A Victory for Supply-Siders

By BRUCE BARTLETT

Last week, in the House Ways and Means Committee's historic hearing room, Congress's Joint Committee on Taxation held a conference that could be the harbinger of important changes in U.S. tax policy. The JCT brought together some of the nation's best economists to discuss the use of "dynamic" scoring methods in estimating federal revenues. Although no decision was reached on how to implement dynamic scoring, there was a consensus that such methods are feasible and would improve the revenue-estimating process. It was a clear victory for those of us who have been advocating dynamic scoring since the 1970s.

The basic issue is this: Under current official methods used both by the JCT and the Treasury Department's Office of Tax Analysis, the estimated impact of tax changes on federal revenues generally excludes any macroeconomic effects -- ways the changes would influence such things as real GDP growth, unemployment and inflation. In the vast majority of cases this is appropriate, since most proposed tax changes have only a trivial macroeconomic effect. But once in a while Congress considers tax legislation that clearly would have a macroeconomic impact. In these cases, the revenue estimates produced by Treasury and the JCT tend to overstate the rise in revenue from a tax increase and the loss of revenue from a tax cut, creating a bias in favor of higher taxes.

Desperate Tactics

Advocates of tax cuts have long been frustrated by the unwillingness of the Treasury and JCT to include the macroeconomic effects of tax changes in their revenue estimates. And proponents of dynamic scoring long believed that only Democratic control of Congress, and hence control of the JCT and the Congressional Budget Office, prevented the adoption of dynamic scoring. Thus when the GOP took over two years ago, many expected reform of the revenue-estimating process to be high on the Republicans' priority list.

They were disappointed. Rather than embrace dynamic scoring, the Republican leadership trashed the idea. A joint hearing of the House and Senate budget committees on Jan. 10, 1995 -- ostensibly an impartial review of the revenue-estimating process -- was heavily and deliberately stacked against dynamic scoring. So desperate were the organizers to bury the idea that they even trotted out people, like former Federal Reserve Chairman Paul Volcker, who have no expertise in the area but could be counted on to defend the status quo.

Had Republicans in the last Congress instituted dynamic scoring, they might have argued persuasively that pro-growth tax cuts would largely pay for themselves.

Why would Republicans suddenly try to destroy an idea they had been advocating for years? The reason would seem to be some combination of fear and bravado on the part of Budget Committee members. On the one hand, they feared that their colleagues would get carried away with tax cutting, thereby undermining the goal of balancing the budget. On the other hand, Republicans were so flush with victory that they believed dynamic scoring was unnecessary to get a tax cut enacted: They would simply cut as much spending as needed to pay for it. Indeed, some Republicans appear to have opposed dynamic scoring precisely in order to force the largest possible spending cuts.

This view, of course, turned out to be incredibly naive. Bill Clinton's veto pen stood in the way of using spending cuts to pay for any significant tax reduction. Thus, faced with budget rules that allow passage of tax cuts only if they're fully paid for, and having abandoned dynamic scoring, Republicans were unable to cut taxes. Had they instituted dynamic scoring, they might have argued persuasively that pro-growth cuts, such as in the capital gains tax, would in large part pay for themselves. Even if dynamic scoring would show a tax cut recouping only, say, 25% of the static revenue loss, that would still be that much less spending that would have to be cut to get a tax cut enacted.

Eventually, the Republicans recognized their error and moved to rehabilitate dynamic scoring. Under the leadership of Rep. Tom Campbell (R., Calif.), the House voted to approve legislation that would force the Joint Committee on Taxation to use dynamic scoring under limited circumstances. This legislation died in the last Congress, but on Jan. 7 the House amended its rules to include a provision allowing the chairman of the Ways and Means Committee to request a dynamic revenue estimate from the JCT on major tax bills.

This is only a first step. The JCT also needs some generally agreed-upon method for actually producing such an estimate. Last week's conference grew out of a yearlong effort by the JCT, working with many of the nation's best economic modelers, to develop such a method. These modelers fall into two groups: academics and professional forecasters. The former include people like Alan Auerbach of the University of California, Laurence Kotlikoff of Boston University and Dale Jorgenson of Harvard. The later include representatives from DRI/McGraw-Hill, Macroeconomic Advisers and Coopers & Lybrand.

All of the modelers, who use complex mathematics and computers to simulate the economic effects of tax changes, were asked to model two alternative tax-reform proposals. First was a comprehensive income tax, of the type long championed by liberal tax reformers at the Brookings Institution. Second was a flat-rate tax on consumption, such as those proposed by Rep. Dick Armey (R., Texas) and publisher Steve Forbes.

The most critical difference between the two proposals for the modelers is that under a consumption tax businesses are allowed an immediate write-off for capital investments, whereas under the comprehensive income tax they must depreciate such assets over their useful lives. (A consumption tax could take one of several forms -- a national sales tax, a value-added tax or a tax on income less savings and investment. For the modelers' purposes, the form is irrelevant.) The modelers were all given other basic assumptions so as to make the results as comparable as possible.

Every model showed significantly more positive economic effects from adoption of the flat-rate consumption tax than from the comprehensive income tax. This would seem to lay to rest the long-running debate over whether we should move toward a consumption tax or broaden the current income tax and lower the rates. Clearly, we must adopt the consumption tax if we want a larger economy.

The conference's other key conclusion is that the higher growth resulting from adoption of a consumption-based tax system will allow the same revenue to be collected at a lower tax rate. For example, the model developed by Mr. Auerbach, Mr. Kotlikoff, Kent Smetters and Jan Walliser shows a long-run comprehensive income tax rate of 25% being necessary to collect the same revenue as a 22.4% consumption tax rate. The model developed by Mr. Jorgenson and Peter Wilcoxen showed a flat-rate consumption tax of between 18% and 19% yielding the equivalent of current revenues.

Important Victories

It is too soon to declare total victory in the debate over dynamic scoring. But important victories have been won, both in the political arena and in the technical sphere. The next step is to take the insights put forward at the Joint Committee on Taxation conference and actually incorporate the necessary equations into the JCT's own revenue-estimating model. Then Rep. Bill Archer (R., Texas), chairman of the Ways and Means Committee, will have to have enough confidence in the results actually to request a dynamic estimate on some important tax bill. Within a year we may finally see dynamic scoring used in the budget process.

Copyright © 1997 Dow Jones & Company, Inc. All Rights Reserved.