Categories
tax administration

Criminal tax prosecutions

Yesterday’s Reuters story, Tax prosecutions plunge as Trump shifts crime-fighting efforts (12/11/205), was no surprise when it reported that DOJ tax prosecutions declined by 27% this year (through 11/1/2025) compared to 2024.

Reuters used data extracted from Westlaw from 1990 on. This year is an all time low as shown in this graph from the story:

The story reports some interesting and troubling details about what is going on (in addition to the dramatic reduction in IRS staffing and shuttering of the DOJ Tax Division). Some excerpts:

The administration made deep cuts to the Internal Revenue Service’s criminal investigative unit, and some of those who remained were ordered to start working on immigration cases or anti-crime patrols in Washington, according to government records and officials, speaking on the condition of anonymity because they were not allowed to discuss their work publicly. At the same time, the Justice Department closed its Tax Division, and officials said a third or more of the criminal lawyers who worked there quit.

Reuters used federal court dockets to count the number of Justice Department attorneys who appeared on behalf of the government in tax prosecutions between January and the beginning of November. Last year, about 420 did. This year, about 160 have.

Top Trump administration officials told prosecutors early this year that tax investigations were not a priority, three people familiar with the discussions said, speaking on the condition of anonymity to discuss the department’s internal deliberations. Participants concluded that the department’s new management was “very skeptical about white-collar crime and whether we should be doing those cases,” one person familiar with the discussions recalled.

At least a third of the roughly 80 criminal prosecutors who worked in the office [DOJ Tax Division] at the beginning of the year quit rather than be reassigned, two officials familiar with its staffing said.

U.S. attorneys’ offices that could pick up some of that slack have also lost prosecutors with experience in white collar cases, current and former Justice Department officials said. Department records show more than 1,000 lawyers have left U.S. attorneys’ offices this year, roughly double the number who quit or were pushed out in previous years.

The IRS investigators who remain “are being pulled in a lot of directions,” some of which are unrelated to taxes, a former Justice Department official who witnessed the changes said. “The damage being done is significant.”

In Washington, the new responsibilities for IRS investigators have included conducting patrols alongside city police officers as part of a show of force Trump ordered this year to combat what he called a crisis of crime in the capital city.

The IRS’ office in Washington initially sent only a few of its roughly 60 agents to assist with those patrols. But after Trump aide Stephen Miller complained, the office upped that amount to more than 20 agents to patrol the streets, two people familiar with the deployment said.

Good grief is all I can say to 20 IRS agents assigned to patrol the streets of Washington DC. These people are not serious about running a government. The pattern squares with multiple inexplicable pardons of white collar and tax criminals, as well as a president who views aggressive tax avoidance (to the edge of evasion) as smart.

Need to assure donors? Just ask.

Another data point on how the IRS is truly a different agency than the one I was familiar with during my working career: this CBS story, Treasury Dept. tells Erika Kirk Turning Point USA not under investigation, following social media rumors (12/9/2025):

The Treasury Department sent a letter last week to conservative influencer Erika Kirk with findings that contradict fraud allegations about the finances at Turning Point USA and could help her refute those claims, sources told CBS News.

Questions were being raised on social media about the finances at Kirk’s organization, Turning Point USA, and podcaster Candace Owens and others were urging donors to demand refunds. That led a few of its small-dollar donors to ask for their money back, one of the sources said. Erika Kirk runs the nationwide conservative college student organization co-founded by her late husband, Charlie Kirk.

The letter said none of the four tax-exempt entities Kirk now runs — Turning Point USA Inc., Turning Point Action Inc., Turning Point Endowment Inc., or America’s Turning Point Inc. — are being examined by or are under investigation by the IRS, and all of the entities “submitted on time” all their 990 forms to the IRS this year. Contents of the letter were shared with CBS News.

Asked about the swift response to aid the conservative organization’s efforts to combat the social media rumors, a senior Treasury official said: “The IRS is able to provide this type of information upon request by the taxpayer. And in this case, it’s hideous that malicious lies and smears obligated her to make the request.”

That’s a service I had never heard of the Service offering.

Categories
tax administration

IRS Budget Detail Out

Last week the Trump administration released more detail on its 2026 budget proposal. This document includes its plans for the IRS – 18 pages of narrative and tables. The narrative is almost exclusively boilerplate descriptions of the various functions, authorizing laws, etc. In short, you’re left making inferences from the numbers in the tables, which include the proposals for authorized funding and the estimated number of employees for big functional categories.

Two high-level takeaways are (1) taxpayer service funding appears to be largely unscathed, essentially restoring the DOGE cuts, and (2) business systems modernization (i.e., the Inflation Reduction Act’s additional money to improve the IT systems) is zeroed out.

On the latter point, Janet Holtzblatt at TPC has a post, What Are The Trump Administration’s Intentions For The IRS’s Modernization? She points out the challenges with the administration’s narrative that it will offset budget and personnel cuts with technology improvements, AI, and increased efficiency. She asks:

Can any administration modernize the IRS with less staff and less funding?

The obvious answer to this presumably rhetorical question is no.1

I assembled a table of the budget’s estimates (pp. 857-74) of the changes in IRS employment by functional category from the actual 2024 numbers to the estimates for FY 2026. See below. Note that the 2024 numbers are likely below the peak employment under the Biden administration. The 2024 numbers do not reflect employees hired in the first four months of FY 2025 (October through mid-January 2025), many/most of whom were likely laid off as provisional employees by DOGE. This may be why the total employment I tallied from the budget document (92k) is below the higher number (100k) reported by the media for IRS employment in January.

The change columns reflect the 2026 estimates compared to 2024 actual.

Category20242025-est2026-estchange % change
Taxpayer service37,99638,849   36,694 (1,302)-3.4%
Enforcement38,48832,07831,699 (6,789)-17.6%
Op Support   13,11813,6688,050(5,068)-38.6%
Modernization6387630     638)-100%
Energy security764570    (764)-100%
Private collection823 1,285  1,285      46256.1%
TOTAL 91,827  86,700 77,728(14,099)-15.4%
Number of IRS employees; Source: Administration Budget

A few observations about these numbers:

  • The protection of taxpayer service, compared to the overall cuts, jumps out. This is a savvy political move. Cutting direct service provided by the federal agency that the public most often deals with would be a foolish political move. For example, creating major problems with 2026 filing season (e.g., delays in refunds) could be political catastrophe before the 2026 midterms.
  • Overall agency employment, per the 2026 estimates, will drop less than some of the media reporting has suggested (e.g., 15% per the budget’s estimates versus more than 20% in some reports).
  • Holtzblatt’s post reports that 20% of the IRS workforce applied for buyouts in April. If accurate (or even close to accurate), that implies there will be a large number of new hires to replace them. (The budget estimate is only a 15% reduction, and the workforce is already down by 5% to 10%.) That means a fair number of experienced employees will be replaced by newbies with the expected reduction in efficiency and quality of service for a time, anyway. Table 3 of the Tax Notes article I reference in the next section calculates (from IRS Data Book numbers) that over 21,000 IRS employees left the IRS in 2022 and 2023. That’s over 20% of the agency’s employees. So, turnover is high under normal circumstances, even without the stress created by Trump/DOGE.
  • Holtzblatt notes, citing other budget documents, that $2 billion in IT cuts would come from “eliminating, renegotiating, and descoping wasteful IT and professional services contracts.” This does not square with the fact that that is about the total amount spent on IT in prior years. Something does not add up. It’s worth noting that there is another potential source of that funding by modifying the purpose of still unrescinded IRA funding. My guess is that funding will not be around for very long under this Congress.
  • An interesting oddity is the 50% increase in employees2 in the Private Collection Agent Program. This is a niche program that the Service uses to collect debts (not for tax compliance). I’m sure there’s a back story. I have no idea what it might be.

