IRS audits of high earners are incredibly effective at raising tax revenue and reducing deficits.

New research shows that for each $1 spent on an audit of a high income individuals, the audit recovers $2 in tax revenue immediately, and then another $10 over the next 14 years as audited individuals voluntarily pay more in taxes due to deterrence effects

https://policyimpacts.org/research/67/a-welfare-analysis-of-tax-audits-across-the-income-distribution

A Welfare Analysis of Tax Audits Across the Income Distribution

Will Boning, Nathaniel Hendren, Ben Sprung-Keyser, and Ellen Stuart

the image above is from the Washington Post writeup of the research

https://www.washingtonpost.com/opinions/interactive/2023/irs-enforcement-costs-congress-funding/

Congress just cut IRS funding. It costs even more than we thought.

New research finds that for each extra dollar the IRS spends auditing wealthy taxpayers, it can collect more than $12 in return.

The Washington Post
@benzipperer great example of nonlinear policy effects
@benzipperer Well, here's why CONservatives have been squawking about funding the IRS... though if I were going to place a bet, I'd bet that 90% of the people waxing apoplectic about "armed IRS enforcers" or whatever florid fantasy are ones whose income makes them unlikely to be an audit target in the first place... average income from payroll rather than millionaires' magic tax dodges.

@benzipperer

If you ever wondered why Republicans like firing tax collectors....

@Stalkholm @benzipperer

I saw a number recently that under payment of taxes, intentional or not, costs the US hundreds of billions every year. And if you include use of loopholes to avoid tax - like offshore shell companies - it rises to two trillion dollars a year. That's about $6000 per man woman and child in the US. That's surely enough to fully fund social security, provide healthcare for all, feed the hungry, house the homeless, and probably still have some left over for a rainy day.

@benzipperer Gee, no wonder those rich fuckers are against funding the IRS. Shocking! 🙄

@benzipperer

Gosh, does that mean high earners are cheating on their taxes? (snort)

@benzipperer This is both good and bad.
@benzipperer Worth noting that this is likely to be an underestimate of deterrence effects, because it only estimates the individual deterrence due to being audited, but collective deterrence (increased tax compliance because you are aware others are being audited) is likely real and not measured by this methodology. (This point is made explicitly in the last paragraph of the conclusion but bears highlighting IMO)
@benzipperer Yeah, but then they've less money to line the politician's pockets.
@benzipperer the figure of $12 over 14 years doesn't seem so impressive compared to $6 for the poorest. Is there really so much tax they can squeeze out of the poorest?
@ianchanning @benzipperer Haven't read the full paper yet, but I'm guessing the bump there is due to EITC audits, which I've read elsewhere make up a large number of the IRS's audits, and target a credit used by low-income taxpayers. I'd imagine that, while no one audit of that type recovers a lot, they're relatively easy and cheap for the IRS to do, especially when the targets of the audits are in less of a position to put up a fight than rich taxpayers.

@benzipperer

But…but…but I was told that collecting more taxes would result in lower revenue!

@benzipperer How much confidence do we put in this averages with projections style of estimation? I know a lot of policy writing typically uses identities and structural models.