No stimulus is better than bad stimulus

Posted: Nov 29, 2001 12:00 AM
The odds that a stimulus package could now be passed in time to actually stimulate the economy before it recovers on its own are close to zero. Nevertheless, Congress continues to work on stimulus legislation, with no likelihood that final action will be forthcoming before it reconvenes in January. One problem is that new proposals keep being brought forward that take time to gestate. The latest is a plan by Sen. Pete Domenici, New Mexico Republican, to provide a payroll tax "holiday." The idea is to relieve both employers and employees from paying Social Security taxes for one month as a way of putting money into peoples' pockets. The theory is that such a tax cut could be implemented more quickly than other proposals under consideration, and it would be oriented toward those with lower incomes more likely to spend their tax savings. This is a bad idea. First, this proposal is not nearly as easy to implement as it might appear. The notion is that it could take effect faster than another round of tax rebates, which take about 2 months to send out. But in fact, it is much more complicated. Recently, a coalition of payroll professionals, whose job it is to get accurate payroll checks into the hands of workers, issued a statement saying that the payroll tax holiday would be far more difficult to implement than Domenici imagines. They say that it would take 6 months to make all the accounting and software changes necessary to make such legislation effective. By that time, most economists believe that the recession will be over. Another problem is that the distribution of payroll taxes is far more progressive than most people think. It is generally assumed that the payroll tax is highly regressive, taking much more from those with low incomes than high incomes. The reasons are that the tax applies only to wage income, thus excluding such things as interest and dividends, and because it applies only to wages up to $80,400 ($84,900 in 2002). Any wages above that amount are free of tax. Thus, when we look at effective tax rates (payroll taxes as a share of income) the tax does appear to take more from those with low incomes. According to a recent Congressional Budget Office report, those in the middle quintile (20 percent of households) pay 9.7 percent of their income in payroll taxes, while those in the top quintile pay just 6.7 percent. However, this does not mean that the distribution of the benefits of a payroll tax holiday would accrue primarily to those with lower incomes. Although they might receive more in percentage terms, those at the top end would receive far more in absolute terms. Someone earning the minimum wage would get $53 more in take-home pay for one month; someone earning at least the maximum wage covered by Social Security would get $415. According to CBO, 44 percent of all payroll taxes are paid by those in the top quintile. Thus, presumably, they would get 44 percent of the benefits of any payroll tax holiday. Moreover, this analysis doesn't take into account the fact that the Domenici proposal would apply equally to the employer's and employee's share of the Social Security tax. At present, each pays 6.2 percent of covered wages. The CBO assumes that ultimately all the tax is borne by the employee in the form of lower wages. But if the payroll tax is cut only temporarily, it is doubtful that most employers will share their lower taxes with employees. Hence, half the benefits will go to business owners, most of whom undoubtedly have high incomes. Another point worth keeping in mind is if the payroll tax holiday occurs in December, its benefits will be much more progressive than if it does not take place until January. That is because by the end of the year, most highly paid workers have already stopped paying Social Security taxes, since they are over the limit. In January, even Bill Gates would get a tax cut. Finally, the basic idea of cutting the payroll tax simply repeats the error of the earlier tax rebates. The idea is to put dollars into peoples' pockets. They then spend the money, which stimulates growth. But economic theory and experience show that people mostly save windfalls. Thus, only about 15 percent of earlier rebates were spent, according to various surveys. There is no reason to think that a payroll tax holiday will have a different effect. In conclusion, the payroll tax holiday will have no stimulative effect whatsoever -- especially if it is not implemented until May, by which time the economy will almost certainly have recovered. It also sets a bad precedent that will make fundamental Social Security reform harder. Senator Domenici should quit while he's ahead.