Two of my takeaways are (1) basic taxpayer service, processing, answering phones, responding to simple inquires, paying refunds and so forth, is likely to continue as usual in 2026 filing season and (2) enforcement efforts with resulting impacts on compliance will degrade over time. This will inevitably result in long term declines in federal and state revenues (for states with income and/or corporate income taxes). Those long-term effects will probably not be immediately obvious but will be real and concerning.

Useful historical context

Tax Notes, Forced Attrition Will Make Tax Evasion Great Again (4/7/2025) [no paywall] is well worth reading. In particular, see Tables 1 and 2, which compile the change in IRS staffing levels from 1992 through 2023 and compares it to changes in the US population. Table 1 shows that total IRS personnel in 2023 was about 30% lower than in 1992 (the number of revenue agents and officers were more than 50% lower!). Table 2 shows that over the same period when IRS personnel declined by 30%, the US population increased by 38%, which I have copied below.

Schedule of IRS Staffing Consistent with U.S. Population Growth 1992-2023
 Total IRS Staffing LevelRevenue OfficersRevenue AgentsSpecial Agents
Fiscal 1992117,9459,70415,9472,943
U.S. population growth factor (1992 to 2023)38.9%38.9%38.9%38.9%
Projected increase in IRS staffing based on U.S. population growth45,8813,7756,2031,145
Projected IRS staffing levels consistent with U.S. population growth163,82613,47922,1504,088
 
Actual staffing levels in 202382,9902,6987,8532,067
Difference between projected and actual staffing levels80,83610,78114,2972,021
Percentage of actual personnel compared with projection50.7%20%35.5%50.6%

Source: Table 2, David P. Weber, Robert A. Warren, Timothy J. Fogarty, Vilson Dushi, Forced Attrition Will Make Tax Evasion Great Again (Tax Notes, April 7, 2025) (notes omitted).

Update on IRS layoffs

This from 6/6/2025 WaPo article, Trump administration races to fix a big mistake: DOGE fired too many people:

Across the government, the Trump administration is scrambling to rehire many federal employees dismissed under DOGE’s staff-slashing initiatives after wiping out entire offices, in some cases imperiling key services such as weather forecasting and the drug approval process.

Trump officials are trying to recover not only people who were fired, but also thousands of experienced senior staffers who are opting for a voluntary exit as the administration rolls out a second resignation offer. Thousands more staff are returning in fits and starts as a conflicting patchwork of court decisions overturn some of Trump’s large-scale firings, especially his Valentine’s Day dismissal of all probationary workers, those with one or two years of government service and fewer job protections.

….

Some fired federal employees, especially those at retirement age or who have since secured jobs in the private sector, are proving reluctant to return. So the administration is seeking work-arounds and stopgaps, including asking remaining staff to serve in new roles, work overtime or volunteer to fill vacancies, according to interviews with 18 federal workers across eight agencies and messages obtained by The Washington Post. A Post review found recent messy re-hirings at agencies including the Food and Drug Administration, the IRS, the State Department and the Department of Housing and Urban Development. In some cases, the government is posting new online job listings very similar to positions it recently vacated, a Post review of USAJobs found.

At the IRS, managers received a notice on May 19, a Monday, that all probationary workers would be coming back to the office on Friday, according to a copy obtained by The Post. The turnaround was so swift that some probationary staff probably wouldn’t have a desk or a laptop initially, the announcement acknowledged: “If a seat assignment is not available … your employees should begin teleworking until local management secures a seat assignment for them.”

None of this should be a surprise, given the shoot-first approach they have taken and that Trump 2.0 no longer has competent, normie Republicans modulating decision making. The most rudimentary competence would have prevented the worst of this. One wonders how Musk’s businesses have survived. Running them must be much less complex than conventional wisdom believes.

Notes
  1. This difference still seems too big. I’d speculate that some IRS employees are not reported in the budget document’s functional categories. ↩︎
  2. This will not be the case for inquiries and services related to more complex problems. For example, the Taxpayer Advocate Service, which often helps resolve more complex and intractable problems for ordinary taxpayers is in for major cuts ↩︎
Categories
income tax tax administration

Random tax stories

This post is a compilation of some recent minor federal tax developments, along with my reactions to them. Most should hit delete now; I’m just archiving links to and my thoughts on these stories for my future reference:

  • Pro-publica leaker pleads guilty
  • New tax gap estimates
  • Hunter Biden’s taxes
  • IRS picks states for free file pilot program
  • IRS delays pay-fors
  • Paul Ryan’s presentation at Brookings

Leaker caught

It took over two years, but the source of the illegal disclosure of tax return information, leaked to the NY Times and ProPublica has been caught (WaPo 9/29 story) and plead guilty to a felony count with a maximum 5-year prison term (WaPo 10/12 story). He is a contractor who worked for the IRS. The stories say nothing on this front, but the plea may be a prelude to his providing prosecutors with help in charging and convicting others. It’s hard to imagine one person pulling this off alone and it’s unclear why a contractor would have such extensive access to the universe of return data for multiple years. If he was writing and testing software, for example, access to a sample of returns would seem adequate.

It’s not clear what he was doing as a contractor. Tax Notes (no paywall) reports:

Prosecutors said the returns were accessed from an IRS database after the contractor used “broad search parameters designed to conceal the true purpose of his queries” and was able to evade IRS protocols for preventing or detecting the downloads.

IRS Contractor Pleads Guilty to Leaking Tax Data to News Outlets (10/12/2023).

According to WaPo, the scope of the leaks was stunning to congressional and IRS officials. Breaches of confidential tax data historically have been rare. The leaks obviously were politically motivated either to damage Trump or to create a climate favorable to increasing taxes on high-income filers or both.

According to CNN, because there were so many victims:

[P]rosecutors have asked Reyes [the judge] for permission to create a public website to notify the victims without reaching out to each person individually.

Former IRS contractor who allegedly leaked Trump’s tax information set to plead guilty, CNN (10/6/2023)

Here’s a link to the DOJ website. At least one hedge fund magnate has sued the IRS for damages. He may be embarrassed that his peers now know he paid a 29% effective rate, much higher than most of them, all of whom benefit from the carried interest tax break (taxing their compensation at long-term capital gains rates). I assume he has a tall legal hill to climb to prevail, given the rules of governmental immunity.

