and consumers) are not tech experts and will naturally turn to intermediaries to dispose of their cryptocurrencies.”). 272 United States v. Coinbase, Inc., No. 17-CV-01431-JSC, 2017 WL 5890052 (N.D. Cal. Nov. 28, 2017). 273 For a discussion of the imposition of costs on anonymity, see Marian, supra note 271, at 64–65. 274 See, e.g., Alex Biryukov, Dmitry Khovratovich & Ivan Pustogarov, Deanonymisation of Clients in Bitcoin P2P Network, CORNELL U. (2014), https://arxiv.org /abs/1405.7418. 275 See Marian, supra note 271, at 65 (“The gross tax should be set at a rate that is more likely to result in overcollection of taxes, so as to incentivize consumers to identify themselves, or otherwise force them to internalize the cost of their tax evasion.”).

40

required to collect the tax. The tax can be waived if the purchaser properly identifies himself or herself, relinquishing anonymity.276 It is much more difficult to address the challenges stemming from blockchain’s tamper resistance and its use for autonomously executed applications. As previously explained, governments can neither undo blockchain transactions nor prevent transactions from taking place;277 that is, there is no “off” switch for blockchain. Once an application is released to the internet and is made operational, it remains as long as people choose to keep using it. It logically follows that the only way to prevent malicious applications is to regulate them before they are executed. This means that the government should consider regulating blockchain entrepreneurs at the point of business commencement, by requiring them to undertake certain measures, including but not limited to identifying themselves and their application users, withholding taxes on their application users, and introducing fail-safe mechanisms into the software’s code that can be triggered if illicit activity is detected.278 Given global economic mobility, regulatory oversight risks driving blockchain entrepreneurs to jurisdictions that offer less regulated environments. There is some evidence that blockchain start-ups prefer such jurisdictions, in particular countries that have traditionally served as offshore tax havens.279 If this trend continues, these tax havens may offer illicit blockchain users an entry point to a system. The right response to this risk requires international cooperation to develop a coordinated regulatory framework that includes tax haven jurisdictions.280 Some initial steps in this direction have already been taken,281 but international coordination on blockchain regulation is still at an early stage. Finally, blockchain technology itself may offer a new enforcement instrument for tax authorities. Because a blockchain ledger is virtually impossible to manipulate, tax administrators can build blockchain- based applications to perform tax-enforcement functions. For example, in the context, such systems can already be applied.282 The virtues associated with such oversight are numerous, including the fact that the tax would be automatically and accurately collected, almost immune to taxpayer manipulation and less error-prone, and all of this would happen in real time.283 In addition, blockchain can also be used to codify consideration for profit allocation between related companies, thus streamlining compliance with intercompany pricing rules.284 Another attribute of blockchain is that it holds the potential to replace current information reporting. For example, instead of sending information reports to the Service (e.g., Form 1042s), transactions can automatically be recorded on a blockchain application with information delivered to the Service in real time rather than at the end of the year, substantially increasing transparency and accuracy. Whatever the application may be in the hands of tax authorities, it is clear that the technology offers powerful new enforcement tools. This runs counter to the fact that this technology was conceived (at least

276 Id. (“The consumer would effectively be in a position to elect between avoiding the tax by disclosing his or her identity and paying the gross tax to maintain his or her anonymity.”). 277 See DE FILIPPI & WRIGHT, supra note 202, and accompanying text. 278 See Marian, supra note 94, at 566–67 (discussing potential approaches to blockchain “entry-point” regulation). 279 Id. at 550–53 (describing the ascent of blockchain havens). 280 Id. at 562 (explaining the need for a coordinated approach). 281 Id. at 562–64 (describing current international initiatives in blockchain regulation). 282 Since consumption taxes are usually collected on a per-transaction basis and since purchase transactions are contractual executions, tax authorities could require transactions to be registered as smart contracts on a blockchain application. See Gijsbert Bulk, How Blockchain Could Transform the World of , ERNST & YOUNG (Apr. 24, 2018), https://www.ey.com/en_gl/trust/how-blockchain-could-transform-the-world-of-indirect-tax. 283 Id. 284 See How Blockchain Technology Could Improve the Tax System, PWC, https://www. pwc.co.uk/issues/futuretax/how- blockchain-technology-could-improve-tax-system.html (last visited Feb. 28, 2020).

