Senate Changes Make Recovery Package Less Effective
The Senate today passed a version of the American Recovery and Reinvestment Act (H.R. 1) that makes a number of changes in the House-passed bill. Contrary to their proponents' claim, these changes — in Senate committees and on the floor last week, as well as by a group of Senators led by Ben Nelson and Susan Collins — have reduced the package’s effectiveness as economic stimulus. Though it costs modestly more than the House bill — $838 billion, compared to the House bill’s $819 billion — the Senate package would likely preserve or create tens (or possibly hundreds) of thousands fewer jobs.
The Senate has reduced spending, a fair amount of which was well-designed to stimulate the economy such as funding for state fiscal relief and school construction, and substituted new or expanded tax cuts that are not targeted and are unlikely to provide a substantial boost to the economy.
These changes fly in the face of the consensus of mainstream economists about how best to provide the boost in aggregate demand that is needed to help stem the current economic downturn and speed a recovery. Those economists conclude that:
Well-designed spending measures tend to be significantly more effective than tax cuts in stimulating aggregate demand because much of tax cuts are saved rather than spent; and
Tax cuts are most effective as stimulus when they are targeted on low- and moderate-income households that will likely spend a high proportion of the benefits rather than on high-income taxpayers, who will likely save a large proportion of the tax benefits, or to businesses that are unlikely to spend the tax to expand capacity or hire workers when their sales are depressed.
This paper focuses on three tax cuts added or expanded by the Senate — a greatly expanded homebuyer credit, deductions related to new car purchases, and relief from the Alternative Minimum Tax — that would cost a total of about $116 billion over the next 11 years but do little to boost the economy. None of these provisions would provide more effective stimulus than most of the spending provisions that were reduced or eliminated.
The paper also addresses a provision in the House bill — the repeal of a significant provision enacted in 2006 that would help put a dent in the more than $300 billion a year “tax gap” — that would not stimulate the economy and would permanently set back efforts to collect more of the federal taxes that are owed.
The House bill includes a provision that would amend a temporary first-time homebuyer credit enacted last summer waiving the requirement in last summer’s legislation that the taxpayer repay the credit over time. (That requirement would still apply if the house is sold within three years of purchase.) The Senate Finance Committee recommended a similar provision and modestly extended the period over which the credit would apply. According to the Congressional Joint Committee on Taxation, the House provision would cost $2.6 billion over 11 years, while the Finance Committee provision would cost $3.7 billion over that period.
During Senate consideration of the recovery bill, Senator Johnny Isakson successfully offered an amendment that greatly expanded this provision by doubling the size of the tax credit (from $7,500 to $15,000), making it available for any purchase of a principal residence (not just purchases by first-time homebuyers), and removing income limits on eligibility for the credit. These changes increased the cost of the provision by more than $35 billion (or by about ten times), to $39 billion over 11 years. The additional $35 billion would have very little effect on the economy, because the bulk of the money — probably more than 90 percent of it — would go to taxpayers who would have bought a new house anyway. In addition, the people most likely to benefit from the credit would be among those who are suffering the least from the recession, as they would need to have sufficient current income to qualify for a new mortgage.
Since the credit would have a limited effect on the number of home sales (and thus also would not lead to a large increase in purchases of home furnishings and other items new homebuyers might make) and most of the benefits of the credit would go to people who do not have low- or modest-incomes and thus are likely to save a large portion of the tax cut, the $39 billion would have very little bang for the buck as stimulus. Most of the $39 billion would be saved, rather than spent, as homebuyers withdrew less from savings, or took out a smaller mortgage, to make the home purchase.
Deductions Related to New Car Purchases
The Senate also adopted an amendment that Senator Barbara Mikulski offered to provide tax benefits for purchases of new vehicles between November 12, 2008, and January 1, 2010. Taxpayers would be allowed to deduct interest paid on loans related to these purchases. In addition, the amendment would allow a taxpayer to deduct state sales and excise taxes related to such a purchase. The Joint Committee on Taxation estimates this provision would cost $11 billion over 11 years.
Like the homebuyers provision, the deductions related to new car purchases would overwhelmingly be claimed by taxpayers who would have bought a new vehicle anyway, so the provision will generate little in the way of new economic activity. Also like the homebuyers provision, the bulk of the benefits would go to taxpayers who are not among those struggling the most to meet the day-to-day needs of their families and who are thus likely to save a substantial portion of the new tax benefit.
Moreover, because the new subsidy is structured as a deduction, it is only available to filers with positive income tax liability, and the size of the benefit would be much larger for those with higher incomes — the people who save more and spend less of their tax cuts — than for those with modest incomes. Many low- and moderate-income families would not benefit at all from the credit (a family of four with two children would benefit from the provision in 2009 only if it had income of at least $26,000), and many moderate-income families will receive only a fraction of the full benefit.
A family of four with two children and an income of $40,000 would see its tax reduced by $150 if it could claim a deduction of $1,500 under these provisions in 2009.
In contrast, a family of four with two children and an income of $250,000 would receive a tax benefit of $500 even if the deduction claimed is the same $1,500. And the difference in tax savings between lower- and higher-income taxpayers is likely to be much greater because the upper-income family is much more likely to buy a more expensive car and have larger interest and tax payments to deduct.
Alternative Minimum Tax Relief
At the recommendation of the Senate Finance Committee, the Senate added a provision to the recovery package that would extend relief from the Alternative Minimum Tax through the end of 2009. This provision would cost nearly $70 billion over the next 11 years (all of the effect would occur in 2009 through 2011).