Victim statements are typically given at sentencings. Trump couldn’t wait and sent an attorney to the hearing who called for imposing the maximum 5-year sentence. The judge’s response, as reported in the 10/12 WaPo story, was that the former president could come and deliver his own statement at the sentencing in January. That seems a little caddy, since victim statements are typically received in manners that accommodate victims. The DOJ website says you can email in your victim impact statement. Objectively, it’s unclear how much Trump has cause to complain, since Ways & Means Democrats legally released his returns to the public later on. So, all the defendant’s conduct did with respect to Trump was to speed up public access to his returns, it would seem to me anyway. Of course, there is the matter that the 2020 election happened in between the two releases.

From what I can tell, the disclosure has not affected congressional actions on tax policy, contrary to what ProPublica expected or hoped for. Rather than sparking outrage (unlike somewhat similar but legal disclosures in the 1960s), it likely confirmed what the public had assumed. Many top earners pay low tax rates and in some years little or no tax. The fact that Trump essentially bragged about paying very little tax during the 2016 presidential debate and suffered no apparent ill political consequences for doing so is another data point supporting that view. The risk, as I always feared, is further undercutting confidence in the IRS and in the security of tax data.

As an aside, WaPo quotes the NY Times reaction as follows:

Charlie Stadtlander, a Times spokesman, said in an email Thursday night: “We remain concerned when whistleblowers who provide information in the public interest are prosecuted. The Times’s reporting on this topic played an important role in helping the public understand the financial ties and tax strategies of a sitting president — information that has long been seen as central to the knowledge that voters should have about the leader of our government and the candidates for that high office.”

Salvador Rizzo, Former IRS contractor pleads guilty to leaking Trump’s tax returns, 10/12/2023 Washington Post

Referring to this guy as a whistleblower is a mischaracterization. He was not revealing wrongdoing, illegal misconduct, corruption or similar but rather releasing protected, confidential tax data to serve his own personal or political interests. That sort of justification demeans legitimate whistleblowers and leads to a narrative that journalists and media organizations just want to publish stuff that serves their agendas, sells newspapers, or attracts eyeballs. It may be that presidents’ and presidential candidates’ tax returns should be released to the public. If so, it needs to be done by enacting a law that requires it.

New tax gap numbers

On October 12th, the IRS released new tax gap numbers (summary and full report) that showed “a significant jump from previous estimates.” According to the release:

The tax year 2020 and 2021 tax gap projections translate to about 85% of taxes paid voluntarily and on time, which is in line with recent levels. After IRS compliance efforts are factored in, the projected share of taxes eventually paid is 86.3% for tax year 2021, down slightly from the 87.0% for tax years 2014-2016. This drop in compliance does not factor in any changes in compliance behavior; instead, it is due to changes in the types of income and how that income is reported to the IRS.

IRS, Tax Gap Estimates for Tax Years 2014-2016 (10/12/2023)

These are the first estimates of annual amounts and IRS also announced it will begin releasing yearly estimates. Both seem to me to be good moves to focus more attention on the gap and to provide more detail.

The estimates are based on audits from 2014-16 and, thus, should reflect the first effects of congressional starvation of the IRS’s budget that began in FY 2012. The tax gap is growing faster than the tax base or the economy. Since the estimates assume constant compliance rates, the growing gap estimates are likely caused by changes in the composition of income – e.g., income with lower compliance rates (e.g., more business income relative to wages) – and reduced auditing and enforcement.

Table 1 in the full report (p. 10) shows that the Gross Tax Gap grew by 38.7%, while Enforced and Late Payments slightly declined (-0.1%). The gross tax gap is used to compute the Voluntary Compliance Rate (i.e., what taxpayers report on their own), while the Enforced and Late Payments reflect IRS enforcement efforts (amounts attributed to audits, examinations, etc.). I don’t know how the estimated growth in enforcement are made, so it is hard to know how much is due to reduced IRS resources, rather than more income with lower compliance rates even after IRS enforcement efforts. (This is how the report cryptically describes the methodology: “Estimate for a given type of tax & tax year is the sum of late payments to date plus a projection of future late payments based on payment patterns observed for earlier tax years” Table 8, p. 20.) In any case, the growth in the net tax gap (i.e., the headline gap everyone focuses on) is attributable to the enforcement amount not growing at pace with the tax base. That follows necessarily (it seems to me anyway) from the assumption of constant compliance rates. One inclined to be suspicious of the IRS would suspect a thumb on scale might be involved.

Tax compliance funding fight from yesteryear

On a tangential but related matter, Joe Thorndike, has a piece in Tax Notes, Tax History: Treasury Promised a 20-1 Return on Tax Enforcement, but Congress Slashed Funding Anyway (9/25/2023) (no paywall) that describes the political fight over BIR (the IRS predecessor) funding waged by the 1947 Republican Congress with the Truman administration. I was completely unaware of this history and found it interesting and revealing. Thorndike’s account even includes a quote from a Minneapolis Star editorial opposing the cut.

I’m sure he wrote the piece and it is interesting because of the parallels to the current fight between the Biden administration and Congressional Republicans over IRS funding. Thorndike ends the piece with the observation that “In Washington, some debates — even ones surrounded by a sea of relevant data — just go on forever.” He casts it as a debate over the return on money appropriated for tax enforcement, which is what it was. The current debate is nothing like that.

What struck me on reading the piece was how the substance of the 1947 debate, as portrayed by Thorndike, starkly contrasts with the current debate over increased IRS funding.

The Republicans had taken over Congress in the 1946 mid-term election. It was their first control post-New Deal and they were determined to cut federal spending. The BIR’s budget had greatly expanded during WWII when the income tax was expanded from applying only to the affluent to nearly everyone with any reasonable amount of income, a 6X increase in the number of returns. The agency’s budget had similarly grown. The House’s budget cut the agency’s $208 million request by $30 million.

The Truman administration responded by saying that would cause a $600 million reduction in revenue (the 20:1 ratio in Thorndike’s title). And that all the cuts would come from enforcement. The House’s response – diametrically opposite to the current debate over increased IRS funding – is described by Thorndike as follows:

GOP lawmakers resented the BIR’s effort to strong-arm the committee with catastrophic projections about lost revenue. And they instructed the BIR to leave enforcement funding intact. “There is a specific item in the report that the committee will not countenance a reduction in the enforcement personnel,” noted Rep. Everett Dirksen, R-Ill. “We want them to get out of Washington some of these people in the administrative offices who are falling all over themselves.”

Joe Thorndike, Tax History: Treasury Promised a 20-1 Return on Tax Enforcement, but Congress Slashed Funding Anyway Tax Notes (9/25/2023) [emphasis added].

