Scott Sumner's False Dichotomy
Anyway, he comes close to saying that you either have to believe in efficient markets or you ought to be rich. If so, then that is a false dichotomy. In 2003 and 2004, I thought that markets were out of whack in several respects. I thought that interest rates were too low and the dollar was too high. So I did not believe in efficient markets. And, to some extent, I allocated my portfolio according to my beliefs. But I did not get rich.
The moral of the story is that it is easy to believe that markets are wrong. Most recently, I spelled out my differences with the market on the outlook for long-term interest rates.
But one can believe that the markets are wrong and still not get rich. The markets can be right. Or, the markets can be wrong in ways that you did not expect. As an investor, the prudent approach may very well be to act as if markets are efficient. Load up on those stock index funds and those inflation-indexed Treasury securities, and be done with it.
Sumner points out that the 1987 stock market crash was not followed by a recession. As he says, this is bad news for the theory that stock market fluctuations cause economic fluctuations. However, it is equally bad news for the theory that stock market fluctuations represent rational responses to news about economic fluctuations. What news were investors processing in August of 1987?
Sumner also points out that macro models cannot predict sharp recessions. I can think of two reasons for this, one theoretical and one statistical.
The theoreteical reason is that modelers tend to impose constraints that force the models to have long-run properties, such as a return to full employment. The way these constraints are imposed often results in mean-reversion with a vengeance–it makes the models unable to get off trend growth by a lot or for a long time.
The empirical reason is that models tend to be dominated by recent data. Given that it has been more than 25 years since we experienced a major recession, any model that fits the recent data is bound to have pretty mild cyclical properties.
So here is where I stand:
1. Macroeconometric models are pretty useless. I’ve said that in many times and in many ways, so I won’t elaborate here.
2. Perfectly predictable financial crises should not occur. Both private sector actors and government regulators should have incentives to act on information that predicts financial crises.
3. Unlike models, individual human forecasters can make predictions that include financial crises and severe macroeconomic events. Most of those predictions will be wrong. Occasionally, a predicted crisis will come to pass.
4. I think that in 2005, even if one had known that housing was a bubble, the policy implications were not clear. We survived the Dotcom crash, so why should we not survive a collapse in the housing bubble? The real “news” in 2008 was the vulnerability of large financial institutions to the collapse of the bubble.
Even now, it is still somewhat mysterious how the housing boulder was able to cause such an enormous financial avalanche. The mainstream answer is that the financial sector had become highly leveraged and fragile. I believe that Sumner would argue that contractionary monetary policy, resulting from the decision to pay interest on reserves, caused the economic outlook to deteriorate, which in turn caused stocks to collapse.
I wonder if there is a way to test Sumner’s view against the mainstream view. Maybe we can compare how financial stocks performed in this period with their typical cyclical performance. If financials fell much more relative to the market than you would have expected, then the mainstream view has support. If financials were no more cyclically sensitive than usual, then maybe Sumner’s view deserves more consideration.