The Problem With the IRS Vow to Avoid Stepping Up Audits of Earners Under $400,000

The audit rate has newfound relevance because of the IRS’s new $80 billion in funding. Treasury Secretary

Janet Yellen

had directed that “any additional resources—including any new personnel or auditors that are hired—shall not be used to increase the share of small business or households below the $400,000 threshold that are audited relative to historical levels.”

The new IRS commissioner,

Danny Werfel,

confirmed by the Senate last month, repeated that pledge this week. He told reporters the IRS had no plans to increase “the most current audit rate we have for households making less than $400,000…We will not come close to hitting or exceeding any historic average rate.”

At all income levels, the audit rate has plunged over the past decade amid declining IRS resources and head count, as a Government Accountability Office report released last year shows. 

For the 2019 tax year, there were 22.9 million individual tax returns reporting between $100,000 and $200,000 in income, of which 37,000 were audited. That works out to an audit rate of just 0.2%. That is down sharply from 128,000 audits, or 0.8%, out of 15 million returns in 2010. (The IRS has three years to initiate an audit, so the figure for 2019 could still rise somewhat because of audits that started after the GAO report was written.) Over the 10-year period, the average rate was 0.45%, meaning the IRS could still have some wiggle room for audits without going beyond the historic average.

Still, it might raise the question: Why bother? True, $400,000 is a fairly high threshold—fewer than 2% of tax returns have adjusted gross income above that, according to IRS data. But why pursue those families when there might be multimillionaires and billionaires dodging millions of dollars in taxes?

Well, one reason is that the standard calculation of audit rates falls short. They are based on what taxpayers report their income to be. But tax dodgers are by definition underreporting their income.  

“People might think it’s very clear who is and who isn’t high net worth,” said

Natasha Sarin,

a Yale law professor and former Treasury official. “The whole point of tax evasion is people with high net worth are doing a lot of maneuvering and accounting tricks to look like they’re not.”

So while most people who report, say, $300,000 in income are upper-middle-class households with straightforward finances—salaries, a mortgage, children—some are very rich people whose actual income is far above that level but are especially audacious or creative in reporting a lower figure. Those are the people the IRS might very much want to audit.

The problem, simply put, is that reported and actual income are different. Indeed, exempting taxpayers with reported income below a certain threshold can create incentives that widen the gap.  

An unscrupulous person who earned $500,000 might decide it is better to hide $100,000 instead of just a few thousand dollars, because they would then have the reported income of an ordinary person that the IRS has pledged not to pursue too hard.

(Underreporting doesn’t just distort audit rates, it might distort measures of inequality that are based off tax data. In a 2021 paper, a quintet of economists, including two from the IRS, studied random audits to determine how much income was being missed. They estimate that 7% of actual income goes unreported among the bottom 50% of taxpayers, but over 20% for the top 1%, meaning the latter’s share of total income is larger than reported.)

Audit rates also present an incomplete picture of how efficiently the IRS allocates enforcement efforts. Why would the IRS ever audit a lower-income taxpayer when auditing a wealthy taxpayer presumably yields more tax? Because from the IRS’s perspective, not all audits require the same effort. It spends a little over six hours on average auditing a taxpayer reporting $25,000 to $200,000 in income, but almost 34 hours on one reporting $500,000 to $5 million. So audit efficiency is best measured by additional revenue per hour of work, rather than audit rate. 

By this metric, the most effective use of IRS time is going after returns reporting over $5 million in income, which yield an extra $5,000 of tax owed per hour of work. But the second most effective use is auditing households at the opposite end of the scale: low-income people who claim the earned-income tax credit, which yields nearly $3,000 per hour. The IRS has said these audits take little time and have a high rate of funds being recouped because the audits are typically of people seeking a refund, and occur before the refund is issued.

Little time spent by the IRS doesn’t necessarily mean little time spent for the taxpayer. I learned this lesson the hard way a number of years ago when a mistake reporting a withdrawal from a Health Savings Account triggered an automatic letter. This is something less than an audit, and probably took the IRS mere minutes. But for me, it took dozens of hours plus hiring an accountant and a lot of stress.

So to answer the original question—what are your odds of being audited?—the audit rate turns out not be a good metric, because audits aren’t random. They are triggered by something unusual or a conspicuous mistake. The task facing the IRS: What is the trigger for identifying a taxpayer pretending to be middle class?

