It is well known that the Christina Romer / Jared Bernstein report, issued in January 2009 when Obama first came into office, got things exactly backwards. At the time, unemployment in the U.S. was 7 percent and change. Romer, who was the Chair of the Council of Economic Advisors to the new president, put her name on the Keynesian analysis which got things laughably wrong. The report projected that without stimulus, unemployment would break 9 percent. However, with the Obama “stimulus package,” unemployment would not quite break 8 percent. The following chart (click to enlarge) is taken from their report, but with reality superimposed on it:
As the chart shows, unemployment with the stimulus actually broke 10 percent. In other words, actual unemployment with the stimulus ended up being worse than what the Obama team warned Americans would be the case without the stimulus.
In response to this debacle, Keynesians of course have claimed that they just got the baseline wrong. Why, the economy was even worse than they had realized in early 2009, meaning that the stimulus still helped.
In policy wonk circles, the Romer/Berstein stimulus goof has achieved iconic status, epitomizing the non-falsifiable nature of Keynesian economics, even though the Keynesians are the ones who keep claiming they have all of the empirical evidence on their side. Yet we have also seen the mirror image outcome occurring with the so-called “sequester,” in which the U.S. federal government reduced spending (relative to the original baseline) because Congress couldn’t come up with a budget deal.
For example, in February 2013 Paul Krugman wrote a piece titled “Sequester of Fools,” in which he referred to the “fiscal doomsday machine” that was being unleashed upon the nation. He claimed that the sequester would cost 700,000 jobs, and justified this number by linking to a Macroeconomic Advisers analysis, which was completely Keynesian in its approach. It said (just as Krugman and the others said, in decrying the sequester) that reduced government spending would reduce economic growth. The Macroeconomic Advisers analysis first made a baseline forecast of U.S. growth without the sequester, then showed what GDP growth would be (broken down by quarter) if the sequester occurred. The bulk of the action occurs in the 2q and 3q of 2013. We are now in a position to compare the Keynesian forecasts to reality (as codified by the Bureau of Economic Analysis):
As the table (click to enlarge) shows, we have the mirror image of the stimulus debacle: Actual U.S. GDP growth in both quarters was greater with the sequester than what the Keynesians told us would be the case without the sequester.
As always, no single outcome “proves” anything in macroeconomics. Krugman and the rest can claim, “Oh phew, the economy was stronger than we realized back in early 2013. The sequester still hurt, but in absolute terms we’re doing better than we previously would have guessed.” They can also quibble with the actual fate of government spending compared to the specific projections that went into the Macroeconomic Advisers forecast. But especially with the government shutdown in the fall, no Keynesian can possibly claim that 2013 was a case of a boost in “stimulus” spending, and in fact the year was littered with Keynesians constantly warning of the dire effects of “unprecedented austerity.” The 3q growth of 4.1% is completely anomalous for the Keynesian paradigm.
The arguments over fiscal policy show the importance of Ludwig von Mises’ stress upon the primacy of theory over empirical “testing.” There are 57 moving parts in macro; a partisan can always explain away the facts, whatever happens. Yet Austrian theory shows why government spending merely transfers resources away from the productive private sector and into the inefficient political realm. We shouldn’t be surprised to see the data easily accord with the Austrian theory.