41

by some) as a mechanism to usher in a libertarian tax-free utopia. Instead, technology may unleash a torrent of forceful tax enforcement.

2. Reforming Tax Laws Applicable to Transactions in a Cross-Border Setting

In an international setting, technology also holds the promise of curtailing some cross-border noncompliance that it may have facilitated. For example, advances in information and computer technology make possible the implementation of FATCA’s foreign financial institution reporting requirements, thereby mitigating through enforcement and deterrence the traditional tax-noncompliance device of hiding financial assets abroad.285 Technological advances have also made possible new administrative reporting requirements aimed at helping tax administrators better identify aggressive tax practices via and other cross-border tax-reduction techniques. For example, there is a concerted global effort to secure consistent tax-reporting practices across borders by multinational business enterprises. Evidence for this proposition is found in the BEPS minimum standards (Action 13), which call for large multinational enterprises to file Country-by-Country (CbC) reports using a common template.286 Consistent with these efforts, the Treasury Department recently implemented CbC reporting for large U.S. multinational enterprises with annual revenue of at least $850 million.287 In order to facilitate the exchange of information with partner jurisdictions, the Service is encouraging the filing of CbC reporting electronically, utilizing, in the jargon of computer aficionados, Extensible Markup Language schema, which is intended to be compatible with the format approved for OECD CbC reporting.288 The collection of such extensive data, its analysis and comparison, and its exchange with other countries would have been unthinkable even just a few years ago as a technological matter. While the utilization of technological advancements by tax administrators is a necessary component of addressing tax noncompliance in an increasingly digital age, it alone might not be sufficient. As digitalization becomes further interwoven into the economy, century-old tax principles for determining the amount of income appropriately subject to a nation’s tax laws have to be revisited, particularly as they apply to multinational business enterprises. Recognizing this challenge, the OECD/BEPS package of measures has attempted to create “the first substantial renovation of the international tax rules in almost a century.”289 The goal of the OECD/BEPS package is to ensure that, through “changes in domestic law and practices, and via treaty provisions,”290 eventually “profits will be reported where the economic activities that generate them are carried out and where value is created.”291 The BEPS Action 1 Report acknowledged that “[b]ecause the digital economy is increasingly becoming the economy itself, it would be difficult, if not impossible, to ring-fence the digital economy from the rest of the economy for tax purposes.”292 While the specific details of the BEPS project are beyond the scope of this paper, they signify that, at some point in

285 See Foreign Account Tax Compliance Act, Pub. L. No. 111-147, §§ 501–62, 124 Stat. 71, 97–118 (2010). 286 See 2018 Digitalisation Interim Report, supra note 238, at 115. 287 See Treas. Reg. § 1.6038-4 (2016). 288 See Frequently Asked Questions (FAQs): Country-by-Country Reporting, IRS (last reviewed or updated Jan. 10, 2020), https://www.irs.gov/businesses/international-businesses/frequently-asked-questions-faqs-country-by-country-reporting. 289 BEPS ACTION 1 REPORT, supra note 237, at 3. 290 Id. 291 Id. 292 Id. at 142.

