Thanks for the insightful comments – I will have a quick go at listing down a couple of my thoughts with regards to them.

@Nick Rowe

“But what determines the equilibrium is not whether forms *really do* set q or p in advance, but whether the other firms *think* they do.

From my cursory reading, this strategy space problem is pervasive in game theory.”

Agreed, as far as I understand it the folk theorem and the strategy space issue are two of the biggest methodological issues in game theory. However, we have to ask what we do when faced with a methodological quandary – do we ignore it, or do we try to take a step back and see if we can endogenise the process that gets us there.

The solution to the Cournot-Nash vs Bertrand game was to take an extra step back – and make the strategy space endogenous.

Now in the case of monetary policy we do hit a dilemma, because the choice of instrument is exogenous. However, before we state that there is a framing problem we need to ask whether changing policy really matters.

“If inflation expectations become unanchored will framing policy in terms of interest rates still work? And what about the US Fed, which doesn’t really have an inflation target?”

I agree here that inflation expectations ARE the central issue. Ultimately, best policy involves a central bank blatantly targeting inflation expectations.

If we do reach a situation where “inflation expectations are unanchored” this means that when the bank prints money (which it can still do when the interest rate hits a zero bound – and it can still communicate, by saying it is going for inflation) it just turns into inflation, and we get no recovery in output.

However, here the real issue is that inflation expectations are become unanchored. In this case we need to ask how the hell that happened – and how prices became so much more responsive to contemporaneous changes in the money supply.

I do agree with you that, if banks only mentioned nominal interest rates, and they hit zero, and as a result this managed to unhinge inflation expectations then we have a problem – however, I was under the impression that inflation expectations were still anchored, and that if there is an issue it is a lack of printing action from central banks.

“Why aren’t things like share prices or commodity prices “appropriately related to “monetary policy””?”

@Nick Rowe

The key term for me here is “appropriately”.

Inflation expectations, the money supply, nominal interest rates – these are all factors that are determined on the basis of monetary policy directly. The idea of using these variables is to ensure so type of nominal change. As a result, they seem appropriate.

Commodity prices, the stock market, both strongly influenced by monetary policy for sure. However, they are also separate – supply shocks should have a direct influence on these variables. Do we want the public to view changes in commodity prices and the stock market as an effective lead on monetary policy?

Say we have a drought (negative supply shock), commodity prices rise, the public looks at this as a defacto signal of tighter monetary policy as well as a normal supply shock – this leads to downward pressure on the price level exacerbating the problem. New Zealand has the occasional issue with drought, and so this sort of issue concerns me.

Sure, economists are unhappy with the Cournot/Bertrand result (that the equilibrium can depend on something as ephemeral as the strategy space). And applying some more structure to the game can seem to resolve this problem (suppose that firms *really do* have to produce output in advance, or *really do* have to advertise prices in advance). But what determines the equilibrium is not whether forms *really do* set q or p in advance, but whether the other firms *think* they do.

From my cursory reading, this strategy space problem is pervasive in game theory.

Back to monetary policy. I wrote in some earlier post, something like “We always knew that setting interest rates couldn’t work in theory, but it seemed to work in practice, so eventually we stopped worrying about the theoretical problem”. *Provided* that inflation expectations remain well-anchored, I think we can understand that interest rate policy can work. But what happens if people lose faith in central banks’ ability to keep inflation on target? (And over the last year, that loss of faith was a very real danger). If inflation expectations become unanchored will framing policy in terms of interest rates still work? And what about the US Fed, which doesn’t really have an inflation target?

Here’s where I really disagree with you though: “Targeting arbitrary variables that are positively related to an economic recovery, but not appropriately related to “monetary policy” seems both sort of aimless and potentially dangerous.”

Why aren’t things like share prices or commodity prices “appropriately related to “monetary policy””? Go back 100 years, when “monetary policy *meant* fixing the price of gold (a commodity price)! You have been so captured by the current way of framing monetary policy that you can’t see that it is just one of many possible framings! Escape The Matrix!