U.S. Fiscal Stimulus Revisited

• The stimulus that became law in February should reach its point of maximum economic benefit this summer.

• Most of the benefit so far comes from checks to state and local governments and expanded unemployment insurance benefits to workers.

• The plan is performing about as expected, but policymakers should be prepared to do more if the economy flags.

The moment of truth is at hand for the U.S. fiscal stimulus plan. The stimulus that became law in February should reach its point of maximum economic benefit this summer. If the plan is working, retailing will improve soon, and businesses should respond by curtailing layoffs measurably. Early results suggest the stimulus is performing close to expectations, but policymakers should be prepared to provide more help to the economy if things don't work as expected in coming months.

The existing stimulus includes temporary tax cuts and increased government spending worth just over $700 billion (about 5% of GDP), with the bulk of the money distributed this year and next. This includes some $250 billion in tax cuts and income support to households and unemployed workers, $150 billion in aid to state and local governments, almost $100 billion in tax cuts to businesses, and $200 billion in various kinds of infrastructure projects. This breakdown differs from the official $787 billion price tag for the plan, as it does not include the one-year cost of adjusting the alternative minimum tax, which was included as part of the stimulus legislation but would have become law anyway.

Sluggish flow of funds

Stimulus funds have flowed from Washington a bit slower than expected. Through the end of May, over $50 billion was distributed, mostly to help fill gaping state budget holes and as one-time checks to Social Security recipients in lieu of the lower payroll taxes that most working Americans are now enjoying. Spending on unemployment insurance benefits has also increased substantially as the number of unemployed and the duration of unemployment increase. Outlays on infrastructure spending have been slow to get going, but this isn't particularly surprising even for "shovel-ready" projects given lengthy contracting processes for even the most mundane public works.

What matters most for economic growth is the change in the stimulus payout, and the biggest increases in the stimulus payout are occurring right now. The payout was near $10 billion in the first quarter; it is expected to rise to $80 billion in the current quarter, remain effectively unchanged through this time next year, and then fade quickly after that. The impact on the economy is not immediate, as measured by real GDP, employment and unemployment. The extra take-home pay that workers saw after payroll taxes fell, for example, won't be spent for several months. If the 2001 and 2008 rebate checks are a guide, it will take through the end of this year for workers to spend about two-thirds of their extra cash.

Accounting for these lags, the maximum contribution from the stimulus should occur in the second and third quarters of this year, when it will add more than 3 percentage points to annualized real GDP. This suggests that if policymakers had not been able to pass a stimulus plan, real GDP would have declined nearly 6% in the second quarter and by more than 3% in the third. With the stimulus, GDP is expected to fall close to 3% in the second quarter and rise a bit in the third. The contribution of stimulus to growth fades quickly, adding just over 1 percentage point to annualized growth in the fourth quarter of this year and the first quarter of 2010 and actually detracting from GDP growth by the second half of 2010. The impact on jobs and unemployment is also significant, as the stimulus results in approximately 2.5 million more jobs by the end of 2010 than would have been the case without it, and leaves the unemployment rate almost 2 percentage points lower.

It is important to note that estimating the economic impacts of the fiscal stimulus is not an accounting exercise but rather an econometric one. It is not feasible to identify and count each job that results from the stimulus; economic impacts are estimated using a macroeconometric model, a statistical representation of the U.S. economy based on historical relationships. The economic impact of the stimulus is determined by simulating such a model with fiscal stimulus and then without it. The results presented here are based on the Moody's Economy.com model, which is used regularly for forecasting, scenario building and policy analysis. The Obama administration has derived its estimates of the stimulus' impact using a similar approach.

To date, most of the benefits from the stimulus plan have gone to state and local governments to pay for Medicaid and educational programs and expanded unemployment insurance benefits. This stimulus is defensive—it helps forestall draconian cuts in government services or tax increases that would have otherwise occurred. In the nomenclature of the debate surrounding the merits of the stimulus, this preserves jobs. A bigger economic kick from stimulus should be seen this summer, when the stimulus goes on the offensive and begins to create jobs, via personal tax cuts and infrastructure spending. First-time homebuyers also can also receive a tax credit worth up to $8,000 if they close on a home purchase before December 1—meaning they will need to shop this summer and sign a contract not long after Labor Day.

Are shovels ready?

A good gauge of whether the stimulus is working as expected will be whether infrastructure outlays increase significantly over the next several months. If they don't, the benefits will likely be delayed until next spring as winter will forestall many projects. There is also a growing concern that state and local government officials may push projects into next year so they ramp up just as political campaigning starts for the 2010 election. Better home sales will also be a positive signal. The sharp recent increase in fixed mortgage rates—they've jumped almost a percentage point—is particularly bad timing.

But the most important sign of whether the stimulus is working is if job losses slow substantially in coming months. Monthly employment losses have already slowed from close to 700,000 in the first quarter to 500,000 in the second quarter. The stimulus plan will be on track if the job losses moderate further to near 300,000 in the third quarter, 100,000 in the fourth, and cease altogether by next summer.

Risks to this sanguine script are skewed to the downside. Odds remain uncomfortably high that the economy will enjoy a bounce from the increased stimulus this summer but then fade with the waning stimulus by the summer of 2010. This scenario is more likely if the administration's foreclosure mitigation efforts don't quickly begin to reap benefits. Without a measurable increase in mortgage loan modifications, foreclosures will continue to surge, further undermining house prices, housing wealth, the financial system, and the economy's prospects for a sustainable recovery.

Prepare for the worst

Policymakers should thus be quietly preparing another round of fiscal stimulus for early 2010. Effective additional stimulus might include more help to state and local governments, whose budget problems will probably be even worse next year; an expanded housing tax credit to address the foreclosure crisis; and a payroll tax holiday. Delaying increases in marginal personal tax rates, now legislated to occur at the start of 2011, would likely also make sense. Higher-income households may begin to rein in spending in 2010 as they prepare for the higher tax rates.

It is premature for policymakers to publicly consider all this now; the current stimulus should be given a chance, and the nation's long-term fiscal challenges are daunting. But if the Great Recession has taught us anything, it is to prepare for the worst.