In a recent post titled Money, Inflation And Models for his NYT blog, Paul Krugman clearly states that “normal equilibrium macro models” say that the direction of causality is from money to prices. Krugman says:
Consider the relationship between the monetary base — bank reserves plus currency in circulation — and the price level. Normal equilibrium macro models say that there should be a proportional relationship — increase the monetary base by 400 percent, and the price level should also rise by 400 percent. And the historical record seems to confirm this idea.
His post is about how this fails when the economy is in a “liquidity trap”. The post hence is the clearest proof of Paul Krugman’s struggle in getting the causality right. Keynes quote from the GT is appropriate here:
The difficulty lies, not in the new ideas, but in escaping from the old ones, which ramify, for those brought up as most of us have been, into every corner of our minds.
Krugman presents a chart which shows the relationship between money and prices but it does not occur to him that the causality is reverse to what he is assuming, whether or not there is a liquidity trap. The simple causal story that a rise in the level of expenditure leads to a rise in the stock of money seems alien to Krugman.
It is of course also true that a rise in the stock of money such as via an asset purchase program by the central bank (“QE”) may have an effect on prices. This happens via a wealth effect: demand has an effect on prices of goods and services. This effect is likely small. The main mechanism — the reverse causality — seems to never occur in Krugman’s mental model of the way the world works.
Krugman has of course written about the “dark age of Macroeconomics”, but has shifted his position since then. Although it is not clear what exactly Krugman’s model is, one can still make an inference: normal equilibrium models work outside of liquidity traps, but in liquidity traps a lot changes. This model is chosen by Krugman so that he can confidently claim that there is hardly anything wrong in Macroeconomics.