Thus, they explicitly prohibited cutting auditing and enforcement to preserve revenue collections. Contrast that with the current GOP falsely claiming Biden and Democrats funded 87,000 new gun-toting agents who will appear at your business or home to make sure you pay up. Both parties are still on the same sides of the debate, but the substance has shifted seismically. The 1940s Republicans obviously thought preserving tax collections was crucial, making the return on investment relevant. I may have missed it, but I have detected no concerns on that front by the 2023 congressional Republicans. The 1947 debate appears sedate and rational compared to today’s, animated as it is by extreme claims and outright misinformation.

To finish the story, the House’s proposed cuts were about halved in a final compromise. But that was not before the administration sent out layoff notices to BIR employees and encouraged them to lobby Congress against the cuts and Truman decrying the cuts when he signed the bill.

Hunter Biden’s taxes

Disclaimer: this is largely about tax compliance and enforcement, not Hunter or his taxes.

Senator Wyden, chair of the Senate Finance Committee, sent a letter to IRS Commissioner Werfal (with a press release, of course) urging the Service to go after high-income non-filers. I assume the exercise had two purposes – (1) to counter the GOP drumbeat on Hunter Biden’s tax problems and (2) to shore up the case for IRS funding, while making the GOP look hypocritical (typically an easy lift). But the disclosures in the letter are eye-popping and somewhat (to me anyway) inexplicable.

The committee made a request to the IRS for information on high-income non-filers. They defined high-income non-filers as ones who reported $200,000 in AGI for one tax year between 2017 and 2021 – note that is half what the Biden Administration considers high-income under its goofy tax pledge – and had multiple years in which they did not file returns between 2015 and 2021. Wyden’s letter describes the response the Committee got as “identifying staggering levels of noncompliance and criminality by high-income non-filers.” That is hyperbolic, but the numbers did surprise me:

There were 981 high income non-filers in this population with AGI of at least $1 million. These taxpayers owed $34,277,321,852 in total unpaid assessments (tax, penalties, and interest) as of May 2023. There were 58 high income non-filers in this population with AGI of at least $10 million. These taxpayers owed $16,685,980,639 in total unpaid assessments (tax, penalties, and interest) as of May 2023.

Ron Wyden letter to Commissioner Werfel (9/28/2023), pp. 1 – 2, fn 2.

So, 981 filers reported $1 million or more in AGI in one year of the period and failed to file returns in two years! I would discount or ignore the dollar amounts owed, because they must reflect SFRs constructed by the IRS, plus interest and penalties, all of which often inflate real tax liability, potentially by a lot. But if the $1 million AGI number reflects self-reported income on filed tax returns (rather than something constructed by the IRS), it shocks me that almost a thousand of those taxpayers would skip filing two or more tax returns. Since failing to file does not trigger the statute of limitations, the IRS has unlimited time to come after you. If your reported income was that high, I can’t imagine a well-advised taxpayer (as most of these taxpayers must be) would think they could somehow avoid IRS scrutiny. Of course, in a few cases they may have insufficient income to generate a liability (typically because of business losses) and so there is no urgency to file. But you typically will still file, often to claim carryover losses and get a refund of prior year tax or to establish carryforward losses for the future returns.

It’s a mystery to me. Tax compliance and filing can’t really be this bad, can they? It does suggest that DOJ was not going easy on Hunter by allowing him to plead to a misdemeanor, and that he is more likely getting the book thrown at him now. I don’t feel much sympathy for him (beyond his obvious mental health issues), just to be clear.

In a recent press release, the IRS says it is on the case of pursuing these non-filers, as Wyden requests:

The IRS has been ramping up efforts to pursue high-income, high-wealth individuals who have either not filed their taxes or failed to pay recognized tax debt. These efforts are concentrated among taxpayers with more than $1 million in income and more than $250,000 in recognized tax debt. * * * As announced in September, the IRS has begun contacting about 1,600 new taxpayers in this category that owe hundreds of millions of dollars in taxes.

The IRS has now collected $122 million dollars in 100 of these already assigned 1,600 cases.

IR-2023-194, Oct. 20, 2023

Update 11/9/2023. Tax Notes has a story, Behind the Scenes of the IRS’s Crackdown on Wealthy Taxpayers (11/6/2023) (no paywall), that interviewed practitioners on reasons for high-income taxpayers not filing:

“It was definitely a mixed bag,” said Gerald W. Kelly of Kelly Dorsey PC, a former revenue officer. The most common excuse among these high-income taxpayers was simply that they were living above their means, he said, explaining that while they might have earned hundreds of thousands of dollars in income, their lifestyle expenses had increased commensurately, and they ended up spending all of it to maintain that.

Another common scenario is a taxpayer who doesn’t file or pay one year. They may think what they’ve done is so bad that if they file the next year, it’s going to bring that mistake to the IRS’s attention, Kelly said. Then they assume the worst: that the IRS will send them to prison or seize all their assets if it finds out, so they bury their head in the sand and stop complying altogether, hoping the IRS won’t notice.

High-income taxpayers also might not file returns or pay their taxes because of a major life event like an illness, according to Robert A. McKenzie of Saul Ewing LLP. Then, instead of filing and paying, they may opt to prioritize maintaining their lifestyle, said McKenzie, another former revenue officer.

Another example is a taxpayer in the midst of a divorce, which can delay the tax preparation process because each spouse needs records from the other to complete their respective returns, McKenzie said. “Trying to get a joint return filed by a divorcing couple is not always the easiest thing in the world when they’re busy trying to — at least psychologically — kill each other,” he noted.

In still other cases, taxpayers with a lot of money just make bad financial decisions, McKenzie continued. “I’ve had clients fail to file because they knew they were facing a $400,000 liability, but they saw this opportunity to invest in something that they put their money in instead,” he said. “There are some very bad life choices by some of these people.”

Ultimately, all of these high-income taxpayers know that they’ve messed up. “I can’t think of one case I had where the taxpayer wasn’t aware of their obligation to file and pay every year,” Kelly said.

Ibid.

Collecting from them may not be as easy as one might think.

As an aside on Hunter Biden, WaPo has the backstory on the ups and downs of the prosecution, the whistleblower IRS agent, and more. The story has a lot of interesting details, including this curiosity as to how the tax case against him started (why is sex always involved?):

The investigation into Hunter Biden grew out of an international tax probe of an amateur online pornography company in Britain suspected of failing to pay taxes on payments to contractors, according to internal IRS documents shared with Congress.

One of those people was a suspected escort who bank records showed also received payments from Hunter Biden, according to case file documents. A review of Biden’s tax history showed he had failed to file federal tax returns for several years, prompting agents to open a separate investigation into his conduct.

Devlin Barrett and Jacqueline Alemany, Backroom battles between IRS agents, prosecutors in Hunter Biden tax case, Washington Post (10/3/2023).

The story appears to be a classic case of conflicting strong personalities, DOJ Tax Division bureaucratic machinations, and probably political considerations. A lethal combination in the current environment.