42

the future, a consensus might be reached regarding modernized tax nexus and profit-allocation principles among governments as well as tax policy representatives of major multinational enterprises.293 The OECD, in recognition of the transformation that digitalization is playing in the economy, formed the Task Force on the Digital Economy (TFDE).294 Even as this task force is moving toward issuing reports aimed at a consensus-based long-term solution, it has released a public consultation document outlining possible approaches to the challenges arising from the digitalization of the economy.295 This consultation document proposed “pillars” that could form the basis for renovating international tax rules to address modern challenges. The first pillar focuses on fundamental aspects of the international tax system’s allocation of taxing rights. It revisits the underlying nexus principles that permit a country to exercise taxing jurisdiction, as well as related questions as to how profits (especially, but not exclusively, from digitalized enterprises) should be divided between and among countries.296 In response to digitalization challenges, the first pillar suggests three proposals, enumerated below, for revising nexus and profit allocation: • The first proposal, utilizing what is referred to as a “user participation” approach, focuses on “value created by certain highly digitalised business through developing an active and engaged user base, and soliciting data and content contributions from them”;297 it would allocate profit to “jurisdictions in which those businesses’ active and participatory user bases are located, irrespective of whether those businesses have a local physical presence.”298 This proposal would have particular significance for social media platforms and search engines, whose value is driven by user engagement and participation. The most significant hurdle to the adoption of this approach is that it focuses on a specific sector — digitalized business built on a user platform model — thereby running counter to the BEPS Action 1 Report’s inclination to try not to “ring-fence” the digital economy from the rest of the economy. • The second proposal emphasizes marketing intangibles, such as brand and trade names. Under this proposal, profits would be allocated based upon where businesses sought to capitalize upon such intangibles, requiring “marketing intangibles and risks associated with such intangibles to be allocated to the market jurisdiction”299 generating profitability. A perceived advantage of this standard is that it would apply to a broad range of businesses rather than merely a subset of highly digitalized businesses. • The third and final proposal allows a “significant economic presence” to trigger taxability. This approach challenges the long-standing focus on physical presence as the touchstone for taxing nexus. Instead, it would allow a country to tax “when a non-resident enterprise has a significant

293 While multinational enterprises might prefer to continue with an antiquated system that facilitates aggressive profit shifting, they realize that political and public revenue realities necessitate change, so they would rather have a seat at the table in the development of those changes. 294 The TFDE is a subsidiary body of the OECD/G20’s Inclusive Framework on BEPS, a group of more than 125 countries that have committed to implementing the BEPS package. See EXECUTIVE SUMMARIES, supra note 238, at 5. 295 See Org. for Econ. Co-operation and Dev. (OECD), OECD/G20 Base Erosion and Profit Shifting Project, Addressing the Tax Challenges of the Digitalisation of the Economy, Public Consultation Document (2019) [hereinafter 2019 Public Consultation Document]. The TFDE convened a well-attended meeting to discuss this document with representatives of multinational enterprises and other stakeholders. See Stephanie Soong Johnston, OECD Working on Impact Assessments of Digital Tax Proposals, 93 Tax Notes Int’l 1210 (Mar. 18, 2019). 296 See Org. for Econ. Co-operation and Dev. (OECD), OECD/G20 Base Erosion and Profit Shifting Project, Addressing the Tax Challenges of the Digitalisation of the Economy — Policy Note (Jan. 23, 2019). 297 2019 Public Consultation Document, supra note 295, at 9. 298 Id. at 10. 299 Id. at 12.

