Payroll Tax Holiday a Poor Stimulus Idea
Making Work Pay" Credit a Better-Targeted Alternative
January 27, 2009
Note: this paper analyzes a 2009 proposal by Sen. McConnell to suspend employer and employee payroll taxes for two months. For the Center’s analysis of a different proposal, to extend and possibly expand the reduction in employee payroll taxes in effect for 2011, click here.
This week Senate Minority Leader Mitch McConnell suggested suspending the Social Security payroll tax for a period of time, as a stimulus measure. A payroll tax holiday, however, would both be costly — a two-month suspension could cost about $120 billion, for example — and likely relatively ineffective as a stimulus measure. Public resources would be better spent on stimulus measures with a higher “bang for the buck,” such as the Making Work Pay tax cut that President-elect Obama has proposed.
Biggest Tax Benefits from Payroll Tax Holiday Would Go to Workers Least Likely to Spend Them
Economic stimulus measures aim to encourage an immediate increase in aggregate demand by boosting consumer spending. The most efficient way to boost consumer spending is to put money into the hands of people who will spend it quickly rather than save it; tax cuts focused on moderate- and low-income households are more effective as stimulus than tax cuts that are larger for people with higher incomes, because people at low-income levels spend a larger share of tax cuts they receive than people at higher income levels do.
A payroll tax holiday does not score well on this front — too little of the benefit goes to lower-income households struggling to make ends meet and too much goes to higher-income taxpayers, who are likely to save a significant fraction of any new resources they receive. Under the payroll tax, employees pay tax of 6.2 percent on earnings up to $106,800. So, for example, a worker earning $10,000 would receive a tax cut of just $103 from a two-month payroll tax holiday, while a worker earning ten times as much ($100,000) would receive a tax cut ten times as big — $1,030. Indeed, the highest-income fifth of households could receive more than half of the benefits that would go to workers from a two-month payroll tax holiday.
A better way to boost consumer spending through a tax cut is President-elect Obama’s proposed Making Work Pay credit, which would offset the worker’s share of payroll taxes for the first $8,100 in earnings. Worth up to $500 per worker, it would cost about $70 billion in 2009. This is roughly equivalent to the revenue from a one-month payroll tax holiday.
The Making Work Pay credit would likely provide considerably greater bang for the buck than a payroll tax holiday for two reasons. First, a substantially larger share of the benefits would go to low- and moderate-income workers. Unlike the payroll tax holiday, the credit would provide the same benefit — $500 — to a worker earning $10,000 as to a worker earning $100,000. Moreover, that $500 credit would far exceed what low-income workers would receive from a payroll tax holiday. A worker earning $10,000 would receive only $103 in tax benefits from a two-month tax holiday even though such a payroll tax holiday would be nearly twice as expensive. Indeed, for a worker to receive more from the employee share of a two-month payroll tax holiday than they would from the Making Work Pay credit, their income would have to exceed $48,387.
Second, as explained below, half of the cost of the payroll tax holiday would go for a tax break for firms — which would likely retain much of the tax break rather than spend it. That would sharply reduce the “band for the buck” of a payroll tax holiday and make it an inefficient and relatively ineffective stimulus measure.
Half of Tax Benefits Would Go to Businesses as Inefficient Windfall
The proposed payroll tax holiday would apply to the employer’s share of the payroll tax as well as the employee’s share. (Each share amounts to 6.2 percent of eligible wages.) Suspending employees’ payroll taxes would immediately translate into higher take-home pay for workers. Suspending employers’ payroll taxes, by contrast, would put cash into companies’ coffers, where it is likely to sit as long as sales are weak and factories are operating below full capacity. Indeed, according to the Congressional Budget Office, suspending employer’s payroll taxes is “not a particularly cost-effective method of stimulating business spending: Increasing the after-tax income of businesses typically does not create an incentive for them to spend more on labor or to produce more, because production depends on the ability to sell output”. In other words, firms will not hire (or retain) more workers than it takes them to produce the goods and services they can sell. Simply giving them a general tax break is unlikely to affect their hiring or investment in most cases, and thus would be largely ineffective as stimulus.