Direct file pilot states announced

The IRA legislation funded an IRS-provided free filing option. On October 17th, IRS announced that Arizona, California, Massachusetts and New York have been chosen as pilot states, meaning they will integrate their state taxes into the IRS-provided platform for 2024 filing season. Taxpayers in states without state income taxes may also be allowed to participate. States without their own government-provided filing option (e.g., Minnesota) were not eligible.

Assuming the pilot goes forward (not always a safe assumption), participation will be limited to:

[T]axpayers with certain types of income, credits and deductions – taxpayers with relatively simple returns. The IRS today announced it anticipates specific income types, such as wages on a Form W-2, and important tax credits, like the Earned Income Tax Credit and the Child Tax Credit, will be covered by the Direct File pilot.

IRS advances innovative Direct File project for 2024 tax season; free IRS-run pilot option projected to be available for eligible taxpayers in 13 states (10/17/2023)

The release has more detail about who’s eligible and the provisions covered. Here’s how it describes the option and how state taxes will be handled:

Direct File will be a mobile-friendly, interview-based service that will work as well on a mobile phone as it does on a laptop, tablet or desktop computer. The service will be available in English and Spanish for the pilot.

The Direct File pilot will be a limited, phased pilot. It will not be available to all eligible taxpayers when the IRS begins accepting tax returns. Because the IRS wants to make sure the program works effectively, Direct File will first be introduced to a small group of eligible taxpayers in filing season 2024. As the filing season progresses, more and more eligible taxpayers will be able to access the service to file their 2023 tax returns.

***

Direct File will cover only individual federal tax returns during the pilot. Also, Direct File will not prepare state returns. However, once a federal return is completed and filed, Direct File will guide taxpayers who want to file a state return to a state-supported tool that taxpayers can use to prepare and file a stand-alone state tax return. For the pilot in 2024, where taxpayers may have state or local tax obligations, the IRS will limit eligibility to states that are actively partnering with the IRS on the pilot.

Ibid.

IRS employees will provide, according to the release “technical support and provide basic clarification of tax law related to the tax scope of Direct File.”

I have always thought something like this is a good idea. Expecting people to prepare and file their own returns may have been reasonable in the past. But the complexity layered on the tax over the years and the reality that most already use software makes requiring taxpayers to shoulder that burden (i.e., to buy a commercial product to file their taxes) harder to justify. The Free File program has been revealed as essentially a sham or close to it with declining participation by commercial software providers. The two biggies, Intuit and HR Block, have withdrawn from it.

Will be interesting to see how this develops and what effect it has several fronts – e.g., inducing states to provide their own software and support, increased electronic filing (probably not much), and sales of commercial software (industry obviously thinks it will hurt them). I will stay tuned.

IRS delays pay-fors

The IRS has announced delays in information reporting for payments through payment platforms and in requiring higher income employees to make their “catchup” retirement contributions to Roth style plans. In both cases, this slightly undercuts intended revenue offsets to tax cuts or increased spending. I hope it does not permanently derail them, which is always the risk. There could be a tiny impact on state revenues for states with conforming income taxes.

When Congress passes tax cuts and new spending, it often includes tax provisions that are intended to offset some or all the spending or revenue loss. I think this is still required by budget rules, even though budget discipline in Congress evaporated long ago. Congressional jargon for them is a “pay-for” as in they are intended to pay for the tax reductions or new spending.

The tendency is to focus them on below-the-radar, narrow provisions that politicians perceive are politically acceptable or may go unnoticed. But that can mean that they are difficult to implement and cynically may not succeed in raising the full amount of their estimated revenue. In these two cases, they caused the IRS, which now is adequately funded, to throw up its hands and delaying implementation. Technically, the IRS probably does not have legal authority to do that, but nobody has legal standing to challenge the decision.

1099-K threshold

Back in December, the IRS announced that the much lower thresholds for 1099-K reporting that Congress enacted in 2021 would be delayed by one year until tax year 2023 (Notice 2023-10). This was a pay-for in the Democrats’ pandemic relief bill, the American Rescue Plan. The Joint Committee estimate (VIII.4) projects $8.4 billion in added collections from the item over the 10-year budget window. Delaying its effect by one year probably won’t equal exactly one-tenth, but close.

The expanded information reporting for payment platforms (technically Third-Party Settlement Organizations or TPSOs – e.g., Venmo, PayPal, and similar) was a dramatic increase – from 200 transactions aggregating at least $20,000 to trigger a Form 1099-K under prior law to a simple $600/year. The change was intended to increase compliance by businesses receiving these revenues, including gig workers like Uber Drivers, Airbnb hosts, and similar.

The prospect of ARP’s change generated a fair amount of media attention (almost hysteria in some quarters). The concern was that Venmo and Zelle payments where friends split restaurant checks, concert tickets and similar would generate 1099-Ks and tax compliance nightmares. (When that happens, the IRS advises attempting to get 1099-K corrected and if you can’t to make offsetting entries on the 1040. FAQ #8.) It’s unclear how the platforms will be able to sort out when payments are gifts, cost sharing, and similar rather than payments for selling goods and services. I could not find any IRS guidance on how they can do that (e.g., by allowing self-reporting by their users). These payments are likely a trivial share of the total payments covered, but an obvious source of friction.

The risk, of course, is that the delay will result in full repeal (something already being proposed in Congress by the usual suspects), rather than using it as an opportunity to implement a better system that generates fewer reporting errors.

In a classic case of having it both ways, GOP Senator Mike Crapo (R-Idaho), the ranking minority Finance Committee member, criticized the IRS for not following the statute in this case (of course the GOP opposes the provision altogether). According to Politico:

“One of the troubling concerns that I’ve seen is a recent trend that, from my perspective, is a willingness of the IRS and Treasury to ignore the plain language in enacted statutes when issuing regulatory guidance.”

“The IRS has simply disregarded statutory deadlines for implementing new Democrat-led provisions,” Crapo added.

To name but one example, a tax provision implemented by the American Rescue Plan that would have increased reporting for users of apps like Venmo, PayPal and Stubhub — and is unpopular among both Democrats and Republicans — was delayed for one year by the IRS in December 2022.

That law was supposed to go into effect for the 2023 tax season, but e-commerce and third-party marketplace companies lobbied heavily against it.

Benjamin Guggenheim, Why Treasury’s tax powers may be in GOP crosshairs, Politico (10/2/20223).
Roth catchup contributions

The Secure 2.0 Act, passed as part of the December 2022 budget deal, loosened taxation of retirement plans in a multiplicity of ways but most prominently by delaying the age at which RMDs must be claimed. (Yes, it also made some good changes that will help ordinary people struggling to fund adequate retirements through the savers match and auto enrollment changes.) With only a few exceptions RMD rules adversely affect those who have plenty saved for retirement and are funding inheritances, rather than retirements. There was no need to do that, especially the increase whose effective date was delayed until the first year outside the budget window.

