Nominal Price Targeting (A Word on Paul Ryan's Monetary Policy)
Matthew Martin 8/16/2012 09:16:00 AM
As an empirical question, the data certainly agrees with the conclusion that Kristula-Green and Thoma have made: nominal price targets cause more price stability than do moderate inflation targets. This is something I've written about before. But when it comes to explaining why this happens in the data, I think Kristula-Green and Thoma are way off base. Here is what Kristula-Green argues:
"A Federal Reserve that were to set monetary policy on the price of commodities would arguably be setting policy based on the demand for commodities in emerging economies, notably China...the Fed would risk tying itself to prices that could change rapidly and on short notice. When demand among developing countries eventually subsided, it would alter the price of commodities..."Thoma makes the same argument a bit clearer:
"But this is not a recipe for price stability as Ryan claims. Every time the price of oil, corn, or other commodities in the basket change due to ordinary fluctuations in supply and demand, which is often, the value of the dollar would change as well. This would make the value of the dollar even more unstable and uncertain than it is now..."I think that both these arguments are wrong, even if they are right in concluding that instability would increase. The fact is that foreign trade is irrelevant to this particular issue. Yes, global demand does impact US prices. But their fallacy is that they assumed that the Federal Reserve would react schizophrenically to global demand shocks--that the Federal Reserve would overreact to these shocks and destabilize the currency. There are two possibilities in the event of a positive global demand shock: one is that the demand shock is permanent, in which case the Fed would tighten policy to deflate the rest of the prices in the economy to preserve the aggregate price level. The other possibility is that it is a temporary shock, which the fed can either ignore (since the price level will return to normal soon anyway) or correct for.
This is, in fact, the exact same decision problem faced by the Fed even when there is a positive inflation target. That is, under the inflation target, if there is a demand shock that raises the price of one commodity, then the Fed will have to tighten policy to lower the rate of inflation of all the other goods. The deviation from the trend in both scenarios is identical.
The keyword that both Thoma and Kristula-Green are missing is "relative"--global demand determines the relative price, of many commodities, not their absolute levels. While it is true that absolute purchasing power parity rarely exists between developed and non-developed countries, it is true that relative purchasing power parity almost always holds, which means that exchange rates will move one-to-one with domestic price levels.
I have previously noted that the higher price volatility during the gold standard was due to the fact that while the peak inflations over the business cycle were about the same magnitude and frequency as today, under the gold standard the Fed was obliged to regularly induce similar magnitude deflations to compensate, which we no longer do now because we have a high enough inflation rate. But this is an incomplete explanation because I have not theorized why those inflationary episodes would occur. I don't really have a theory for this at the moment, but here is one possibility: a positive inflation target gives the federal reserve more room to work to stabilize prices. Here's the gist:
You can think of the Fed as either targeting the nominal interest rate, or as targeting the aggregate money supply (sorry Steve Keen, but Krugman is absolutely right), they are analytically equivalent. In the long run, the relationship between money supply growth and inflation is one-to-one. So that means that to induce higher inflation the Fed needs to lower the nominal interest rate. In the face of negative demand shocks, absent monetary intervention, prices must fall--to achieve the constant price target, then, the federal reserve must bring about an offsetting amount of inflation, which means that it must lower the nominal interest rate. But since inflation was already zero to begin with, the nominal interest rate was already extremely low to begin with, meaning the Fed quickly finds itself helpless against the zero-lower-bound in an economy with falling prices. Thus once demand returns to normal the fed must re-inflate the economy in order to return to the price level target, causing those frequent spikes in moderate inflation we saw under the gold standard following deflationary episodes.
By contrast with a 2% inflation target, nominal interest rates are generally high enough that the Fed rarely ever hits the zero-lower-bound. Hence, it can adequately compensate for demand shocks to avoid destabilizing prices.
The point is, this has nothing to do with international trade flows or exchange rates--it would be just as prominent in closed economies.