More and more my commenters are doing the work for me. When commenting on a recent post, Joe Calhoun linked to this 2003 paper by Bernanke, and singled out these excerpts:
“What I have in mind is that the Bank of Japan would announce its intention to restore the price level (as measured by some standard index of prices, such as the consumer price index excluding fresh food) to the value it would have reached if, instead of the deflation of the past five years, a moderate inflation of, say, 1 percent per year had occurred. (I choose 1 percent to allow for the measurement bias issue noted above, and because a slightly positive average rate of inflation reduces the risk of future episodes of sustained deflation.) Note that the proposed price-level target is a moving target, equal in the year 2003 to a value approximately 5 percent above the actual price level in 1998 and rising 1 percent per year thereafter. Because deflation implies falling prices while the target price-level rises, the failure to end deflation in a given year has the effect of increasing what I have called the price-level gap (Bernanke, 2000). The price-level gap is the difference between the actual price level and the price level that would have obtained if deflation had been avoided and the price stability objective achieved in the first place.”
“A concern that one might have about price-level targeting, as opposed to more conventional inflation targeting, is that it requires a short-term inflation rate that is higher than the long-term inflation objective. Is there not some danger of inflation overshooting, so that a deflation problem is replaced with an inflation problem? No doubt this concern has some basis, and ultimately one has to make a judgment. However, on the other side of the scale, I would put the following points: first, the benefits to the real economy of a more rapid restoration of the pre-deflation price level and second, the fact that the publicly announced price-level targets would help the Bank of Japan manage public expectations and to draw the distinction between a one-time price-level correction and the BOJ’s longer-run inflation objective. If this distinction can be made, the effect of the reflation program on inflation expectations and long-term nominal interest rates should be smaller than if all reflation is interpreted as a permanent increase in inflation.”
This is exactly what people like Woodford (and I) have been advocating. During deflation a price level target is essential, much superior to an inflation target. And notice that Bernanke is advocating that the BOJ exceed their normal (zero percent inflation) price level target. In contrast, I am merely asking the Fed to adhere to its traditional 2% implicit inflation target, with a price level target that rises 2% each year. If they had adopted that policy at the meeting immediately after Lehman failed, then the past 15 months would have been very different. Instead of short term deflation leading investors to expect slower inflation going forward (which is what happened), the deflation would have led investors to expect increased inflation going forward, in order for the price level to return to its 2% growth trajectory. And those inflation expectations were exactly what was needed during the recent deflation.
Yes, I favor NGDP targeting rather than a pure inflation target, consistent with the Fed’s dual mandate. But even a price level target would have prevented most of the damage. Why? Consider the Tabarrok/Cowen AS/AD model. If inflation falls below 2% as it did in late 2008, then to get it back up to 2% you need to boost AD. But because the SRAS is fairly flat, the higher AD will not just raise inflation, it will also raise RGDP. The only problem with the price level target is that (as George Selgin forcefully points out) it gives off false signals during productivity shocks. So money is too easy when productivity is strong (say 1998-2000, or 2004-06) or it is too tight when there is an adverse supply shock, like the high energy prices in mid-2008. But it would at least prevent the most egregious errors from AD shocks.
BTW, I put “Sumnerianism” into quotation markets, because of course most of these ideas are not my own. I have added a few wrinkles, and perhaps put the pieces together in novel ways, but Kling is right that I am mostly trying to defend the standard model. Indeed as Joe’s quotations show, my views on setting explicit aggregate targets, level targeting, could just as well be called “Bernankeism.”
I would also like to add one additional quotation to those cited by Joe. I have recently received a lot of criticism for my “odd” views that a weak yuan actually helped the US economy last spring. Of course Bernanke knew that the policy he was calling for in Japan, a more expansionary monetary policy, would have tended to depreciate the yen. So why would a patriotic American call for a policy that would cheapen the yen and steal jobs from American labor? Doesn’t Bernanke want to defend our national interests first? After all, he was already working for the Fed in 2003. Here is the final portion of his essay:
Banking and structural reform are crucial and need to be carried out as soon and as aggressively as possible. Although the importance of reforms cannot be disputed, however, I do not agree with those who have argued that deflation is only a minor part of the overall problem in Japan. Addressing the deflation problem would bring substantial real and psychological benefits to the Japanese economy, and ending deflation would make solving the other problems that Japan faces only that much easier. For the sake of the world’s economy as well as Japan’s, I hope that progress will soon be made on all of these fronts. (Italics added.)
Who does that sound like? Perhaps someone at the Congressional hearings could ask Bernanke why if these policies would have been good for Japan in 2003, they would not have been equally good for America in 2008-09.