On November 15, 1994, Office of Management and Budget Director Alice Rivlin told reporters that a Congressional Budget Office (CBO) director acceptable to Republican leaders could produce "dishonest" budget numbers. Rivlin was arguing that a departure from current CBO practices would be unrealistic and lead to budget problems. Her claim can best be evaluated by reviewing CBO analysis of relevant policies and examining their accuracy. This paper will focus on the capital gains issue, an important component of the Republican "Contract With America."
The CBO is an organization of congressional staff that estimates the effects of changes in budget policy. Under the existing budget process, the CBO staff has the ability to effectively deny elected Members of Congress the opportunity to offer legislation by triggering stiff procedural obstacles. CBO and the Joint Committee on Taxation (JCT) are often criticized for failing to adequately account for economic and behavioral changes that would occur under new policy initiatives. In so doing the CBO can make it much more difficult, or virtually impossible, for new policies to be considered.
In evaluating CBO the factual record is the most important, if often neglected issue, particularly with respect to capital gains. A review of the facts shows that CBO failed to take into account the effects of higher capital gains taxes after 1986, producing huge forecasting errors. By using grossly mistaken estimates of capital gains income (off by over 100% in most cases), CBO analysis misstated the effects of capital gains legislation introduced in 1989 and 1990, also misinforming Congress about the growth path of federal revenues. Ranking JEC Republican Dick Armey first uncovered and disclosed these massive CBO errors early in 1991 by releasing a study on this subject, later reprinted in the tax specialist publication Tax Notes. Equally disturbing was the fact that CBO had never disclosed or explained its botched scoring of the capital gains legislation to House Members, a majority of whom supported the capital gain tax cut.
In 1989 and 1990, CBO projected that capital gains realizations (income) would be $225 billion in 1989, $254 billion in 1990, $268 billion in 1991, and $287 billion in 1992. This was the tax base from which the projected revenue losses, and supposedly unfair distributional results of the capital gains tax cut legislation, would be calculated. It was also used as part of the tax base determining the level of federal revenues in these years.
A JEC research project requested by Rep. Armey found that the CBO capital gains realization estimates were grossly overstated. By early 1991 it became clear that these figures were inflated by at least $70 billion annually. A follow-up JEC investigation in 1992 revealed that the magnitude of the error had grown to at least $130 billion annually for the years after 1989. By 1992, with the latest IRS data available, the actual capital gains realizations, at $118 billion, were actually about $170 billion lower than CBO had projected. In sum, the actual versus CBO-projected capital gains realizations were $154 billion versus CBO's $225 billion for 1989, $124 billion actual versus $254 billion projected for 1990, $111 billion versus $268 billion projected by CBO for 1991, and $118 billion versus $287 billion projected for 1992. The CBO errors amounted to $71 billion for 1989, $130 billion for 1990, $157 billion for 1991, and $169 billion in 1992, amounting to a cumulative error of $527 billion, or half a trillion dollars, over the four years.
These facts are not in dispute, but are simply ignored by CBO's advocates in Congress and the media. CBO's static methodology created huge forecasting errors and grossly erroneous CBO budget and distributional scoring of the capital gains legislation of 1989 and 1990. Furthermore, CBO failed to disclose its botched scoring of this legislation to the members of Congress, the media, and the public. This disturbing lack of candor raises further questions, especially since the flawed data were included in the already defective CBO family income data used by the Democrats to fabricate the "fairness" issue. After Armey's repeated disclosure of new defects in these family income data, they were finally discontinued by CBO.
The Armey study first exposing CBO's capital gains problem, Distorting the Data Base: CBO and the Politics of Income Redistribution, was released in April of 1991. Not only was the CBO's scoring of the capital gains legislation grossly erroneous, but the overstatement of capital gains income also resulted in an overstatement of revenue growth. Armey pointed out that the result would be higher levels of deficit spending in coming years. CBO dismissed this concern, but later had to make huge "technical" reestimates, i.e. corrections, to its projection of budget revenues and deficits as a result.
Of the many reasons why CBO and JCT analysis of the so-called distributional effects of capital gains are defective, one of the more fundamental is that much of the capital gains tax revenues paid are actually ignored. CBO and JCT acknowledge that a reduction in the capital gains tax rate would unlock investment and generate revenues that would offset at least some of the static revenue losses. These additional revenues from unlocking are included in estimates of the budgetary effects of reducing the tax rate, though they are understated. However, the key point is that in allocating the tax benefits of the tax cut, the revenues from unlocking are simply ignored. Though this is indefensible, CBO and JCT argue that since a taxpayer induced to sell by a lower tax rate is improving his economic situation, the resulting taxes paid should not be counted in the distributional analysis. This is clearly an absurd position, as virtually all taxes could be excluded on analogous grounds with respect to labor and other income, the generation of which also improves economic welfare. The bottom line is that skewed appearance of capital gains benefits tilted to the upper income groups is based on a number of flawed assumptions, including the simple exclusion of tax payments acknowledged to be made but officially ignored for the purpose of distributional analysis.
It is ironic that CBO's alleged superior accuracy in static revenue forecasting could be invoked by its former director, Alice Rivlin, with respect to the Contract, the centerpiece of the debate being capital gains, of all things. The facts on this matter are devastating to CBO, which is why they are consistently ignored by the agency's advocates. It is preposterous to tout the accuracy of the CBO in a debate over dynamic forecasting and capital gains when the truth is that CBO's track record is abysmal if not embarrassing. Perhaps this is why the CBO failed to disclose its huge errors on capital gains to the Congress and public in the first place. In any event, the huge forecasting mistakes of CBO on capital gains, and CBO's lack of candor displayed with respect to them, totally discredit CBO's revenue forecasting methods, at least with regard to capital gains.
Click here to see Graph.
Return Home