Estimating Revenue Losses from the Armey Flat Tax: A Chronology

Rep. Dick Armey introduces H.R. 4585, calling for a 17% flat-rate tax to replace all federal personal and corporate income taxes. The bill is patterned after a flat-rate consumption tax designed by Hoover Institution economists Robert Hall and Alvin Rabushka in 1983. Armey's version, however, entails both a lower tax rate and much higher standard deductions against the wage portion of the tax. Armey claims his plan will lose $40 billion a year in revenues at its proposed 20% (two-year) temporary rate.⁽¹⁾

September 1994:

In an article in the Washington Post (Sept. 25, 1994, p. C2), Citizens for Tax Justice director Robert McIntyre estimates that the Armey 17% flat tax plan, as drafted, "could cost in excess of $200 billion annually." A few weeks later, CTJ provides a detailed outline of its analysis, which is based on the revenue-estimating methodology outlined by Armey's mentors, Hall and Rabushka, in their 1983 book, Low Tax, Simple Tax, Flat Tax. CTJ pegs the revenue cost of the Armey plan at $220 billion a year at its proposed 17% rate. (Because of unresolved transition and compliance issues, CTJ calls the $220 billion revenue-loss estimate "optimistic.")

Rep. Armey immediately challenges CTJ's estimate. He repeats his claim that his bill will lose "only $40 billion"--but again at the plan's temporary 20% tax rate, rather than its permanent 17% rate. Armey calls CTJ's analysis "fundamentally flawed," says that CTJ "manipulated data" and accuses CTJ of "outright dishonesty." Although Armey offers no explanation of exactly what "major error" he found in CTJ's figures, he calls CTJ's analysis "discredited," and falsely tells reporters that CTJ has "admitted its error."

On September 21, 1994, a few days before McIntyre's Post article appeared, the congressional Joint Committee on Taxation writes to then majority leader Richard Gephardt in response his request for an analysis of the Armey plan. The letter notes:

"Viewed in isolation, the effect of the [Armey plan's] changes in the standard deduction, exclusion of investment income, and reduction in the top marginal tax rate could cause decreases in revenue in excess of $200 billion annually."

Joint Tax's letter goes on to say: "However, we cannot provide our normal revenue estimates for the bill's impact on individuals for the following reasons: [basically, because Armey's bill was too poorly drafted to answer some key questions about the tax base]." Joint Tax concludes that "we are unable to provide either a revenue or distribution analysis of H.R. 4585 [the Armey plan] at this time."

Rep. Armey seizes on Joint Tax's complaints about the poor drafting of his bill to argue that no one could possibly cite Joint Tax's tentative revenue-loss estimate to buttress CTJ's analysis.

October 1994:

The Treasury Department releases an extensive computer analysis of the Armey flat tax (relying heavily on Hall and Rabushka's 1983 book for details that are lacking in Armey's bill). In "An Analysis of a Flat-Rate Consumption Tax," Treasury finds that at a 17% rate, Armey's plan will lose $244 billion a year in revenues, assuming retention of the earned-income tax credit for the working poor, or $219 billion annually if the EITC is eliminated (as CTJ had assumed). Thus, despite a somewhat different estimating methodology, Treasury's finding is virtually identical to CTJ's September 1994 estimate.
Treasury notes that (obviously) it would take a much higher tax rate (25% to 26%) or much lower exemptions than Rep. Armey has proposed for his plan to break even. And under a revenue-neutral Armey flat tax, Treasury concludes, the vast majority of American families will pay much higher taxes, while the very rich will get enormous tax cuts.
Treasury's distributional findings match previous conclusions by the authors of the Armey plan, Hall and Rabushka. In their 1983 book, they had conceded that their flat tax would "be a tremendous boon to the economic elite" (p. 67), and admitted:

"Now for some bad news. . . . [I]t is an obvious mathematical law that lower taxes on the successful will have to be made up by higher taxes on the average people." (p. 58)

Rep. Armey responds to Treasury's findings by calling them "garbage" and a "partisan assault by the Democrats."
Armey offers no response, however, to the fact that 12 years earlier, Ronald Reagan's Republican Treasury Department reached exactly the same conclusion. At a 1982 Senate Finance Committee hearing on various tax proposals (including an Armey-style flat tax plan), the Reagan Treasury testified that "any" flat-rate tax "would involve a significant redistribution of tax liability" away from the wealthy and onto average taxpayers.

Rep. Armey also complains that Treasury had misunderstood some of the technical details of his plan. In particular, he says, Treasury failed to take account of the fact that his plan treats employer-paid Social Security taxes as taxable fringe benefits to workers.
Treasury responds that it is surprised to hear that Armey intends to tax employer-paid Social Security taxes. But even with Armey's odd technical revisions, Treasury notes, Armey's plan still loses $186 billion annually (assuming retention of the earned-income tax credit). Treasury also points out a distributional analysis of a revenue-neutral version of this restated Armey plan will almost certainly show the same harsh consequences for most taxpayers as the plan Treasury originally analyzed.

Rep. Armey then falsely declares that Treasury has "admitted" its October analysis of his plan was completely wrong!

