Everything Finance

Jonathan Parker

The American Recovery and Reinvestment Act, signed into law on February 17, 2009, provides for a one-time economic recovery payment of $250 to be disbursed this May to people who receive Social Security, Supplemental Security Income, or Railroad Retirement and Veterans benefits. Anyone eligible for one of these benefits at any time during the months of November 2008, December 2008 or January 2009, is eligible for the one-time payment.

Will this have a noticeable impact on consumer spending? I expect people will spend at least half of these payments within a couple of months of receiving the money. Before calculating the direct macroeconomic effect of these payments, why do I think people will spend so much?

Well, in 2001 the Federal government did what it had not done before—randomized countercyclical fiscal policy: the government sent tax rebate checks out to different households in different randomly assigned weeks.  More specifically, because it was viewed as administratively impossible to send out all the checks in a timely manner and at the same time, the government decided to randomly distribute rebate checks over a ten week period from July to September 2001.  In joint work (here) with David Johnson (of BEA) and Nick Souleles (at Penn), I studied the spending effects of these rebates by comparing the different time-patterns of spending of households who received the rebate checks at different times, much like one would in a medical trial—treating sick patients at different times and seeing whether the rates at which they recovered varied in response. As summarized in the NBER Reporter we found that people on average spent about a third of their check within a couple of months of receipt and about 2/3 within six months of receipt.  Moreover, households with low incomes or low assets responded more quickly and with more spending—those in the bottom third spent roughly the entire check within a few months. I am working now on analysis of the 2008 economic stimulus payments, and while the analysis is not complete, preliminary results suggest that people spent their checks at similar rates to 2001 at least on average. 

This evidence provides the most solid guess for what is going to happen in 2009. So, since in 2009, the targeted population is towards the lower end of the wealth and income distributions, I expect a higher average response than in 2001, but lower than the bottom of the income/wealth distribution in 2001. Since 14.2 billion payments are expected to be disbursed, this should raise consumer spending by at least 28 billion dollars at an annual rate in the second quarter of 2010, which is about 0.3 percent. Not a huge amount, but potentially a help, and probably concentrated in areas like Florida and Arizona, where there are more retirees and where the local economies are worse than the national average.  The big question – largely unanswered by economic research—is to what extent this spending leads has a multiplier effect as household spending raises the both profitability of businesses and the incomes of other households leading in turn to lower unemployment and higher capacity utilization, and to what extent this spending ends up increasing prices today and lowering resources available for tomorrow.