43

economic presence on the basis of factors that evidence a purposeful and sustained interaction with the jurisdiction via digital technology and other automated means.”300 Pillar two of the TFDE’s consultation document points to the deficiencies in the current arm’s-length standard that will continue to offer opportunities for significant risk of base erosion and profit shifting. Accordingly, this pillar proposes two antiabuse provisions, elaborated below, that are intended to prevent such base erosion and profit shifting: • The first proposal, a so-called income inclusion rule based on the United States’ new global intangible low-taxed income provision,301 “would tax the income of a foreign branch or a controlled entity if that income was subject to a low effective tax rate in the jurisdiction of establishment or residence.”302 By way of example, if a shareholder held a significant (e.g., 25 percent) ownership interest in a subsidiary incorporated in a low-tax jurisdiction, that shareholder would need to include a pro-rata share of the low-taxed subsidiary’s income in its income in the higher-tax jurisdiction, regardless of whether a distribution was actually made. • The second provision, loosely analogous to the United States’ new base erosion and antiabuse tax,303 would impose a tax on base-eroding payments. More specifically, it “would deny a deduction or treaty relief for certain payments unless that payment was subject to an effective tax rate at or above a minimum rate.”304 By way of example, if a corporation in a relatively high-tax jurisdiction made an interest payment to a related party in a relatively low-tax jurisdiction, this rule would deny a deduction for the paying corporation unless the interest was taxed at or above some minimum rate in the hands of the recipient corporation. In the international arena, these developments suggest that Congress and the Service join forces and approach tax compliance from two different angles. First, it is important that tax administrators receive sufficient funding to maintain a tactical advantage against both tax evasion and multinational enterprises’ well-funded, aggressive tax-avoidance schemes. Second, in an increasingly digital world, a significant part of the battle to protect the tax base also requires a rethinking (in cooperation with other countries) of the underlying international tax principles and the corresponding tax laws as they apply to the changes wrought by accelerating technology and resulting business models. In addressing both angles, it is important not only that the choices not only make sense from a theoretical perspective but also that the choices enable tax administrators to effectively monitor and enforce compliance in a feasible way given the financial and technological resources available.

3. Reforming Tax Laws Applicable to the Gig Economy

To address the issue of gig taxpayers masquerading their personal expenses as business expenses, this analysis proposes the following four solutions: (a) permit a simplified deduction allowance amount in lieu of actual expenses, (b) augment third-party reporting, (c) utilize technology to enhance taxpayer education, and (d) employ technology to enhance enforcement. Each is detailed below.

300 Id. at 16. 301 I.R.C. § 951A; see 2019 PUBLIC CONSULTATION DOCUMENT, supra note 295, at 26. 302 2019 Public Consultation Document, supra note 295, at 25. 303 I.R.C. § 59A. 304 2019 Public Consultation Document, supra note 295, at 25.

44

a. Permit a Simplified Deduction Allowance Amount

Presently, the Code requires gig workers to expend resources on tax compliance for what often amounts to a small amount of income. Faced with this burden, many gig workers choose not to be compliant. To alleviate some of the compliance burden, one option is for the Code to provide a so-called simplified deduction allowance amount (SDAA). Under this proposal, gross income minus SDAA would be subject to tax.305 Several OECD countries, including Italy and the UK, have considered and, in some cases, adopted similar measures.306 The concept of allowing a hypothetical set of expenses is not novel. Under the Code, this approach already exists, exemplified by the standard mileage deduction,307 the simplified method for the home office deduction,308 and the uniform depreciation deduction rules.309 In instituting an SDAA, a major design issue is whether it should be a percentage of the taxpayer’s gross revenue or a fixed dollar amount. There are shortcomings associated with each approach. Under a system using a percentage of the taxpayer’s gross revenue, there are several concerns. One is that gig workers with low business expenses would benefit compared with gig workers with high business expenses.310 Another is the anticipated tax revenue loss that would occur if this option were elective: taxpayers would select SDAA or the actual expenses incurred, whichever yielded the lowest tax burden. A final concern is that if this mode were elective, it would not obviate gig workers’ record-keeping burdens because they would still maintain a detailed set of records just in case this methodology proved more financially beneficial than reliance upon the SDAA. A different set of concerns surround a system utilizing a fixed dollar amount. One is that gig workers would now be guaranteed that the dollar amount equal to the SDAA would be received tax-free.311 This might discourage some gig workers from earning income beyond this threshold. Furthermore, if the Form 1099 issuing threshold (currently, $600 for Form 1099-MISC and $20,000 for Form 1099-K) were higher than the SDAA (say, equal to $500), gig workers might be reluctant to claim the SDAA at all lest they risk their entire income becoming visible to the Service. Under both the percentage and fixed amount systems, there is a trade-off between the benefits of simplification and the potential revenue loss that the government would experience, the latter of which would, of course, depend upon on the threshold percentage or dollar amount that Congress establishes and