This provision should not be viewed as adding any stimulus to the recovery package since there is no question about AMT relief being provided for 2009 whether or not the provision is included in the stimulus legislation, just as Congress has ensured that AMT relief has been provided for every year since 2001 and clearly will do so again. (Some of the extensions have covered two years, others just one.)
If the Senate merely added AMT relief to the stimulus package and let the total cost of the package increase, there would be no great harm since AMT relief will eventually be provided in separate legislation if it is not included here. But adding the AMT relief to the package and then reducing effective stimulus spending — such as state fiscal relief — in order to hold the total cost of the bill to a total cost limit (such as $800 billion) has the effect of substantially reducing the amount of real stimulus that results from enactment of the package.
The reduction in the stimulative effect of this legislation resulting from substituting AMT relief for effective stimulus spending is clear not just from the fact that AMT relief would happen whether or not this stimulus package is enacted, but also because AMT relief provides very little stimulative effect in any case.
AMT relief would be ineffective as stimulus because its benefits go overwhelmingly to higher-income taxpayers who likely will save a large proportion of the tax benefits they receive, rather than to families at more modest income levels that are likely to spend most of the benefit and help provide a boost to aggregate demand. The Brookings Institution-Urban Institute Tax Policy Center estimates that 80 percent of the benefits of AMT relief would flow to households in the top 20 percent of the income distribution. The bottom 60 percent of households would receive just over 2 percent of the benefits of AMT relief. Nearly 90 percent of the benefits would go to taxpayers with income of more than $100,000.
These figures led the Tax Policy Center to conclude that the AMT provision “would provide virtually no economic stimulus,” and that it “makes no sense as economic stimulus.”
Repeal Withholding on Government Payments
Although the Senate added a number of inappropriate tax provisions to the recovery package, the House bill contains at least one provision that clearly has no place in a stimulus package (or in any other bill for that matter). The House bill repeals the requirement for the federal, state, and local governments to withhold 3 percent of payments they make to companies for goods and services they purchase. (Governmental entities that make payments of less than $100 billion a year are exempt from the requirement.) The Joint Committee on Taxation estimates this would cost $11 billion over the next 11 years.
This provision was enacted in 2006 as a way to begin chipping away at the large federal tax gap — the gap between taxes owed to the government under current law and what is collected — that is estimated to exceed $300 billion a year. In response to a request from the Senate Finance Committee, the Joint Committee on Taxation in 2005 produced a report on “Options to Improve Tax Compliance and Reform Tax Expenditures.”  The very first recommendation in that report was to introduce withholding on government payments to vendors and service provider because of the lower-than-average tax compliance of such companies. The withholding requirement was enacted in 2006 and is scheduled to go into effect in 2011.
The Senate bill delays the implementation of the withholding requirement for one year, at an estimated cost of $291 million. While it may be appropriate to give state and local governments an extra year to implement the withholding requirement given the stress they are under during the current downturn, it makes no sense to repeal the withholding provision in any legislation, much less in a bill that is intended to stimulate the economy.
The withholding tax does not impose any new tax liability. It is merely a measure intended to ensure that taxpayers who do business with the government actually pay the taxes they already owe. Repealing the withholding requirement would effectively allow some businesses to avoid paying taxes they owe. Compliant businesses would receive little benefit from the measure and would be at a competitive disadvantage compared to companies that do not pay the taxes they owe.
Furthermore, the House repeal provision is a permanent provision in what is supposed to be a temporary economic-recovery package. It would permanently worsen the nation’s already bleak long-term fiscal outlook.
State and local governments may complain that the withholding requirement is a burden on them. But it does not seem unfair for the federal government to expect state and local governments to take such a modest step to help ensure taxes owed to the federal government are collected, particularly when the recovery package is providing states and local governments with substantial funds to help meet their budget shortfalls over the next 2 ½ years.
 See Chad Stone, "Attacks on Congressional Recovery Package Don’t Withstand Scrutiny,” Center on Budget and Policy Priorities, February 5, 2009.
 Proponents claim that the new credit would increase home sales by nearly 500,000 in the first year after enactment. Since home sales are likely to total around 5,500,000 in 2009 without the new credit, the home sales prompted by the new credit would represent only a little more than 8 percent of total sales if the proponents are right about its impact. That means that more than nine-tenths of the people eligible for the credit would have purchased a home without the credit. See Barbara Sard and Robert Greenstein, “Costly Isakson Homebuyer Tax Credit Amendment Would be Ineffective Stimulus,” Center on Budget and Policy Priorities, February 9, 2009.
 The amendment limits the amount of the deduction to interest paid on no more than $49,500 in car loans, but it seems likely that loans exceeding $49,500 could be structured so the interest paid on the first $49,500 is higher than the interest paid on the remaining amount, so that the benefit of the credit could effectively extend to amounts borrowed in excess of $49,500.
 The amendment also imposes a $49,500 limit. The limit presumably was intended to apply to taxes paid on the first $49,500 of the purchase price of the vehicle, but it appears to limit the deduction itself to $49,500.
 These examples assume that the families take the standard deduction and do not qualify for other deductions or credits other than the child tax credit.
 See Tax Policy Center Table T09-0064: http://www.taxpolicycenter.org/numbers/Content/PDF/T09-0064.pdf
 Tax Policy Center, “Tax Stimulus Report Card: Senate Bill,” http://www.taxpolicycenter.org/taxtopics/senate_AMT.cfm