Treasury’s withdrawal of an analysis from its website on who bears the brunt of the corporate tax — labor or capital — elicited concerns that the work of Treasury’s nonpartisan tax research wing is at risk of being politicized.
The study in question, Technical Paper 5, was authored by staff within the Treasury Office of Tax Analysis, and argues that 18 percent of the corporate tax incidence falls on labor income and 82 percent falls on capital. That argument stands in stark contrast to positions adopted by top members of the Trump administration, including Treasury Secretary Steven Mnuchin.
“Most economists believe that over 70 percent of corporate taxes are paid for by the workers,” Mnuchin said September 3 on Fox News Sunday. “And the fact that the Treasury Department, I think it was eight or 10 years ago, put out a piece otherwise — I don’t believe in that. Our current economic team does not believe in that. There is lots of economic research.”
In a series of posts on Twitter, Jason Furman of the Peterson Institute for International Economics described the removal of the study as “unprecedented, or at least highly unusual,” noting that while the Treasury website has over 40 years of working and technical papers, “This is the only one removed.”
“The goal of the technical paper series was to be more transparent about the methodology Treasury used for its modeling and analysis,” he wrote, adding that while the incidence of the corporate tax is subjective and still up for debate, the analysis was “in the mainstream of other modelers.”
Later, Furman told Tax Analysts that “transparency is a bright-line concern. It is just wrong to pull numbered technical papers from a website.”
“I would also have concerns about any politicized process to redo the assumptions underlying the incidence of corporate tax,” Furman said. “I do not know whether or not that is happening, but no Treasury Secretary should have any role in what is a highly technical and analytical process — that should be left to the career staff.”
Economist Paul Krugman was less measured in his reaction: “What we don’t know can’t hurt them. Why would anyone believe these guys on anything?” The tweet included a link to a story describing the study’s removal.
For now, Treasury’s website does retain a vestige of the missing technical paper, with a link to a document titled, “Treasury’s Distribution Methodology and Results” dated November 12, 2015, which includes a footnote setting the split between capital income and labor income at 81.5 percent and 18.5 percent, respectively.
The White House repeated Mnuchin’s claim in a release related to a September 29 tax reform speech by President Trump. “According to one study by the [Congressional Budget Office], over 70 percent of the corporate tax burden falls on American workers,” a bullet point in the release read.
When asked about that CBO research, a White House spokesperson referred Tax Analysts to a 2006 working paper. The withdrawn paper itself discusses that CBO document, noting and essentially endorsing scholarly critiques of its assumptions and conclusions before offering a new approach.
As of press time, Treasury had not responded to Tax Analysts’ request to clarify whether it intended to change its corporate income tax incidence ratio for future distributional analyses.
Mnuchin, along with others in the Trump administration, has championed the idea that cuts in the corporate tax rate are a key part of fulfilling their promise for a middle-income tax cut.
“This is about creating jobs, because many surveys show that 70 percent or more of the [corporate tax] burden is borne by the American worker. And this is about putting money back in the American worker’s pocket,” Mnuchin said April 27 on CBS’s This Morning.
In an August 16 paper, Chye-Ching Huang and Brandon DeBot of the Center on Budget Policy Priorities warned that such statements could mean the administration is planning to sell their corporate tax cuts as helping workers.
“Assumptions about who ultimately pays the corporate tax have a substantial impact on how the benefits of cutting the tax are estimated to be distributed across income groups,” they wrote in their analysis. “Mnuchin’s embrace of an abnormally large share flowing to labor — far more than the Treasury Department assumes, or assumed during the George W. Bush Administration — means that a Trump Administration distributional analysis could show low- and middle-income workers receiving unrealistically large benefits.”
Adopting the view that 70 percent of the corporate tax burden falls on labor income could “make a significant difference” in determining whether the administration’s tax reform plan as presented in the framework can be packaged as a middle-income tax cut, depending on how middle-income is defined, Alan Viard of the American Enterprise Institute told Tax Analysts.
If middle-income is defined to include taxpayers earning upwards of $200,000, then that would be “putting a significant share of that into the middle class,” he said. “It does make a significant difference.”
Viard added that labor income is “still fairly concentrated” among higher-income earners, however, so a “high-wage earner is still certainly going to get a bigger dollar tax cut than the people in the middle class.”
Eric Toder of the Urban-Brookings Tax Policy Center, speaking on a September 29 conference call with reporters about a new report on the framework’s distributional effects, agreed that the distributional impact would shift somewhat. Like Viard, he said the corporate tax incidence “still doesn’t fall on middle-income workers because most labor income is earned by very-high-income people.”
But when asked to speculate on what the distributional impact might be, TPC Director Mark Mazur said to do so would be to use assumptions that are “so far out of the mainstream thinking.”
Mazur, who previously was deputy assistant secretary for tax analysis in Treasury and also served as Treasury assistant secretary for tax policy, later said, “It doesn’t really do us a whole lot of good to speculate on what kind of assumptions and gimmicks are going to be used to try to make the [distributional] tables come out right . . . Our analysis is very much in the mainstream of what government agencies do on distributional analyses. And it really is not incredibly helpful of us to speculate on what might happen at some future date.”
Luca Gattoni-Celli contributed to this article.
Follow Jonathan Curry (@jtcurry005) on Twitter for real-time updates.