(Politico) - A simmering dispute between Democrats and the Congressional Budget Office over how much money could be raked in by improving IRS enforcement largely boils down to a single question: How afraid will rich people be of a newly empowered tax collector?
Lawmakers and the White House want to count $400 billion in budget savings toward the cost of their reconciliation package, including $240 billion projected to come from increased audits of corporations and the wealthy.
More controversially, they also want to pencil in $160 billion in savings from people who aren’t themselves audited but who they say would see the IRS cracking down on scofflaws and decide it’s not worth trying to cheat on their own taxes.
That has become a major point of contention with the CBO, which is working on a much-anticipated cost estimate of Democrats’ plans.
The official accounting agency agrees increased enforcement will deter some would-be cheaters but says the effect would be modest, in part because the well-to-do are less likely to be intimidated by tax examiners.
“There’s other analyses that put a lot of weight on the idea of deterrence — that if people know that the IRS is doing more and auditing more that other people then pay more of their taxes,” CBO director Phillip Swagel said in a webinar this week. “The research literature on deterrence, I’d say, is very mixed.”
His agency is expected to say, as soon as Thursday, that Democrats’ IRS plan would generate somewhere in the neighborhood of $150 billion in savings over the next decade — less than half of what Democrats are hoping for.
That would likely show Democrats’ social spending plan is not fully paid for, despite promises by party leaders that it would not add to the budget deficit.
The question then will be how much that matters to lawmakers, particularly moderates who’ve demanded the CBO analysis before putting the package to a vote.
Aware CBO is unlikely to agree with its estimates, the administration has been huddling with centrists to explain why it believes the agency’s numbers will be wrong.
Some lawmakers have signaled they are willing to ignore the nonpartisan budget scorekeeper.
“I don’t think we should be put off for a second on these suggestions that it would only bring in $125 billion or $150 billion,” House Ways and Means Committee Chair Richard Neal (D-Mass.) told reporters. “An investment in the IRS will return considerable dollars.”
All sides agree that boosting funding for the IRS would be a moneymaker for the government because auditors would bring in more money than it costs to employ them. The question is: How much more?
Predicting that is no simple matter, with forecasters taking into account a host of factors, from how long it would likely take the IRS to hire and train new workers — no small thing in a tight labor market — to how long audits would likely last (CBO figures an audit of medium complexity takes two years).
Treasury and CBO differ on all those points, but their biggest disagreement is over how much Democrats’ plan would deter people from cheating on their taxes in the first place.
It’s an inherently tough topic because it means predicting how wealthy people and corporations will perceive the IRS in the future.
“It’s fair to take into account deterrence,” said Janet Holtzblatt, a former top official at both Treasury and CBO. “But it’s difficult to estimate.”
The funding increase Democrats are pushing is unprecedented, so there’s little in the way of historical evidence to consider. And academic research into the question of deterrence has been inconclusive, said Holtzblatt, now at the Tax Policy Center.
“You can interpret the research very differently — as these two agencies have done,” she said.
Government scorekeepers had traditionally omitted deterrence from their projections because of the uncertainties. Treasury itself initially left it out of its own estimates of the IRS funding proposal. As part of its May budget request, it predicted that increasing IRS funding by $80 billion would generate $320 billion in additional revenue, for a net savings of $240 billion.
That did not include deterrent effects, which Treasury said at the time were significant, though “more research is needed to arrive at a better understanding of the magnitude.”
But CBO said in September that it believed deterrence should be taken into account. And Treasury soon thereafter said it had revised its own estimate to include that effect as well — predicting that an $80 billion investment in the IRS would produce $480 billion in revenue, for a net savings of $400 billion.
“Just like having a police officer on the median of an interstate will result in fewer cars excessively speeding, having more IRS agents able to audit potential tax evaders should increase voluntary compliance with the tax system or deter tax evasion,” the agency said in a memo to lawmakers. “Recent academic work speaks to the importance of deterrent effects.”
That $160 billion number is conservative, the administration says, pointing to studies suggesting the savings would likely be much higher.
“There’s some experts who say you could raise substantially more than $400 billion,” said Ben Harris, Treasury’s assistant secretary for economic policy.
“The notion that we would have this generational increase in IRS funding that did not lead to a marked change in taxpayer behavior feels implausible,” he said.
CBO agrees more people will comply if they believe they are likely to get caught cheating — but not everyone.
Some people who get audited will learn what sort of tax maneuvers the IRS is able to detect and move to other ways of slicing their tax bills. Some people may cheat again because they believe they’re unlikely to be audited a second time.
And, CBO says, the high earners and large companies targeted by Democrats are less likely to be cowed by the IRS than average people.
“Research about such deterrence finds varying responses, depending on the taxpayer,” the agency said in a September blog post. “People generally increased their reported income in the years following an audit, but people with higher income generally do not, and neither do corporations.”
By BRIAN FALER
