The Tax Gap: No Trillion Dollar Silver Bullet

(pdf)

 

Internal Revenue Service (IRS) Commissioner Chuck Rettig raised eyebrows in a recent Senate Finance Committee hearing when he appeared to state that the “tax gap,” or the gap between tax revenue that is owed and tax revenue the IRS manages to collect, is much larger than previously thought.[1] But lawmakers who think that the tax gap could be a trillion dollar silver bullet should realize the limitations of both Rettig’s statement and the revenue that could potentially be recouped from increased enforcement efforts.

Not an Updated Institutional Estimate

First and foremost, the context of the Commissioner’s statement shows that it was an off-the-cuff estimate, not the announcement of a new IRS position.[2] Rettig was asked by Sen. Ron Wyden (D-OR) to provide his “personal opinion” about how big the tax gap is. Rettig responded by stating that he thought it was “not outlandish to believe that the actual tax gap could approach and possibly even exceed $1 trillion per year.”

That’s a notable statement, but it is certainly not an official estimate supported by genuine research. The IRS last released an official estimate of the size of the tax gap less than two years ago, using tax data from 2011-2013.[3] That estimate placed the size of the tax gap at a much smaller $381 billion after factoring in late payments and enforcement efforts.

Rettig is on somewhat of an island with his trillion dollar estimate. Nina Olson, the former Taxpayer Advocate, responded to Rettig’s comments by disagreeing with them, arguing that any calculation “still doesn’t add up to $1 trillion.”[4] Former IRS Commissioner Charles Rossotti recently estimated that the tax gap was higher than the IRS’s last analysis, but still only $574 billion.[5] Even the Biden Administration can’t seem to agree on the exact number — Treasury Secretary Janet Yellen placed the per-year figure closer to $700 billion.[6]

That an off-the-cuff statement like Rettig’s was reported as gospel truth is a serious indictment of a press corps whose job it is to inform readers about public policy matters. The misleading stories that followed the Senate Finance hearing have contributed to introduction of new legislation with utterly unrealistic revenue targets from closing the tax gap.

The truth of the matter is that the tax gap is probably not $1 trillion or more, but perhaps more importantly, that enforcement efforts would likely raise just a fraction of the real amount while imposing huge administrative and privacy burdens on taxpayers.

How Close to $1 Trillion Could the Tax Gap Be?

Now, there is reason to believe that the latest IRS estimate of the tax gap at $381 billion may be inaccurate. For example, the official IRS estimate of the tax gap does not factor in the growth of cryptocurrency and new research into offshore holdings.

A rough IRS estimate in 2018 placed the tax gap from cryptocurrency at about $11.5 billion,[7] back when the digital cryptocurrency market was worth an estimated $500 billion. Even assuming the IRS has failed to improve at all on its previously estimated noncompliance rate of 50 percent, this extrapolates to a cryptocurrency tax gap of just under $50 billion based on the 400 percent increase in global market capitalization that digital currencies have enjoyed since then.

That is of course a significant amount, but it is also only around 12 percent of the $381 billion tax gap estimate.

The IRS also has not made a concrete estimate of offshore holdings which are going untaxed. Research the IRS helped fund and authored in part by left-wing economist Gabriel Zucman placed the amount of American offshore holdings in low-tax jurisdictions at just over $1 trillion.[8] They arrive at this number by assuming that 95 percent of wealth in these holdings is concealed from the IRS, an estimate which is likely high.

Nevertheless, even if one assumes this data is entirely correct, the authors still estimate just $15 billion in uncollected tax revenue per year. Combining this number with the estimate of roughly $50 billion in uncollected cryptocurrency tax revenue, we are still shy of $500 billion, let alone $1 trillion.

Researchers have also pointed to pass-through business tax liability underreporting as another potential reason for a larger-than-estimated tax gap. For example, that same Zucman-authored report, which we can safely place on the high end of estimates, assumes 20 percent of pass-through business income goes unreported to the IRS.

IRS data shows that pass-through business income averaged about $982 billion between tax years 2011 and 2013.[9] If the amount of unreported pass-through income was 20 percent higher than reported, this would mean that around $200 billion in pass-through income goes unreported. Given that pass-through income faced effective tax rates between 15 and 25 percent before the 2017 tax law, that would mean roughly $40 billion in lost tax revenue under the most generous interpretation.[10]

The IRS, for its part, blames pass-through income for a far larger share of the tax gap. Of the $441 billion tax gap before late payments and enforcement efforts, the IRS attributes $110 billion to underreported pass-through business income, and another $45 billion to underreported self-employment tax liability.[11] In order for this to be true, closer to 60 percent of pass-through income would have to go unreported.