305 See, e.g., Thomas, supra note 259, at 1454 (suggesting that gig taxpayers be offered a “standard business deduction”); Shu-Yi Oei & Diane M. Ring, Can Sharing Be Taxed?, 93 WASH. U. L. REV. 989, 1062 (2016) (promoting a “safe harbor” of allowable expenses). 306 See 2018 DIGITALISATION INTERIM REPORT, supra note 238, at 197 box 7.2. 307 See Treas. Reg. § 1.274-5(g) (2010). For the standard mileage deduction, taxpayers who use their cars for business purposes can elect to take this deduction instead of deducting actual car-related expenses. Total business miles driven multiplied by the standard mileage rate generates a standard mileage deduction. I.R.S. Notice 2019-2, § 3, https://www.irs .gov/pub/irs-drop/n- 19-02.pdf. 308 See Rev. Proc. 2013-13, § 1, 2013-6 I.R.B. 478. For the home office deduction, a simplified method provides taxpayers a $5 per square foot deduction for qualified business use of a home instead of the standard method that requires the calculation of actual home office expenses; furthermore, the simplified deduction method limits the home office to 300 square feet, therefore generating a maximum deduction of $1,500 per year. Simplified Option for Home Office Deduction, IRS (last reviewed or updated Jan. 16, 2020), https://www .irs.gov/businesses/small-businesses-self-employed/simplified-option-for-home-office-deduction. 309 I.R.C. § 168(a). 310 See Thomas, supra note 259, at 1461 (“[I]mposing a 60 percent [standard business deduction] creates a revenue loss to the government for any taxpayer with a net profit ratio that is higher than 40 percent. . . . The reverse would be true for taxpayers with a net profit ratio below 40 percent. . . . [The] [t]axpayer . . . would forego the [standard business deduction] and claim actual expenses.”). 311 Id. at 1457 (“[A flat dollar amount] would effectively exempt all business receipts from tax up to the amount of the flat [standard business deduction].”).

45

whether or not such systems would be elective or mandatory. In-depth and detailed studies would have to be undertaken to ascertain the appropriate approach.

b. Augment Third-Party Reporting

Although advancements in technology enable noncompliance, they also provide tools to reduce the compliance burden and facilitate traceable transactions, empowering the Service to fulfill its oversight mission. As previously noted, when there is third-party withholding or information reporting, taxpayers underreport a much lower percentage of income subject to the requirements.312 Although it does not rise to the level of withholding or information reporting, app-based and technology-enabled record tracking may facilitate taxpayer compliance. Consider the fact that there are multiple smartphone apps available to help gig workers track expenses. Foreceipt, Smart Receipts, and ReceiptBox are just a few examples of paperless receipt maintenance systems that provide real-time processing.313 In some apps, receipts can be synchronized across several devices, with some linked to cloud-based storage (e.g., Google Drive) or tax- filing software, while others can generate summary reports. Uber provides a case study to illustrate the utility of this technology. Uber supplies all of its drivers with a summary of miles driven when engaged in customer activity.314 What the company does not supply is a summary of miles that its workers may incur when looking for work between passengers or returning to the general pickup locations (e.g., the airport) after dropping off a passenger. Tracking these miles can be accomplished through the use of other apps such as Stride Drive, Hurdlr, MileIQ, Ridar, and QuickBooks Self-Employed.315 Some apps have add-on features of expense tracking, receipt imaging, and tax filing, which would help facilitate compliance. For the Service to harvest the information that gig-oriented apps offer and thus reduce taxpayer noncompliance, the agency needs to designate acceptable software versions and formats. In particular, the Service could enter into collaboration agreements with platform companies to ensure uniformity across the industrial spectrum.316 Once these standards are in place, Congress could require vendors such as Uber,317 Airbnb,318 and other gig companies to utilize particular apps and grant the Service access to this

