In this year’s budget, President Bush is requesting an additional $513,000 for the Treasury Department’s Office of Tax Analysis in order to create a new division for dynamic analysis. The idea is to improve estimates of the revenue effects of proposed tax changes so that they are more accurate. But many critics charge that the goal is obfuscation — making the revenue losses from tax cuts appear smaller than they really are.
The debate over dynamic scoring goes back almost thirty years. Historically, estimates of the revenue effect of tax changes were done by accountants, people who simply took the most recent year’s tax data, plugged in the proposed changes, and looked at the revenue effect. They then would try to estimate the impact on future revenues by making some assumptions about growth in the number of taxpayers and the tax base. The estimates were normally done only for a single year.
Eventually, economists replaced accountants, and computers replaced adding machines. More sophisticated methods of estimating inflation, economic growth, and employment were incorporated into the estimates. However, the economists retained one of the accountants’ operating principles: The tax changes were assumed to have no impact on behavior or the economy as a whole. Hence, this method of revenue estimating came to be called “static” analysis.
One reason for this is that prior to the 1970s economists didn’t really have the mathematical tools to make a dynamic analysis that incorporated all the effects of tax changes on things like work, saving, and investment. Furthermore, there was really no demand for dynamic analysis decades ago because the vast bulk of proposed tax changes were too small to have any impact on the economy as a whole.
Another reason is that until the 1970s, most economic thinking was dominated by the theories of John Maynard Keynes, who believed that fiscal policy affected the economy only through its impact on disposable income. Consequently, the incentive effects of taxation interested very few economists.
The great recession of 1973-75 was a severe blow to Keynesian economics because inflation was high while at the same time there was significant unused capacity in the form of unemployment and idle factories. Theoretically, this wasn’t supposed to happen. Also, the failure of traditional Keynesian medicine, especially the tax rebate of 1975, led economists to search for other causes and cures for economic malaise.
One group of economists fingered the capital-gains tax as a key problem area because it had especially pernicious effects on entrepreneurship and risk-taking. Historically, the tax on long-term capital gains had been fixed at 25 percent. But in 1969, congressional liberals raised the rate to 35 percent in order to soak the rich — those who realized the most capital gains. The result was that venture capital virtually dried up while it became much more difficult to finance new business start-ups.
In 1978, a bipartisan effort was made in Congress to cut the capital gains rate back to 25 percent. The problem was that static scoring showed this to be a big revenue loser because it was assumed that the same amount of gains would be realized, only taxed at a lower rate. But one does not need to be a professional economist to see that when you cut the price of something, sales of it will probably rise.
Advocates of cutting the capital gains rate, including Harvard economist Martin Feldstein, argued that it would produce an unlocking effect that would cause many more gains on old investments to be realized. This would both raise federal revenue and create a pool of capital that would be reinvested in new businesses and industries, thus spurring growth.
After Congress cut the capital-gains tax in 1978, the Treasury Department studied the impact and concluded that the tax cut had indeed raised federal revenue. There was also a huge jump in venture-capital financing that many economists credit for starting the high-tech revolution we have witnessed over the last 25 years.
Subsequently, many so-called supply-side economists argued that there were other types of tax cuts that might also pay for themselves, and those that would do so partially, thus reducing the actual revenue loss below official estimates. Few economists today would disagree with the statement that an across-the-board tax-rate reduction would have reflows of about 35 percent. That is, static-revenue-loss estimates would be 35 percent too high. (Similarly, revenue gains from tax-rate increases would tend to be 35 percent too high.)
This is a long way from saying that all tax cuts will pay for themselves, as some overly exuberant conservatives sometimes argue. In any case, if a few extra dollars will improve the Treasury Department’s ability to analyze the effects of taxation, it is all to the good.
– Write to Bruce Bartlett here.