This is not intended to be a comprehensive analysis of all the potential causes of a higher-than-estimated tax gap. After all, no one can truly claim to know the exact size of the tax gap — if the exact amounts and sources of uncollected revenue were known, it would be far easier than it is for the IRS to collect it. Nevertheless, the picture becomes rather clear: the tax gap is almost certainly not $1 trillion and no bona fide research in existence today adequately supports the notion that it is.

What Could Be Gained from Increased Enforcement?

Wherever the tax gap is on the spectrum between $381 billion and $1 trillion, however, the simple truth is that there is no plausible path to reducing it to zero. The herculean administrative efforts required to do so, and the intrusive and economically damaging actions that are imposed upon taxpayers in the process, render it an impossibility.

In the wake of Commissioner Rettig’s testimony, President Biden announced his intention to seek $80 billion in new IRS funding over the next decade, hoping to use this to collect an additional $780 billion in taxes that would otherwise go untaxed.[12] For an agency as dysfunctional and backward as the IRS, with 60-year-old computer systems and massive paperwork backlogs, inflating its budget by an almost incomprehensible two-thirds is the definition of throwing good money after bad. Even if the agency were a paragon of efficiency however, there is reason to wonder if massive investments in tax enforcement can prove successful even just on revenue-generation terms.

For one thing, the IRS’s ability to close the tax gap thus far has proved limited. According to the latest agency estimates, enforcement efforts following voluntary compliance have only succeeded in collecting an additional $60 billion per year, and this number includes late payments that may have little to do with IRS enforcement efforts.[13] All told, enforcement efforts have succeeded in raising the compliance rate by an average of 2.2 percent. It’s not at all clear that an agency with this kind of history is capable of efficiently spending billions in new funding.

Between 2011 and 2013, the IRS dedicated just under half of its $12 billion budgets towards enforcement efforts.[14] Though over $5 billion a year in enforcement efforts was able to close the tax gap by $60 billion, it is unrealistic to extrapolate roughly this same rate of return as the Administration appears to be doing.

After all, tax enforcement sees diminishing returns over time. While a lower tax gap is generally a good thing, it’s not feasible nor desirable to aim for the tax gap to be $0. At a certain point, the cost of enforcing the collection of the next dollar in unpaid taxes is outweighed by the cost of doing so.

Neither are all increased enforcement efforts fruitful. Discussions of the tax gap often seem to presuppose that closing it is a matter of simply choosing to do so. Yet uncollected tax revenue often goes uncollected because it is difficult to identify. As an example of this, between 2015 to 2018, an IRS program selected business returns for audit. Yet 55 percent of these selected businesses saw no change in their tax assessment after audit.[15] 

It’s also the case that a holistic effort to reduce the tax gap would have consequences progressives may not like. For instance, improperly paid credits make up a significant portion of the tax gap, roughly 10 percent of the total.[16] The IRS estimates that between 21 and 26 percent of Earned Income Tax Credit payments are made in error,[17] and the same is true for 25 to 30 percent of Child Tax Credit payments and 30 percent of American Opportunity Tax credits.[18] Stricter enforcement of these refundable credits would primarily impact lower-income taxpayers, subjecting them to invasive audits and expensive enforcement actions.

Data from the Congressional Budget Office (CBO) would also recommend skepticism of the Administration’s revenue generation claims. The CBO estimates that increasing IRS enforcement funding by $20 billion over the next decade would increase revenue collections by $61 billion, while increasing enforcement funding by $40 billion over that same time period would increase revenues by $103 billion.[19] Based on these estimates, it is exceedingly unlikely that increasing IRS funding by $80 billion over a decade would yield $780 billion in additional revenue.

It is also worth pointing out that the vast majority of the tax gap is generated not by knowing efforts to defraud the federal government but instead by inadvertent errors on the part of taxpayers dealing with a confusing and expensive tax code that eats up nearly eight billion hours in compliance time each year.[20] As former Taxpayer Advocate Olson stated in a recent webinar:[21]

Equating the entire tax gap to “tax evasion” is just so disingenuous. And it’s also wrong; it’s incorrect. Evasion has a technical meaning under the law, and generally it requires mens rea, a criminal intent. So much of it is error, or inadvertent...there are a bunch of different types of noncompliance. And if you say [it’s all tax evasion] - and that’s what the Washington Post called it, that’s what the New York times called it - then that creates distrust among the taxpayers of the tax agencies. [They’re asking], “why am I paying if you’re letting all these people off the hook?”

Broadly speaking, as both a legal and a moral issue, intent matters and any effort to address the tax gap should start with a recognition that our tax code creates more liars than it allows. Congress and the IRS have a duty to simplify tax laws and regulations so that taxpayers have a better knowledge of how to comply.