312 See supra note 30 and accompanying text. 313 The authors reviewed top-rated apps and their functionalities in Apple Inc.’s iTunes store. The developers’ websites were also reviewed if accessible. The three apps all received at least a 4.5-star user rating (out of 5); however, this is not an exhaustive list. 314 See Campbell, supra note 264 (images of Uber and Lyft tax summaries and Form 1099). 315 This is similar to using GPS technology to ensure correct calculation of automobile expense deductions. James Alm & Jay A. Soled, The Internal Revenue Code and Automobiles: A Case Study of Taxpayer Noncompliance, 14 FLA. TAX REV. 419, 433– 34 (2013). 316 See Andrzej Posniak & Piotr Nowicki, INSIGHT: Taxation of the Digital Economy — Service Platforms, BLOOMBERG TAX (Oct. 25, 2018), https://news.bloombergtax.com /daily-tax-report-international/insight-taxation-of-the-digital- economyservice-platforms (recommending a voluntary method of cooperation, where tax authorities and platforms enter into data- sharing agreements). 317 See Intuit and Uber Partner to Simplify Filing Taxes for On-Demand Economy Workers, ACCOUNTINGWEB (Jan. 28, 2015), https://www.accountingweb.com/community /industry-insights/intuit-and-uber-partner-to-simplify-filing-taxes-for-on- demand-economy (Intuit and Uber partnered to offer a free version of the expense-tracking software and enable the ability to send data from this software to its tax-compliance software. The goal is to provide drivers “easy to use tools for managing business revenue and expenses.”). 318 See Resources for the 2018 US Tax Season, AIRBNB, https://web.archive.org /web/20190801000506/https://www.airbnb.com/help/article/1213/resources-for-the-2018-us-tax-season (last visited Apr. 1, 2019) (Airbnb partnered with Stride and TurboTax to offer resources to hosts, including tax guides, tax- management mobile apps, and discounts on additional help by H&R Block and TurboTax).

46

information.319 As a quid pro quo, Congress could grant tax credits related to the costs associated with adopting these technological advances.320

c. Utilize Technology to Enhance Taxpayer Education

In 2016, the Service launched its Sharing Economy Tax Center.321 This center functions essentially as a reservoir of available tax documents and links, designed to be helpful to both tax professionals and gig workers alike.322 Due to the complexity of the tax rules surrounding the gig economy, the National Taxpayer Advocate (NTA) has advocated for publications that are more relevant to help workers in these industries understand their civic obligations.323 To identify those areas of tax compliance that are particularly problematic, the Service could use analytics. For example, analytics would probably find that one source of taxpayer confusion is business expense items, which are subject to high levels of reporting errors. Another example would be nondeductible commuting miles. Technology applications could help the agency proactively identify areas where additional guidance and clarification could help taxpayers.324 The Service could also nudge taxpayers before noncompliance happens. For example, the Service could partner with app developers and platform companies to promote pop-up messages to remind taxpayers about estimated payment deadlines or warn drivers when an app detects unusual expense classification patterns (e.g., for five days in a row, in categorizing his rides, a driver does not select “personal expenses”). Expense-tracking apps could flag any large or unusual amounts classified as a business expense and require the worker to review or overwrite the entry.

