Are Stimulus Multipliers Higher During Times of High Unemployment? Not in the United States.

One of the arguments for additional fiscal stimulus over the past few years is that, sure, multipliers for deficit-financed stimulus are typically low enough that it's not worth doing. But when the economy is sluggish, like it is now, the multiplier effect grows larger, and the payoff for additional government spending becomes worth it. 

A trio of researchers from the St. Louis Federal Reserve, the University of California, San Diego, economics department, and the Bank of Canada decided to look at the historical evidence in both the United States and Canada to see if this might be true. And what they found was that multipliers do appear to be higher during times of slack in Canada, but not in the United States.

The research team looked at gross domestic product data, government spending, population, and the unemployment rate from 1890 to 2010 in the U.S. and 1921 to  2011 in Canada. And they tracked the difference in multiplier effects for periods of high unemployment  — above 6.5 percent in the U.S. and above 7 percent in Canada — versus periods of unemployment below those thresholds. In Canada, the authors report that multipliers appear to be a good bit bhigher during periods of high unemployment: about 1.6 compared with about 0.44 for periods below the threshold.

But in the United States, the effect is quite different. Not only are multipliers always below the 1.0 threshold where a dollar of government spending results in a dollar of economic activity, they're actually very slightly lower during high unemployment, ranging from about 0.64 to about 0.64 versus a range of 0.63 to 0.78 when below the 6.5 percent unemployment threshold. 

Are these results just an artifact of the particular threshholds picked by the researchers? They tested other threshold measures and say the results are similar. No matter how they test, the conclusion is the same. The authors say they "find no evidence that multipliers are higher during periods of slack in quarterly U.S. data from 1890 to 2010."