One of the main pay-fors was to require more contributions to be made to Roth plans, including “catchup contributions” for employees with wages exceeding $145,000. That change was scheduled to go into effect next year, but the IRS delayed that effective date by two years as an “administrative transition period.” In the IRS’s words:

until taxable years beginning after December 31, 2025, (1) those catch-up contributions will be treated as satisfying the requirements * * * [to be made to Roth plans} even if the contributions are not designated as Roth contributions, and (2) a plan that does not provide for designated Roth contributions will be treated as satisfying the requirements * * *.

IRS Notice 2023-62, p. 6.

The Joint Committee revenue estimate does not separately break out this component, so I can’t tell how much revenue is at stake. It’s peanuts, for sure. The complexity of implementing a change like this is real – consider people with multiple jobs and how one employer should apply the dollar cap. So, I don’t doubt the need for more time.

Moreover, I have made the point before that the permanent revenue from switching to the Roth structure (aside from being bad policy IMO) is largely illusory. The Roth model requires tax to be paid up front (during the budget window) not during the retirement (often outside the budget window). The revenue loss appears to be made up only IF you look at the budget window and ignore the longrun. It is mainly being accelerated with little change on a permanent revenue basis. Of course, the time value of money – here interest payments on federal debt – is a real factor, especially now with the return of real interest costs of federal borrowing after accounting for the effects of inflation.

There is, however, a likely exception to that for these catchup contributions. These are higher contribution maximums that can only be made by older employees with high incomes (> $145k). They are often used in the years right before retirement to shift earnings into retirement when the retiree’s income and tax rate drop, a self-help income averaging provision. Requiring them to be made as Roth contributions thwarts that strategy and results in a real increase in revenue. The amounts are small, but the effect is likely heavily skewed to moderately high-income taxpayers (often below Biden’s definition, though).

Two observations about the IRS administratively undoing pay-fors
  • There is a disconnect between the precision with which the process attempts to balance these minor tax cuts fiscally, while ignoring practical administrative reality of the pay-fors. The bigger disconnect is that Congress is careful in balancing these small changes, while using reconciliation to pass stuff like TCJA that adds massive amounts to the debt, which is growing faster than the economy when there is no macroeconomic (Keynesian style) justification for doing so.
  • These temporary deferrals are trivial compared with the IRS’s questionable decision to allow PTEs to deduct entity-level state taxes. That blew a large hole in TCJA’s pay-fors. I have not seen an estimate, but it must be very large and will grow, especially if the SALT limit is made permanent without repealing or limiting it.

Paul Ryan speaks

Brookings and the Hamilton Project hosted an event that focused on the path forward when TCJA’s individual income tax cuts are scheduled to expire, Taking on tax: The past, present, and future (video replay, transcripts, etc.). Extending all the tax cuts will reduce revenues by something like $3 trillion over ten years, but the average voter obviously expects them to be extended and allowing them to expire would be cast politically as a tax increase (by Republicans and most of the general public), notwithstanding that is the what the law says. Official federal deficit and debt estimates assume they will expire. So, the debt estimates are unrealistically low as a practical and political matter. Biden’s budget calls for extending the tax cuts that apply to 98% of the population. The situation is a fiscal and political mess.

The first panel included Paul Ryan, the former Ways & Means chair and Speaker, who was a principal architect of TCJA and who (I assume) has a good a grip on what is politically feasible to pass, especially if Republican support is needed and probably even if it is not. Thus, I regard his comments as particularly interesting and maybe insightful for what the future holds. Here are a few excerpts (taken from the transcript rather than trusting my notes; the software used to create the transcript is really bad, so a lot of fixing was required, especially on acronyms but more broadly. For example, “corpse” ≠ the intended “corps” and similar. I did not note all of my fixes.):

Why TCJA’s corporate provisions were made permanent:

Kevin Brady and I decided that we would make permanent that which we thought for *** economic * * * and political reasons needed to be made permanent. The corporate rate, the territorial system. And we made temporary that which we thought had a better chance of withstanding extension under any conceivable political arrangement in the future.

Transcript p. 3.

My take: This was obvious to me: the corporate cuts were politically vulnerable, so they were made permanent. It’s a cynical use of a “temporary” tax cut to de facto put federal revenues and those who want to fund the federal government services at the current level in a vise. The corporate tax cuts would be unpopular (i.e., average Joe and Jane voters would be indifferent to them), so we made them permanent. The individual tax cuts would be hard for the Democrats to allow to expire because we’ll argue they’re raising your taxes – never mind that we enacted the law that causes them to expire. That’s clever and cynical and he is not ashamed to admit to it. I give him credit for honesty.

It also gives the lie to Ryan’s oft-expressed concern about the deficit or at least subordinates it to other parts of his agenda. Limiting the size of government is a higher priority than paying for the government we’re delivering. I’m old fashioned but that’s not conservative or transparent and uber cynical. But Ryan obviously believes in the hypothesis that low tax rates = growth. So, I’m sure that’s his rationalization. Probably because he spends too much time hobnobbing with GOP donors and starts to believe their cocktail chatter about tax rates holding back growth. If he spent more time with Warren Buffett and less with Rupert Murdoch and similar, his perspective might be different.

Regarding Moore, the SCOTUS case challenging the constitutionality of the repatriation tax as a mechanism to abort a wealth tax in utero:

I think [the Moore plaintiffs and their conservative backers are] misfiring on this one. I mean, a lot of the tax code would be unconstitutional if that thing prevailed. * * * So I think it’s a misguided challenge, in my opinion. * * * I’m not for a wealth tax, but I think if you use this as the argument [? it’s a constitutional prohibition, not an argument] to spike a wealth tax, you’re going to basically get rid of, I don’t know, a third of the tax code.

Transcript, p. 4.

My take: One-third of the tax code is hyperbole. I’m sure if the Court rules for the taxpayers, it will attempt to craft a narrow ruling that limits the collateral damage to both TCJA’s revenue offset and the Code, while preemptively killing wealth or other odious taxes according to their priors. But even a narrow ruling for the taxpayers will be bad news. That this case is on the docket at all is a consequence of the Republican selection process for SCOTUS nominees and their agenda to quickly rewrite constitutional law to align with their political philosophy.

There are echoes of Pollock, where the Court reached to rewrite constitutional law to be consistent with its laissez faire economic philosophy. The harm was temporary because in those days it was still practical to amend the constitution. See Professor Ackerman’s account (pp. 28 – 33) for the details (entire article is good).

In any case, the genie is out of the bottle and in SCOTUS’s hands and may become some sort of bogus history lesson on the meaning of “income” in the early 20th century America lexicon. We have got to hope that two of Roberts, Kavanaugh, and Coney-Barrett recognize the gravity of the matter and stay with the conventional wisdom that was taught in law school when they attended. I’m not feeling good about this but I’m a Cassandra.

Will QBI (20% deduction for business income) be extended:

199A [QBI’s code section] we’re going * * * to bring [the tax rate that] pass through [entities pay] up to, what, 44.6% when the C corps are staying at 21%? I don’t think that’s going to happen.