Standard economic analysis suggests that over the long run, a permanent reduction in the employer payroll tax would increase wages, as competition forced employers to pass on the benefits of the tax cut to their workers. But a two-month holiday on the employer share of the payroll tax would not have that effect: according to the Congressional Budget Office, “[s]uspending the employers’ portion of the tax for a short period of time is unlikely to alter wage rates by very much and so would not alter consumers’ resources very much.” Firms generally would not raise wages for two months and then cut them, and the reduction in wage costs would be too brief to make it worthwhile for employers to increase hiring. Instead, businesses would likely retain all or nearly all of the benefits from the tax holiday.
Would infusing cash into businesses in this manner constitute effective stimulus? Probably not. The primary problem that employers face in a recession is a shortage of demand for their products, not a shortage of cash. Therefore, most firms would likely keep much or all of any tax windfall they receive — or pass it on to shareholders and business owners, two groups that tend to have higher incomes and thus quickly spend relatively little of any additional income they receive.
To be sure, the current lack of liquidity in credit markets could make an infusion of cash more valuable to some companies that would, in past recessions, have been able to borrow to meet payrolls and make needed investments to maintain their capital stock. The problem is how to focus a tax cut on businesses that would actually spend an infusion of cash; without an effective targeting mechanism, too much of the expenditure on a business tax cut would be retained rather than spent and hence not have a stimulative effect.
Such mechanisms are hard to design, but a holiday for the employer share of payroll tax would be a scattershot approach and not constitute an effective or well-targeted way of addressing this liquidity concern. Mechanisms such as well-designed business tax incentives tied to new investments and business purchases, or tax credits for new hires, would at least attempt to deliver liquidity in a targeted way and could be a step in the right direction. Other possibilities are zero-fee loans administered by the Small Business Administration or loans via the Troubled Asset Relief Program or a similar mechanism. All of these should be more effective than a holiday on the employer’s share of the payroll tax in targeting assistance so that it is more likely to generate spending that firms would not otherwise undertake. Moreover, the public would ultimately be able to recoup a portion of the value of any such loans.
 Policymakers have proposed payroll tax holidays of varying durations. The 2008 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds estimated that OASDI payroll tax collections in calendar year 2009 will total $723 billion, so the pro-rata share for a two-month holiday would be about $120 billion. However, the exact cost would depend on when in the year the holiday occurred and how high earners were treated (as well as the duration of the holiday). A tax holiday late in the year would cost less because it would not benefit people who had already exceeded the maximum earnings subject to the payroll tax.
 The Urban-Brookings Tax Policy Center estimated that in 2006, 51.2 percent of payroll taxes were paid by the top 20 percent of tax units. The exact percentage of the value of a payroll tax holiday flowing to upper-income taxpayers would depend on the timing of the holiday; as explained in footnote 2, upper-income taxpayers would receive a smaller share of the overall tax cut if the tax holiday were implemented late in the year.
A payroll tax holiday could be designed so that people with annual earnings somewhat above the payroll tax cap (i.e., above $106,800) receive a partial tax cut and those with earnings substantially above the payroll tax cap receive no tax cut. This could be done by requiring that amounts earned during the holiday period not count against the payroll tax cap. Doing so would add some complexity to this provision, and bills that have been introduced to institute a payroll tax holiday do not contain this feature.
 Workers earning $100,000 may in fact receive less than $500, depending on the income level at which the credit begins to phase out.
 Congressional Budget Office, “Options for Responding to Short-Term Economic Weakness,” January 2008, p7. http://www.cbo.gov/ftpdocs/89xx/doc8916/01-15-Econ_Stimulus.pdf.
 Congressional Budget Office, “Options for Responding to Short-Term Economic Weakness,” January 2008, p11. http://www.cbo.gov/ftpdocs/89xx/doc8916/01-15-Econ_Stimulus.pdf.
 See Congressional Budget Office, “Economic Stimulus’ Evaluating Proposed Changes in Tax Policy,” 2002.
 No perfect measure would exist for such a tax credit: there will inevitably be some expanding firms that would have made the new investment or new hire even without the credit, and there will inevitably be some contracting firms that would have laid off fewer workers or cut back less on investment if they had been eligible for the credit.