319 See 2018 DIGITALISATION INTERIM REPORT, supra note 238, at 199 (“[I]ntroducing legislative measures which require platforms or other third parties to report payment and identification data of [peer-to-peer] users . . . could provide tax administrations with information needed to improve compliance or to enhance selection of cases for audit. The fact that data is reported would also be likely in itself to encourage greater self-reporting.”). 320 Although these apps and software keep digital records, determined taxpayers can still find ways to disguise personal expenses as business ones, or simply not report income. For example, drivers can still manually classify a personal trip as a business one on expense-tracking apps. Even with the apps that meet IRS requirements or are approved by platforms, it is virtually impossible or uneconomical for the IRS to detect on a case-by-case basis whether an expense is for a personal or business purpose. Gig workers can also engage in “off-the-platform” transactions with customers, such as providing clients with personal contact information for future services that do not use the platform and are conducted in cash. 321 See I.R.S. News Release IR-2016-110 (Aug. 22, 2016) (According to IRS Commissioner John A. Koskinen, “[t]his rapidly evolving area often presents new challenges for people engaged in these economic activities. . . . The IRS is working to help people in this area by providing them the information and resources they need to file accurate tax returns.”). 322 See NAT’L TAXPAYER ADVOCATE, 2 OBJECTIVES REPORT TO CONGRESS: FISCAL YEAR 2019, at 131–32 (June 27, 2018), https://taxpayeradvocate.irs.gov/Media/Default /Documents/2019-JRC/JRC19_Volume2.pdf (In response to the NTA listing sharing-economy noncompliance in the “Most Serious Problem” category, see supra note 256 and accompanying text, the IRS states that it has made significant efforts to provide critical tax-compliance guidance to gig economy participants and organizations, chief among them the Sharing Economy Tax Center webpage. In reply, the NTA encourages the IRS not to view the webpage “merely as a repository for a data dump.”). 323 See Taxpayer Advocate Serv., supra note 267 (“[A]n Airbnb host would have to sift through the 24-page Publication 527, and an Uber driver would have to navigate through the 50-page Publication 463, and they still might not understand how these rules apply to themselves.”). 324 See 2018 DIGITALISATION INTERIM REPORT, supra note 238, at 204 (“The increase in data availability and advancements in analytics are also leading to improvements in taxpayer services. This includes identifying ways to make it easier to understand and report tax obligations, for example by use of analytics on large data sets to identify areas of uncertainty or errors in reporting, as well as to understand where guidance and communication needs to be clearer for taxpayers, or where tax administration processes may need to be redesigned.”).

47

d. Employ Technology to Enhance Enforcement

Technology is expanding the Service’s capabilities, including the ability to enforce tax laws and audit noncompliant behaviors. However, it is no secret that the Service is resource-constrained;325 and a recent TIGTA report focusing on the gig economy states that even when the Service computer system identified repeated discrepancies, many were not pursued due to a lack of resources.326 Although the Service does not separately track noncompliance within the gig economy, the agency possesses a broad range of data, and it could use data analytics and AI327 to more effectively evaluate and define high-risk, noncompliant groups.328 Application of this process could result in more productive audit cases, yielding more revenue, improving current review procedures, and even automating some functions previously performed by revenue agents.329 In short, technology could prove transformative in making the Service a more effective collection agency.330 Another way of leveraging technology is for the Service to engage in targeted audits. It could conduct short-term, highly publicized audits that focus enforcement resources on the gig economy. Because of the tech-savvy nature of gig workers, they are likely to discuss and share their audit experiences within the gig community, in online chat rooms or internet discussion forums.331 Assuming that such audits are effective, the word on the proverbial street would be that tax compliance makes sense, lest one endures the Service’s scourge.