Write to Josh Zumbrun at [email protected]

The audit rate has newfound relevance because of the IRS’s new $80 billion in funding. Treasury Secretary

Janet Yellen

had directed that “any additional resources—including any new personnel or auditors that are hired—shall not be used to increase the share of small business or households below the $400,000 threshold that are audited relative to historical levels.”

The new IRS commissioner,

Danny Werfel,

confirmed by the Senate last month, repeated that pledge this week. He told reporters the IRS had no plans to increase “the most current audit rate we have for households making less than $400,000…We will not come close to hitting or exceeding any historic average rate.”

At all income levels, the audit rate has plunged over the past decade amid declining IRS resources and head count, as a Government Accountability Office report released last year shows. 

For the 2019 tax year, there were 22.9 million individual tax returns reporting between $100,000 and $200,000 in income, of which 37,000 were audited. That works out to an audit rate of just 0.2%. That is down sharply from 128,000 audits, or 0.8%, out of 15 million returns in 2010. (The IRS has three years to initiate an audit, so the figure for 2019 could still rise somewhat because of audits that started after the GAO report was written.) Over the 10-year period, the average rate was 0.45%, meaning the IRS could still have some wiggle room for audits without going beyond the historic average.

Still, it might raise the question: Why bother? True, $400,000 is a fairly high threshold—fewer than 2% of tax returns have adjusted gross income above that, according to IRS data. But why pursue those families when there might be multimillionaires and billionaires dodging millions of dollars in taxes?

Well, one reason is that the standard calculation of audit rates falls short. They are based on what taxpayers report their income to be. But tax dodgers are by definition underreporting their income.  

“People might think it’s very clear who is and who isn’t high net worth,” said

Natasha Sarin,

a Yale law professor and former Treasury official. “The whole point of tax evasion is people with high net worth are doing a lot of maneuvering and accounting tricks to look like they’re not.”

So while most people who report, say, $300,000 in income are upper-middle-class households with straightforward finances—salaries, a mortgage, children—some are very rich people whose actual income is far above that level but are especially audacious or creative in reporting a lower figure. Those are the people the IRS might very much want to audit.

The problem, simply put, is that reported and actual income are different. Indeed, exempting taxpayers with reported income below a certain threshold can create incentives that widen the gap.  

An unscrupulous person who earned $500,000 might decide it is better to hide $100,000 instead of just a few thousand dollars, because they would then have the reported income of an ordinary person that the IRS has pledged not to pursue too hard.

(Underreporting doesn’t just distort audit rates, it might distort measures of inequality that are based off tax data. In a 2021 paper, a quintet of economists, including two from the IRS, studied random audits to determine how much income was being missed. They estimate that 7% of actual income goes unreported among the bottom 50% of taxpayers, but over 20% for the top 1%, meaning the latter’s share of total income is larger than reported.)

Audit rates also present an incomplete picture of how efficiently the IRS allocates enforcement efforts. Why would the IRS ever audit a lower-income taxpayer when auditing a wealthy taxpayer presumably yields more tax? Because from the IRS’s perspective, not all audits require the same effort. It spends a little over six hours on average auditing a taxpayer reporting $25,000 to $200,000 in income, but almost 34 hours on one reporting $500,000 to $5 million. So audit efficiency is best measured by additional revenue per hour of work, rather than audit rate. 

By this metric, the most effective use of IRS time is going after returns reporting over $5 million in income, which yield an extra $5,000 of tax owed per hour of work. But the second most effective use is auditing households at the opposite end of the scale: low-income people who claim the earned-income tax credit, which yields nearly $3,000 per hour. The IRS has said these audits take little time and have a high rate of funds being recouped because the audits are typically of people seeking a refund, and occur before the refund is issued.

Little time spent by the IRS doesn’t necessarily mean little time spent for the taxpayer. I learned this lesson the hard way a number of years ago when a mistake reporting a withdrawal from a Health Savings Account triggered an automatic letter. This is something less than an audit, and probably took the IRS mere minutes. But for me, it took dozens of hours plus hiring an accountant and a lot of stress.

So to answer the original question—what are your odds of being audited?—the audit rate turns out not be a good metric, because audits aren’t random. They are triggered by something unusual or a conspicuous mistake. The task facing the IRS: What is the trigger for identifying a taxpayer pretending to be middle class?

Write to Josh Zumbrun at [email protected]