Transcript, p. 5.

My take: Ugh, this confirms what I had assumed. It is very bad news, since this is one of the worst features of TCJA. As various academics have pointed out, it’s complex, distortive, and violates both horizontal and vertical equity. It hard to find provisions that simultaneously violate the simplicity, efficiency, and equity principles. Ryan’s rate comparison is misleading as commentators have pointed out. It’s nominally for small businesses, making it highly attractive politically. (Disclosure: I personally benefit because of our investment holdings, which are far from being in “small” businesses.)

SALT deduction:

I would get rid of all of the SALT [deduction] now. I couldn’t. We kept [the threshold] at $10,000 because that’s where the votes were. We wanted to [eliminate the deduction]. It’s $10,000; that’s going away. That pays for a lot. I would do the KTC [likely child tax credit] with that personally[.]

Transcript, pp. 5-6.

My take: I think Ryan is saying a TCJA extension will restore the old SALT deduction. At least, I assume it’s the threshold that is going away, not the deduction. If he is right about that, it’s bad news for the deficit, since it would increase the revenue loss from extending the TCJA individual cuts as is. As Ryan notes, limiting the SALT deduction was one of TCJA’s biggest pay-fors.

The SALT deduction has been transformed into the standard GOP offset for proposed tax cuts, especially for tax cuts that the Democrats like such as the child tax credit. Mitt Romney and others have introduced bills to do that. Of course, media reports said Jim Jordan was willing to restore more deductibility (much higher thresholds) to get votes in his race to be Speaker. Politics is truly weird and torques policy.

How to fix the long-run federal fiscal problems?

Ryan thinks that the GOP’s new antipathy for big corporate America (e.g., Disney and other “woke” corporations) will not extend to tax policy (i.e., raising the corporate tax). So, a corporate tax increase is a non-starter if GOP support is required, in his view.

In terms of finding a fix for the long-run debt crisis, Ryan is still looking for a grand bargain:

I think tax reform is the way out of this. I can tell you the canyon we’re going into here in this conversation, the way out of this box canyon is tax reform. * * * I think you can get a higher revenue line [increase taxes] without sacrificing economic growth, but you’re not going to get the kind of distribution tables [progressivity] that I think this building [the liberals at Brookings] wants. And if you can drop it [concerns about regressivity], you drop that as a priority and focus on growth and revenue. I think you can get a decent tax reform system that that you could probably get some conservative buy in if and when it is brought together with entitlement reforms [benefit cuts to Social Security, Medicare, and Medicaid]. If there is ever going to be a grand bargain, it’s when we’re running into this debt crisis.

Transcript, pp. 8-9.

My take: This makes my head spin.

First, he’s living in a past that didn’t exist. The GOP Congress he was part of (i.e., in 2013) rejected a somewhat similar grand bargain with Obama. Okay, the Democrats in the Senate screwed things up a bit too, but Tea Partiers in the House were going to kill it no matter what, according to Boehner’s book. It’s hard for me to imagine the current GOP House (Jim Jordan, MTG, and the whole lot) even talking about something like this. No chance they’d ever agree to a tax increase.

Ryan is the eternal optimist. He’s convinced that the force of events can change the political dynamic. To me it seems like everything is going in the opposite direction. The debt growing faster than the economy is a silent and slow killer (like chronic high blood pressure or diabetes). My observation is that Congress is impelled to action only by obvious and immediate crises that the public gets. That’s not the debt. Constructed crises (e.g., the legal requirement to cut SS benefits when the trust fund can no longer full pay benefits) can easily be turned off shifting payment to the general fund, unless both parties agree that is unacceptable (as in the early 1980s). I doubt they will. Ryan obviously thinks otherwise. I guess successful politicians (retired or not) need to be eternal optimists.

Second, his outline appears premised on a view that the Democrats highest priority is fiscal responsibility and would give Republicans their preferred policies (more regressive taxes and lower entitlement benefits) to get it. That presumptively fails. Why would Dems ever agree to that? So much for the supposed conservative party being a champion of fiscal responsibility. Instead, their negotiating position is to use the other side’s supposed concern about fiscal probity to extract policy concessions.

Third, he’s transparent about his willingness to increase regressivity. I get that he thinks that will stimulate growth by allowing lower tax rates on income, given his world view. The empirical support for that is next to nonexistent. His core Randian philosophy shines through (his professed concerns for the poor based on Catholic values, not so much). His preference for doing that is a border-adjusted carbon tax. (The border adjustment allows carbon emitting American businesses to compete in foreign markets without the drag of a US tax. That complicates administration and punts responsibility for the emissions to the nation in which the consumption occurs.) He also has not let go of TCJA’s Destination Based Cash Flow Tax, despite its swift death in 2017 because of business opposition. The US needs a broad-based consumption tax to fix the debt/deficit problems. Substituting the DBCFT for the corporate tax would have made doing that harder.

Bottom line, if something really needs to be done sooner rather than later about the debt, things don’t look good. Ryan is (relative to the Republicans in Congress now) thoughtful and reasonable. Truly depressing for anyone who thinks we need to align long-run revenues more with entitlement spending commitments.

Categories
Uncategorized

A Trillion Dollar Free Lunch?

Larry Summers (former Secretary of the Treasury, former President of Harvard, and blue-chip economist) and Natasha Sarin (professor at Wharton and the U Penn Law School) are out with an article on how to raise a trillion dollars in new federal revenues without increasing taxes.  Hint: it involves closing the tax gap (collecting taxes now legally owed but not paid) by increasing IRS funding and implementing other compliance efforts. 

Summers is a centrist Democrat who has been vocal in his concerns about proposals by Democratic presidential candidates for wealth taxes and big income tax rate increases on top earners, so this can be viewed as a partial counter to those proposals.

Based on Summers’ and Sarin’s analysis, if Congress appropriated an additional $100 billion to the IRS for compliance, federal tax revenue would rise by a net of $1 trillion over the next decade. This would occur without the political pain of increasing taxes, e.g., by raising rates, curtailing deductions, etc.  This may sound too good to be true and it probably is.  Their estimates are higher than government estimates (by CBO, e.g.), some of their policy proposals (more information reporting) will run into strong lobbying in Congress if history is any guide, and they themselves characterize the estimates as “naïve.”  But there is a lot of money to be collected by returning to more normal IRS funding levels and enacting reasonable changes in information reporting and/or expanding withholding. That doesn’t mean any of that will be politically easy, though.

The article is well worth reading, either in the full but still relatively short Tax Notes version (unlike most Tax Notes content, it is not gated) or their WaPo op-ed, which covers the main points. (Since I wrote this, the Strib editorial page has also published an op-ed by them.)