Equally important to the idea of how big a budget increase the IRS can accommodate is where any funding hike should be spent. While advocates of the Biden administration’s approach tend to cite impressive dollar-return figures to justify hiring more auditors and criminal investigators, they rarely mention the potential of other expenditures to yield even higher returns. In a 2016 working paper on revenue estimates for various IRS program integrity proposals before Congress, CBO concluded that a dollar put toward IRS efforts to resolve ID theft and refund fraud had an initial rate of return to the Treasury that was over three times greater than a dollar spent on increasing audit coverage.[22]

These dynamics have long been well-known to tax administrators,[23] who realize that audits and investigations often result in protracted disputes with taxpayers that involve some of the highest-paid personnel in government. Thirty years ago, Ernest J. Dronenburg, then Vice Chairman of California’s State Board of Equalization, told a House Committee how creating  “better compliance within the system” and reducing “the number of people who drop out of the system because they are afraid or unsure of it” yielded major revenue gains:

For example, a .5% increase in voluntary compliance resulting from taxpayer education and changing attitudes would increase revenue in my state by over $400 million annually. Conversely, doubling our current audit coverage from 3% to 6% would produce less than half that amount.

Such projections do not reflect the collateral damage that can be avoided by abuses of the IRS’s law enforcement authority, which is a decades-long chapter in the Service’s history (see below).

Few Agencies Are Less Prepared for Budget Increases Than the IRS

The Internal Revenue Service remains mired in problems unrivaled by nearly any other agency of government. It employs some of the most antiquated computer systems still in modern use, has often sent out erroneous correspondence to taxpayers, suffers from near-constant processing delays, flouts the will of Congress in enforcement actions, and has acted overzealously toward taxpayers to such a degree that Congress saw fit to create the Taxpayer Advocate Service as a bulwark against IRS overreach. Earmarking piles of new IRS funding for enforcement purposes only is the equivalent of spending $1 million to build an addition on a house with a crumbling foundation.

While the agency’s historical struggles are the stuff of legend, its recent history offers little reason for optimism that it has improved in such a way that it could efficiently deploy billions in new funding. In March, the Government Accountability Office released an analysis of the IRS’s handling of the 2020 filing season that identified a number of significant problems that contributed to huge backlogs.[24] While the IRS naturally cannot be blamed for the outbreak of COVID-19 and the related disruptions to normal life, this episode was useful in illustrating how years-long failures to modernize the agency made it uniquely susceptible to COVID-related setbacks.

As an example, nearly one-quarter of business tax returns are still filed on paper and the GAO found that the IRS “has not comprehensively identified barriers to business-related e-filing nor taken specific actions to increase e-filing.” The result is an agency already buried in paperwork must sift through more paper submissions and, in the event of COVID-related shutdowns, was unable to complete timely processing of huge swaths of tax returns. Among other issues, this led to the agency being responsible for 50 percent more in refund interest paid out to taxpayers due to processing delays.

By 2021, the IRS had more than 30 years of experience with e-filing. The last 10 years in particular have seen rapid proliferation in the use of electronic documentation for major “paperwork events” like mortgage applications, and fewer and fewer transactions of money or information are conducted on paper.

Yet, the agency would be nowhere close to even keeping up with (much less getting ahead of) the technological curve without technologies and partnerships developed in the private sector. The Free File initiative, for example, was created precisely because the IRS was struggling to meet e-filing objectives established by Congressional directives in the IRS Restructuring and Reform Act of 1998. Most recently, it was the Free File platform (constructed with private sector know-how) that served as a way for millions of moderate-income Americans who had not filed tax returns to claim Economic Impact Payments online.[25]

Meanwhile, the agency still has a backlog of 1.7 million returns it is working through, many of which were filed in 2019 before COVID-19 was even discovered.[26] When accounting for returns from the most recent tax year, the backlog is more than 30 million returns.[27] A recent Treasury Inspector General for Tax Administration (TIGTA) report found that a major contributor to this backlog is a lack of working printers and copiers at processing facilities, a shortage that is evidently so bad that some IRS employees worry it could lead to a work stoppage due to insufficient capacity. According to TIGTA, fully 42 percent of printers and copiers at the Ogden, UT and Kansas City, MO facilities were broken or unusable.

If the IRS’s processing systems are this antiquated, it raises serious questions as to whether its fraud detection and enforcement systems are equally deficient. This is particularly notable since some efforts aimed at increasing tax compliance have suggested that additional funding be earmarked for enforcement purposes with mandatory audit percentages attached, as opposed to being available for investing in systems upgrades.[28]

Even former IRS Commissioner John Koskinen, who was a vocal advocate of increased funding for the agency during his leadership, expressed doubt about throwing $80 billion at the problem. He’s quoted as saying, “I’m not sure you’d be able to efficiently use that much money.”[29] His newfound uncertainty is a few years too late but nonetheless instructive as to the scale of the problems the agency faces.