325 See Hearing on the Tax Filing Season Before the Subcomm. on Oversight of the H. Ways & Means Comm., 116th Cong. (2019) (statement of Nina Olson, National Taxpayer Advocate) (“IRS’s inflation-adjusted budget is more than 20 percent lower today than it was in FY 2010, and the IRS workforce overall is down by about 23 percent.”), https:/ /waysandmeans.house.gov/sites/democrats.waysandmeans.house.gov/files/documents/National%20Taxpayer%20Advocate%20T estimony-W&M%20Oversight%20Filing% 20Season%20Hearing....pdf; INTERNAL REVENUE SERV., supra note 4, at 25 tbl.9a (showing that the IRS has a low audit rate, less than 2 percent, for individual returns with business income under $1 million). 326 See TREASURY INSPECTOR GEN. FOR TAX ADMIN., supra note 249, at 9–11 (A determinant for audit case selection is taxpayers with prior discrepancies. One of the four repeater designations (codes) that could be assigned to a taxpayer with a potential discrepancy is “repeater not worked,” meaning that within the last three years, the taxpayer was identified as having a potential discrepancy at least once but the case was not further reviewed. Of all the gig economy taxpayer cases identified, 59 percent were assigned this code.). 327 For example, a group of Canadian professors recently created an AI-enabled tool that can predict, with 90 percent accuracy, the results of twenty-five tax law issues, including worker classification and deduction of home office expenses. This program’s strength originates from its ability to synthesize tomes of Canadian case law and then predict how the Canadian Revenue Agency or courts would view a particular fact pattern with specified input variables. The algorithm is continuously learning new rules as cases are added to the software. This essentially means that the computer system learns from input data, adapts to certain external circumstances, and replicates human skills used to perform relatively repetitive and low-value-added activities. Bryan Borzykowski, How AI Can Predict Tax Outcomes, CPA CAN. (July 4, 2018), https://www.cpacanada.ca/en/news/accounting/advisory/2018-07-04-ai-can-now-help-predict-tax-outcome-scenarios. 328 See supra notes 125 and 129 and accompanying text. 329 See 2018 DIGITALISATION INTERIM REPORT, supra note 238, at 203 (“A growing number of tax administrations are increasingly using algorithms to review the broad range of data to which they now have access in order to more effectively define risks. These new processes are replacing some audit actions, including audit selection, and other verification checks previously performed by people.”). 330 Id. at 202 (“Technology is in fact expanding the capabilities of tax administrations in a wide range of ways, to enhance the effectiveness of compliance activities, improve taxpayer services, and reduce compliance burdens.”). 331 See Shu-Yi Oei & Diane M. Ring, The Tax Lives of Uber Drivers: Evidence from Internet Discussion Forums, 8 COLUM. J. TAX L. 56, 85 (“The risk of IRS audit was a palpable concern in forum discussions, though one that posters might have overestimated.”); Dubin, Graetz & Wilde, supra note 37, at 395, 405 (noting that there are significant “spillover benefits” of IRS audits: whether taxpayers are audited or not, they report higher taxes because they believe that there is a higher chance of being audited).

48

V. CONCLUSION

Given the availability of information and the ability to process it, there are sound reasons for believing that the trends in technology point to a Service that will be empowered to increase tax compliance. However, this same information produced via technology is also available to taxpayers who may direct its use toward evading their tax responsibilities, particularly those utilizing electronic currency, participating in the global marketplace, and operating in the gig economy, all of whom have the incentive and resources to engage in new forms of tax evasion. There are also other considerations that are likely to negatively shape compliance trends. In particular, privacy considerations may arise that preclude or at least limit the Service’s use of these many technological innovations. Furthermore, technology does not stand still; thus, even if the Service is able to address current modes of tax noncompliance, newer and currently unknown methods are certain to emerge, all of which will require that the agency remain vigilant and responsive. Some likely developments will have uncertain effects. For example, technology may well make possible new tax policies (e.g., the introduction of a , a tax based on biometric information) that in the past have not been feasible. These policies will certainly have impacts — both positive and negative — on tax noncompliance. Despite these negative and uncertain effects on tax compliance, there are aspects of technology that will certainly positively affect compliance. In particular, improved technology may make advances in research possible, which in turn may lead to new strategies for dealing with noncompliance and for adding to the tools of the enforcement, trust, and service paradigms of tax administration.332 Overall, it seems plausible that the dominant trend of these technologies will be for the Service to contain tax noncompliance. However, this prediction comes with one crucial qualification that cannot be overemphasized: Congress must grant sufficient funding to the Service for the agency to gather, analyze, and utilize information that technological advances make possible. Absent such resources, tax noncompliance will remain entrenched and even flourish. Unfortunately, given the severe erosion of the Service’s budget in the last decade and its uncertain future prospects, it is difficult to remain overly optimistic about what the future holds for reducing noncompliance.333

332 Included here are such innovations as changing the wording on tax returns to increase the psychological cost of cheating and to appeal to social norms. See Kathleen DeLaney Thomas, The Psychic Cost of Tax Evasion, 56 B.C. L. REV. 617 (2015). 333 See supra note 38 and accompanying text.

49