April 1995:

Treasury puts out an update of its October 1994 analysis of the Armey plan, taking account of Armey's proposed treatment of employer-paid Social Security taxes as taxable fringe benefits. As already noted, the cost of the plan is pegged at $186 billion annually. Treasury points out that for the restated Armey plan to break even, either its exemptions must be cut by about two-thirds or its rate must be raised to about 23%. Not surprisingly, Treasury also finds that the distributional effects of a revenue-neutral version of the restated Armey plan are virtually identical to those found earlier, i.e., the very rich get huge tax cuts, while everyone else pays much higher taxes.

May 1995:

Representative Armey announces to a Texas television audience that his plan will repeal both the earned-income tax credit for the working poor and the federal estate and gift tax on the wealthy. In combination, these offsetting changes cut the revenue loss from his plan by about $8 billion--thus to $178 billion annually at a 17% rate based on Treasury's figures.
On the same television show, Armey again attacks Treasury's and CTJ's revenue estimates. He promises that "honest" estimates will soon arrive from the congressional Joint Committee on Taxation that will confirm his claims that the bill does not lose huge amounts of revenue.

Meanwhile in May, Armey's gurus, Hall and Rabushka, publish a new version of their book, called The Flat Tax. In it, they say that a break-even flat tax will take both a higher tax rate and much lower standard deductions than Armey proposes.
In fact, once Hall and Rabushka's calculations are adjusted to (a) fix a technical error, (b) include Armey's newly-proposed repeal of the estate tax and (c) bring revenues fully in line with current law, their figures indicate that the flat tax rate would have to about 21 percent, even with standard deductions one fourth less than Armey proposed in his 1994 bill.⁽²⁾

July 1995:

On July 19, Rep. Armey (with Sen. Richard Shelby (R-Ala.)) introduces a revised version of his flat tax plan. It features standard deductions about a sixth lower than he had originally proposed--apparently to shave the plan's revenue shortfall. Armey declares that this amended plan, too, will lose $40 billion annually at its 20% temporary tax rate--thus tacitly repudiating his earlier assertions about the cost of his original proposal.
Nevertheless, Armey continues vociferously to deny Treasury's (and CTJ's) estimates of the huge cost of his plan at its 17% permanent tax rate. Instead, he again claims that he has "asked the Joint Committee on Taxation to score our bill"--although the idea that the tax staff would have refused to respond to the majority leader's alleged earlier request seems dubious.

On July 25, CTJ points out that Armey has not only explicitly backtracked on his earlier assertions about his plan's revenue losses, but also has implicitly confirmed Treasury's findings about the cost of his plan at its 17% permanent tax rate. In fact, CTJ notes, simple arithmetic shows that a plan that loses $40 billion at a 20% tax rate must lose $146 billion at a 17% rate.⁽³⁾ (For more details, see CTJ's "Notes on Dick Armey's Revised 'Flat Tax,'" July 25, 1995.)

August 1995:

On August 11, at a United We Stand America conference in Dallas, Rep. Armey tells deficit-conscious Ross Perot supporters that he "would 'gladly' adjust his [flat tax] legislation (HR 2060) . . . to ensure that it remains revenue neutral."⁽⁴⁾ Armey does not specify what adjustments he would "gladly" make, but instead emphasizes that his revised plan requires a three-fifths majority vote in Congress to raise its tax rate or reduce its standard deductions.⁽⁵⁾

1. Thus, even from the beginning, Armey's own figures seem to imply that his plan would lose close to $150 billion at his 17% proposed permanent tax rate. See July 1995 below.

2. Hall and Rabushka suggest a 19% rate for their plan at 1993 levels, assuming retention of the estate tax. In computing that 19% rate, however, they compared actual fiscal 1993 income tax revenues to calendar year GDP. Correcting for this error adds about 0.6 percentage points to their calculated tax rate. Taking account of Armey's proposed repeal of the estate tax adds another 0.4 percentage points. And bringing their figures in line with current tax revenues adds still another 0.7 percentage points. That raises the total tax rate required to break even using Hall and Rabushka's methodology to 20.8 percent. And even that rate may be too low. Hall and Rabushka also note: "Our estimates assume that the IRS will learn about all the . . . unreported income. [I]t is possible that our estimates of the base for the flat tax are a little optimistic. . . . We do not think we are far off, however." See The Flat Tax (1995), pp. 57-58

3. Current personal and corporate income taxes and the estate and gift tax raise $747 billion at 1995 levels. If Armey's plan loses $40 billion at a 20% tax rate, then it must raise $707 billion. If the rate is reduced to 17%, then Armey's plan raises only $601 billion ($707 billion × 17/20ths). That leaves a revenue loss (at 1995 levels) of $146 billion a year. The difference between that figure and Treasury's $178 billion estimate for the earlier version of the plan stems from Armey's reduction in his proposed standard deductions.

4. BNA Daily Tax Report, Aug. 15, 1995, p. G-1.

5. At the same conference, "Armey also denied that the plan would exempt unearned income [such as interest]. He said the plan would tax unearned income the same as wages--at 17 percent with no exceptions." Armey is quick to contradict himself, however. "He said the elimination of the [home mortgage] interest deduction would be offset by eliminating the tax on interest earnings." BNA Daily Tax Report, Aug. 15, 1995, p. G-1.

Back To Reports