Some points I gleaned from Summers’ and Sarin’s article that I think are worth noting (along with a few of my reactions) include:

  • The IRS estimate of the tax gap is about 15% of collections or $7.5 trillion over the next 10 years. Summers and Sarin think reasonable efforts could shrink that by 15% or $1.1 T.
  • Funding of IRS compliance has declined significantly – its overall budget has dropped by 15% in real terms since fiscal year 2011. The decline in enforcement spending has been even larger (audits have declined by 45%). The number of criminal tax cases filed dropped by 25%. Me: I doubt the number tax crimes dropped by 25% – if anything declining enforcement probably results in more tax crimes.
  • Summers and Sarin don’t say this, but the big decline in funding occurred exactly when Congress saddled the IRS with big new responsibilities – e.g., to administer much of the Affordable Care Act and to implement the complicated provisions of the 2017 tax cut. Those responsibilities undoubtedly required the IRS to siphon resources away from compliance.
  • I had forgotten about the changes in congressional scoring rules that Summers and Sarin point out likely reduced Congress’s incentive to appropriate money to the IRS. These scoring rule changes prevent Congress from using the estimated revenues from increased IRS appropriations to offset spending or tax cuts.  Congress routinely did that as “pay for’s” – i.e., to satisfy budget rule requirements that bills be balanced.  That practice largely came to be regarded as phony, which led to its elimination.  But, according to Summers and Sarin, an unintended effect may be underfunding of IRS compliance.
  • I personally think that Congressional Republicans’ response to charges that the IRS illegally or unfairly targeted Tea Party groups seeking exempt organizations determination letters is a bigger factor in the reduction in IRS appropriations. Whatever you think about the validity of the Republicans’ concerns, underfunding IRS has had bad consequences for tax compliance, unrelated to exempt organization issues.
  • Summers and Sarin have three basic recommendations – (1) increase funding of the IRS compliance and target more of that to high income filers and corporations where the return is proportionately higher, (2) increase information reporting – as academics their recommendations are so general to be pretty much meaningless from a practical tax administration point of view, and (3) make substantial IT investments.
  • Their estimates for the yield on increased spending on exams or audits, which they characterize as “back of the envelop” and “naïve,” do not assume much if anything in the way of diminishing returns. They justify that because compliance revenue has declined proportionately to the decline in IRS spending on those activities. I’m skeptical. Just because the IRS was unable to offset its loss of funding by dropping its least productive audits does not (to me) mean that the IRS will be able to deploy additional spending on audits (above its old baseline) as productively as it is using current resources. My instinct is the exact opposite.  I would guess Summers and Sarin’s estimates are on the high side – particularly for increases that substantially exceed prior IRS funding levels, which Summers and Sarin assume.
  • They point out that expanded compliance funding – particularly in the way they suggest – will be progressive. The revenues will disproportionately come from higher income individuals. That occurs because the types of income with poorer compliance rates (e.g., capital gains and various forms of business income) are concentrated at the top.  Using distributional data on income by source, the article has tables that show how dramatic this effect is.  For example, the net misreporting percentage for individuals with less the $200k in income is under 3%, while it is nearly 14% for those with incomes of $10m or more.  And, of course, if the IRS focuses the new compliance money where the highest return is, that will also disproportionately affect the higher income filers. In their words: “In 2013 the IRS estimated that an extra hour spent auditing someone who earns $200,000 annually generated only $650. An extra hour spent auditing someone who earns $5 million or more per year generated around $4,900.”
  • Better and more information reporting has regularly been proposed in Congress but lobbying typically prevents enactment or (in some cases) has caused repeal before implementation of enacted provisions.  I’m skeptical about the ability politically to get that done. It is a good idea, though, and should be done. But there is a long history of the government having trouble effectively implementing large IT projects. See the debacle with HealthCare.gov?
  • Investing in IRS IT improvements, as the Taxpayer Advocates has recommended, is essential. Summers and Sarin make a telling comparison of IT spending by Bank of America and the IRS: “[I]n 2018 the IRS spent only $2.5 billion on IT investments. By comparison, Bank of America spent around $16 billion but serves only a quarter of Americans.”  Me: the IRS probably has much more complicated IT demands than BoA does, so the difference based on population understates the IRS underspending.  Since this is an annual difference, it compounds over time!
  • Summers and Sarin also point out (as many others have) that it’s hard to make financial or economic sense out of the large amount of resources that the IRS spends on auditing low income filers (particularly those claiming the earned income credit), when the revenue yield of high income audits is so much greater.  I have some thoughts on why that occurs (under both Republicans and Democrats) related to behavioral economics and the reality that most people (other than economists), including policy makers, consider a dollar of direct government spending (including refundable credits like the EITC) to have a higher value than a dollar of tax expenditure or unpaid taxes. But that will need to be the subject of a separate blog post at some point.

My bottom line:

It certainly makes sense to undertake reasonable proposals to collect taxes already owed before enacting new taxes.  Increasing explicit taxes without better compliance enforcement will only make matters even worse; higher taxes encourage noncompliance and will increase the inequity between voluntary payers and the noncompliant.  As Summers and Sarin point out, this inequity is very regressive, benefitting the affluent.

This issue should be taken up by Congress and, at a minimum, normal funding of the IRS restored.  The agency’s underfunding is close to a scandal, in my opinion.  Based on my experience with the Minnesota legislature, politicians (including tax averse Republicans) are willing to entertain compliance spending and other measures, but it’s not as politically easy as it may seem to academics and without a hard budget constraint (e.g., the state’s requirement to run a balanced budget or strict congressional budget rules), probably not much will happen despite the apparent political appeal from reading Summers and Sarin.

SALT/Minnesota angle:

From a parochial perspective, it would be advantageous for the Minnesota state budget if Congress took Summers and Sarin up on some or all of their recommendations. Because Minnesota’s budget is so heavily dependent on income tax revenues and because the Minnesota tax is directly tied to the federal tax, increased federal compliance is especially important for the Minnesota.  Applying the usual formulas used by state revenue estimators, Summers and Sarin’s trillion-dollar estimate could yield increased Minnesota state revenues in the hundreds of millions per year; the exact amount depends heavily on the extent to which their projected revenues are attributable to increased reporting of capital gains or corporate income. Minnesota’s effective tax rate on capital gains and corporate income is  proportionately higher (relative the federal amounts, that is) than for ordinary income: the relative relationships (for the highest income taxpayers) are: (1) for ordinary income, Minnesota rate is about 28% of the federal rate, (2) for long-term capital gains it is about 40% and (3) for corporate income it is about 45%.

Conversely, the persistent underfunding of IRS compliance has undoubtedly also eroded Minnesota compliance and revenues. When I was still working, I observed more efforts by the Department of Revenue (DOR) to audit compliance with federal rules (that directly affect Minnesota as well, of course).  I witnessed this as a result of constituent complaints about DOR that legislators referred to me.  Earlier in my career that almost never occurred, because (I assume) DOR largely left it to the IRS to audit those matters and, then, notify DOR of its adjustments. DOR must no longer feel it can rely on the IRS to the same extent that it had.

Design a site like this with WordPress.com
Get started