We Need Inflation to Spur a Recovery

Last Updated Jul 3, 2009 1:49 PM EDT

Lee Ohanian makes a very persuasive argument that zero percent inflation does not harm an economy. Sometimes that's true, but I don't think it's the case this time.

Because we are all consumers, we naturally think lower prices are good. Yet the U.S. just experienced the lowest inflation in 58 years, with the CPI falling 1.3 percent in the past 12 months. How did that feel? I thought so. There's no way one can understand why deflation (and even zero inflation) is so harmful, without considering the causes of low inflation.

Two scenarios where zero inflation is good...

There are three scenarios that could produce zero inflation. Two of them support Ohanian's benign view and one supports mine. Suppose we had zero percent inflation for years, as was the case in the 1920s. The public would get used to it, and wage and debt contracts would reflect that expectation. This type of price stability would be very good.

Or suppose we had gotten used to two percent inflation, but a tremendous burst of technological progress lowered the cost of production and brought inflation down to zero. This sort of positive "supply shock" would also be good since our real incomes would rise rapidly due to the technological progress. Both these scenarios are good, since even mild deflation can be beneficial if brought about by rapid productivity gains.

...And one scenario where zero inflation is very, very bad
Now consider a negative demand shock. Demand falls sharply, and instead of rising at the usual rate of five percent per year, nominal GDP (or national income) drops. In the short run, output falls, too, and prices stop rising and perhaps fall a bit. This type of zero inflation would cause a massive rise in unemployment. And if the banking system was already in a precarious state, it could cause a severe financial crisis. This basically describes our current situation.

Most people think low inflation is good. But they aren't connecting the trend in prices to the trend in income. It's true if your income stays constant, the lower the inflation rate the better. And it is also true if national income stays constant, the lower the inflation rate the better. But if inflation falls because national income has started falling, that is disastrous.

Deflation over the past year has been tiny, but it has in fact had a devastating effect on the economy. This is because in the modern economy, it takes a severe drop in aggregate demand to lower inflation from the normal two percent or so to negative 1.3 percent.

An increase in demand is needed
Ohanian asks how a small rise in inflation from the current near-zero level up to two percent could do much to improve the economy. And he's right that what I propose seems insignificant. But with very high unemployment, it would take a large increase in demand to push inflation up to even two percent. How large? It is hard to say, but nominal GDP (or total spending) might have to rise five or six percent, or even more.

Ohanian points out that the Fed has done a lot already, having increased bank reserves from $40 billion to $900 billion. But this liquidity injection was not what it seems -- indeed, if it was, we'd now have hyperinflation. In reality, the Fed completely neutralized the injection by starting a new policy of paying interest on reserves, causing banks to simply hoard these "excess reserves," instead of lending them out. The money never made it out into the economy, so it did not stimulate demand.

Monetary policy is obviously complex, so perhaps I should merely reiterate that whatever mechanism it chooses, the Fed's policy should be expansionary enough to boost inflation expectations for the next couple years to about two percent. I believe that such a policy would boost asset prices, and lead to a more rapid recovery and smaller budget deficit. If we don't do this, then I expect near-zero inflation and sluggish growth for the next six to 12 months. And there is still a significant risk of mild deflation. But don't be fooled by the adjective "mild" -- if it occurred, it would be very bad news for the economy indeed.

Follow Blog War on the inflation-versus-deflation debate:
  • Scott Sumner

    Scott Sumner has taught economics at Bentley University for 27 years. He studies monetary economics, focusing on the role of the gold standard in the Great Depression. He also blogs on economic issues at TheMoneyIllusion.