As long since noted by Krugman, for the monetary authority to be able to stimulate demand when the safe short term interest rate is almost exactly zero, it has to be able to credibly pre-commit to causing high inflation, by keeping that interest rate extremely low in the future when the economy has recovered and is overheating. This is enough of a problem that the leading theorists and practitioners of monetary policy at the ZLB (obviously Krugman, Michael Woodford, Ben Bernanke Sinzo Abe and Haruhiko Kuroda *) insist that efforts to stimulate with unorthodox monetary policy be complemented by expansionary fiscal policy.
I think, however, that there is another problem. I do not find arguments including “credibility” to be credible. At the very least the word should be “credited” that is believed, not “credible” that is believable. The assumption of rational expectations has snuck into the language, which is a problem since it has nothing to do with reality. Once we understand that the issue is what people do believe and not what they should believe, we must ask which people ? My efforts to check on the effectiveness of non-conventional monetary policy have been total Fred. I look (often) at interest rates on TIPS and normal nominal Treasuries. I am finding out what bond traders seem to believe. But my interest is in the effect on aggregate demand most of which is not purchases by bond traders (a large fraction of them are rich, but there aren’t all that many of them).
I will hint at a model with three sectors non-durable goods, houses, and financial assets. Note the goods and houses are made from pure labor without any productive capital and the assets are traded on street corners or something without any bank buildings. I think this corresponds to one of the two real puzzles and problems with the US economy, which is why housing construction has not recovered (the other is why does the Republican party still exist but I have no clue on that one). Financial assets include overnight loans and 30 year fixed interest rate mortgages.
There are two ways in which expected inflation can affect demand, in both cases demand for housing. First higher expected inflation implies higher nominal wage growth so the fixed nominal mortgage payments correspond to fewer hours of work. Second house prices generally rise proportional to other prices (except for a relative decline during the depression, increase during WWII and the recent bubble). Expected house prices should move roughly one for one with the expected CPI (price of the non-durable). So if potential home buyers and home builders understand these basic stylized facts expected inflation will cause higher expected demand and a pony.
Look economists don’t know much, but we do know for sure that people do not believe that inflation causes high wages and house prices. If people are asked what’s wrong with inflation (which they totally hate) they tend to say it means you can’t buy as much. That is, they assume that inflation means higher prices but not higher wages so real wages are lower. This means that if people believe there will be high inflation, they will tighten their belts reducing demand. Similarly people other than Robert Shiller didn’t notice the almost perfect correlation of housing prices and the CPI.
I don’t really know this. I am assuming the result of a poll which I haven’t conducted. But I am willing to bet that if people are asked whether high inflation makes it rational to buy a larger and more expensive house or a smaller cheaper house more will say that inflation should cause reduced demand for housing. I am willing to make this bet mostly because I am confident the poll won’t be conducted so I won’t risk losing it. I do not think it is wise to bet the economy on the confident assumption that I have it backwards.
OK so let’s assume I’m wrong and that people will buy more and or larger houses if they expect higher inflation. Let’s assume that home builders know this, so they hire more construction workers and build more houses if they expect higher inflation. Does this mean that we can tell if policy is effective by observing asset prices ? Well we can if we assume a representative agent so there is only one subjective expected inflation rate. Not so much if we imagine that maybe the average bond trader pays more attention to hints of possible tweaks to monetary policy than the average home buyer does.
I will present a model in which H. Kuroda can influence asset prices but not aggregate demand. Above I mentioned financial assets which include overnight loans and 30 year mortgages. As that suggests, I assume that there are other financial assets whose value depends in part on what H Kuroda does. Can we determine the effect of monetary policy on aggregate demand by observing H Kuroda and those asset prices ? Obviously not.
To be more clear H. Kuroda is Hiroki Kuroda, the baseball player, not Haruhiko Kuroda, the central banker, and the financial assets are bets on the outcomes of baseball games. Yes if H Kuroda gives up, asset prices will change, But the point spread on Yankees games is not the key to recovery. Of course this is silly. Point spreads have very little to do with the real economy while every twist and turn of asset prices on Wall street is worthy of obsessive attention. Suuuure.