Sep 27 2009, 5:56PM
The Psychology of Stimulus
Anyone who still thinks that stimulus expenditures in the second quarter of this year increased GDP or employment would do well to read John F. Cogan, John B. Taylor, and Volker Wieland, "The Stimulus Didn't Work," in the September 17 issue of the Wall Street Journal. They note that, consistent with theory, the transfer payments (the major component of the stimulus that was actually executed in the second quarter) appear not to have resulted in any measurable increase in personal consumption expenditures, as they constituted transitory income and therefore were largely saved; and that the modest stimulus spending on investment in that quarter probably had no effect. Rather, they attribute the reduction in the rate of decline of GDP in the second quarter to military spending unrelated to the stimulus and to a decline in the rate at which business investment was declining that began in January, before the enactment of the stimulus law in February.
Their argument, if understood as I think the authors intended it to be understood as a root-and-branch criticism of Keynesian deficit spending as an anti-depression weapon, is not as compelling as it may appear to be. The fact that personal consumption expenditures didn't increase after the stimulus program was enacted is inconclusive, because, had it not been for the transfers, they might have fallen (though as I have emphasized in previous blog entries the initial stimulus spending was so limited that it is doubtful that it could have had much effect on consumer expenditures). The fact that the increase in military spending was unrelated to the stimulus law doesn't mean it wasn't an effective form of stimulus. And the fact that the rate of decline in business investment slowed in January may have been in anticipation of the stimulus, as it was certain by then that there would be a stimulus program.
The narrow criticism of the Administration's touting the successes of the stimulus program is that it emphasizes actual expenditures, which have been as yet (and certainly during the second quarter, when the rate of decline of GDP fell markedly) both modest and heavily weighted to transfers; and that criticism, which I have emphasized in my blog entries, I continue to believe is sound. The program may, however, still have had an important positive effect on business and consumer psychology. Economists both left and right systematically neglect the psychological dimensions of a depression, properly emphasized by Keynes.
An exception, however is Daniel Indiviglio, who is not an academic economist, and who in his blog entry argues that:
"Perhaps knowing that the government was throwing $787 billion at the economy in order try to reduce the pain of the recession helped the sentiment of business as well. Maybe businesses decided that the economy can't possibly continue to suffer given such extraordinary government intervention, so built more plants, ordered more equipment and ramped up inventories in the hopes of imminent recovery built on that government action."
In the same vein, with regard to the transfer payments, he argues that the "money must be going somewhere, so where is it going? Maybe it's being used to pay down debt; maybe it's being used for investment; or maybe it's just being saved. I would argue that, though not consumption, those are still actions that ultimately help a stumbling economy get a little healthier. Having more money in your pocket certainly makes you feel better, and consumer sentiment matters a lot during a recession, even if that doesn't translate to immediate consumption. Maybe people would have saved even more and spent even less without the payments, for example."
Those are excellent points.
Comments (3)
Comments on this entry have been closed.
Richard A. Posner
I tend to see the stabilization of employment levels as a much more potent confidence booster than merely a short-term increase in GDP or consumption. So far, this primary psychologically encouraging development has yet to be achieved, which makes "it hasn't worked" criticism of this particular implementation of a massive stimulus program fair.
Lets say the government wanted to focus on employment levels by subsidizing new hiring. If the average employer needed an extra $40000 a year in this uncertain economic environment to justify an additional hire, then at $787 Billion, the stimulus could have kept 5 million extra men employed for 4 years. That could have virtually arrested the spike in unemployment - and the worry that one might lose his job is a significant contributor to hoarding behavior.
Maybe we need a better defined "mission statement" of what a stimulus is supposed to accomplish so we can better assess whether or not "it worked".
For instance, if a stimulus is merely supposed to increase short-term consumption, then the cash-for-clunkers program worked very well. From July to August, light automobile sales increased from less than 11 to over 13 million at the Seasonally Adjusted Annual Rate (SAAR) for the cost of around $3 Billion.
However, September sales are estimated to fall below 9 million, the worst per-capita sales figure in over 40 years. It appears that the project mostly pulled future demand forward and destroyed a good number (about 750,000) of last-leg used-cars. Did this stimulus "work" or not? Did it leave us worse off?
If we could have forecast these results earlier this year, would that have been part of a coherent argument against certain kinds of stimulus? If so - do the limits of the "what you shouldn't do" argument encompass many of the other efforts attempted in the current stimulus plan?
I might add, the idea of a tax credit for new-hires as a form of stimulus that directly attacks employment levels is starting to make the rounds:
See http://www.nytimes.com/2009/10/07/business/07tax.html
It is too bad such an employment-focused effort comes so late while ridiculous and inefficient automobile and housing giveaway programs passed quickly early on. Hiring credits even help to pay for themselves since they are offset by the otherwise required expenditures of unemployment insurance payments.
In the mean time, the US has lost 8 million net jobs since December 2007 - already more than the population of Switzerland and about 12,000 per day on average. How effective - in spurring the consumption level of overcautious and rationally paranoid workers - can any deficit spending program be without having first solved this problem?
Indiviglio's argument that the stimulus helped sentiment which in turned helped the economy ignores the fact that the administration said that the stimulus would prevent the unemployment rate from rising above 8%. The rate is now at 9.8% and appears to be headed higher. I would argue that as people have watched the unemployment rate climb well above 8% any beneficial psychological benefit of the stimulus has worn off and has likely termed negative. People may now be concluding that stimulus was a failure because it did not accomplish what the administration promised that it would do. If that is the case then consumers will reduce consumption and businesses will reduce investment.