Consumer Spending and the Economic Stimulus Payments of 2008

06/01/2011
Featured in print Digest

Stimulus payments boosted personal consumption expenditures somewhere between 1.3 and 2.3 percent during the second quarter of 2008.

A recent NBER study by Jonathan Parker, Nicholas Souleles, David Johnson, and Robert McClelland finds that households spent an average of 12 to 30 percent of the government's 2008 economic stimulus payments on nondurable goods and another significant portion on durable goods, especially cars and trucks. In Consumer Spending and the Economic Stimulus Payments of 2008 (NBER Working Paper No. 16684), the authors conclude that, in total, consumers spent an average of 50 to 90 percent of the one-time payments they received during the three months in which they received them.

The federal government distributed $100 billion to some 130 million taxpayers as part of the stimulus program. The authors conclude that the stimulus payments directly boosted personal consumption expenditures somewhere between 1.3 and 2.3 percent during the second quarter of 2008. This direct boost to spending does not include any effects that operate through the macroeconomy, such as from the multiplier effect of fiscal stimulus or increases in prices.

The spending response on nondurable goods in 2008 was a little smaller than many of the estimates associated with a previous stimulus program, that in 2001. However, the 2008 payments were roughly twice the size of the earlier payments, individual tax filers got $300 to $600, couples received $600 to $1,200, and an additional $300 per child. Also, studies of the 2001 stimulus found little evidence of change in durable goods spending, while this study of the 2008 payments finds a significant boost in that sector, primarily in sales of new vehicles.

Generally, older and low-income households spent more of their stimulus than others. Homeowners spent more than renters.

The authors estimate the causal effect of the stimulus payments by taking advantage of questions added to the ongoing Consumer Expenditure Survey and the randomized nature of the timing of the payments. The majority of households received their stimulus payments via paper checks that were mailed over a nine-week period between early May and early July. Households who had set-up direct deposit for their regular spring-time tax refunds received their stimulus payments more quickly, over a three-week period between late April and early May. In each case, the particular timing of receipt was determined by the last two digits of the tax filer's Social Security number, digits that are effectively randomly assigned. The researchers also use a new question added to the consumer survey to determine the accuracy of consumers' self-reports of how they used their stimulus payments. Indeed, those consumers who said they spent most of their payment did spend more than those who claimed to have saved most of their payment. But the authors find that even those self-avowed savers spent a statistically significant portion of their payments. "[R]elying on self-reports can understate the actual amount of spending," the authors conclude.

--Laurent Belsie