Delong vs. (Mankiw vs. Romer) on the Multiplier of Tax Changes
Brad Delong rang up Greg Mankiw last week, touching on something I noted from Mankiw’s post. Mankiw links to a series of his previous writings as evidence that he’s never said the stimulus was ineffective, but rather that there may have been better ways to go about spurring recovery. In his references, again and again, he cites Romer and Romer’s 2007 paper as indicating that tax cuts have a multiplier of three ($1 in tax cuts begets $3 of economic growth), comparing this to Valerie Ramey’s 1.4 multiplier for additional spending.
Delong appropriately pointed out that Mankiw was comparing the Romers’ apples to Ramey’s oranges – it’s like saying one movie reviewer’s “three stars” is better than another’s “two thumbs up.” My comment is along different lines.
In Mankiw’s shorter references, coincidentally the ones that would reach a wide audience, he notes that in January 2009, Christy Romer claimed that the multiplier for tax cuts was 0.99, and that in March 2007, she claimed it was 3. He lists these two outcomes matter of factly, not questioning the apparently significant disconnect between the two numbers.
Why? Maybe because of word count limits, he felt he didn’t have space to provide context. Perhaps he’s letting readers decide for themselves; present the data, let the reader run with it and make a conclusion.
But isn’t the point of policy-related articles and posts to influence people? Isn’t Mankiw, and Delong, and Taylor, and Krugman, and Becker and Posner and everybody else — aren’t they all trying to get people to understand and agree with their points of view? Then why wouldn’t Mankiw point up Romer’s apparent contradiction, if he was trying to say that the 2009 study didn’t show the right way out of this recession?
It would be different if Mankiw summarized the two papers, or at least indicated that the methodologies of the papers were dramatically different. But he doesn’t; he throws out the 0.99 number, then says “oh looky here, less than two years earlier the same author said the same exact thing was 3 times higher.”
Maybe Mankiw includes but doesn’t provide context because he can count on his audience to reach wrong conclusions based on his partial information. What inference would a casual reader make, other than “I guess you can’t believe anything Romer says”?
Inferences, implications, and wrong conclusions that point people down the path Mankiw wants them to go. Since he is, of course, trying to influence people.