Other Costs of Enforcement

Historically, many efforts to increase IRS enforcement budgets have been followed several years afterward with taxpayer rights and protection initiatives. Between 1988 and 2019, Congress has enacted five major legislative packages in reaction to tax administration maladies and abuses. Why such actions after the fact? One answer is that unfocused funding with an implied message of “almost anything goes” can result in overzealous IRS actions – and not just against “the rich.”

Prior to the passage of the Taxpayer Bill of Rights of 1988, Congress heard numerous “horror stories,” typified by  Pennsylvania businessman Tom Treadway, who lost his business due to a bogus $247,000 tax assessment that was later thrown out on appeal. The tax agency didn’t just go after Treadway though – his girlfriend’s bank account was raided in the process. Although some Members of Congress might dismiss accounts like these – including suicides of taxpayers later proven innocent – these occasional tragedies and the more common instances of bureaucratic snafus have been documented in hearings for more than 30 years.[30] Throughout this time, the IRS has all too often utilized enforcement tools provided in one context to pursue unsuspecting taxpayers in quite another situation.[31] The Service’s decision to employ asset forfeiture laws against innocent small business owners accused of “structured transactions” was a primary reason why the late Congressman and civil rights champion John Lewis (D-GA) helped lead to passage the Taxpayer First Act (TFA) of 2019.[32] 

Since passage of TFA nearly two years ago, concerns over taxpayer rights and concern about IRS misallocation of resources have unfortunately not abated.[33] For example, the Service has engaged in a relentless assault on what it calls “syndicated” charitable deductions for conservation and historic preservation,[34] including retroactive interpretations of tax guidance, capricious enforcement tactics, and arbitrary litigation strategies. Meanwhile, key taxpayer rights safeguards, such as requirements for supervisor approval of certain audit penalties and due process for appraisers, have been seriously undermined. Even the National Taxpayer Advocate has called upon the IRS to avoid further expenditures of time and money on this dangerous course, recommending that the Service “[d]evelop and publish additional guidance that contains sample easement provisions to assist taxpayers in drafting deeds that satisfy the statutory requirements for qualified conservation contributions.”[35]

Neither the Biden Administration nor its allies in Congress have specified how the bipartisan structure of checks and balances at the IRS can be repaired and strengthened before any more enforcement funding goes out the door.[36] The IRS Oversight Board, for example, was created in the IRS Restructuring and Reform Act of 1998 as one of three pillars (along with the Taxpayer Advocate and the Treasury Inspector General for Tax Administration) of agency accountability, providing private and public sector managerial expertise on how the tax agency can fulfill its mission. The 1998 law was designed out of recognition that while the IRS needed more funding, it needed smarter funding, guided by a culture of customer service, accountability, and respect for taxpayers. That entity has been dormant for the better part of 20 years.

The more recent problems of scarce resources at the Independent Office of Appeals have yet to receive any attention either.[37] Providing more money to the tax agency for enforcement without addressing these and other taxpayer rights matters will increase the likelihood of a civil liberties crisis across the tax filing population in the not too distant future.

Conclusion

It’s certainly true that there is a sizable tax gap, and that new technologies bring with them enforcement challenges. Nevertheless, irresponsible statements about the tax gap and the revenue that could be raised from attempting to close it do little more than mislead voters and generate phantom pay-fors, whereby politicians promise to pay for a spending item with nebulous promises to increase tax enforcement.

When it comes to the tax gap, efforts to close it should be grounded in data-derived expectations about what revenue actually could be gleaned from such an effort, not pie-in-the-sky estimates. And perhaps more importantly, it should be done with a keen understanding of the IRS’s structural failures and a comprehensive plan to address them before giving them greater audit and enforcement responsibilities.

If the IRS is to receive more funding, and a case can be made for doing so, it must be directed intelligently to fulfill the IRS’s mission: to “provide America's taxpayers top quality service by helping them understand and meet their tax responsibilities and by applying the tax law with integrity and fairness to all.” That mission, established more than 20 years ago, is not being met today.

Finally, policymakers must be transparent about the goal of this exercise: is it truly to increase voluntary compliance with the U.S. tax system, or merely to claim that the federal government is getting even tougher with some ill-defined group of “tax cheats?” In an era when law enforcement agencies at all levels of government are under increased scrutiny for their strategies, tactics, and budgets, public officials need to step up with more thoughtful, responsible proposals than an ill-targeted $80 billion IRS funding boost.

Simply throwing more money at the IRS than it knows what to do with is a recipe for disaster. Sadly, it’s a recipe that Washington has followed for decades.

---