Thursday, November 16, 2017

Not all FRB economists know their ass from a hole in the ground


FRBSF - stock market evaluation and the macroeconomy. It pretty much illustrates why economists are so stupid when it comes to markets:

One valuation metric, the cyclically adjusted price-earnings or CAPE ratio for the Standard & Poor’s (S&P) 500 stock index was originally developed by Campbell and Shiller (1998) to help judge whether the stock market is overvalued. The CAPE ratio is computed as the real, that is, inflation-adjusted, value of the S&P 500 divided by the real earnings of companies in the index averaged over the most recent 10 years. Campbell and Shiller found that higher values of the CAPE ratio predicted lower future real returns on stocks over subsequent 10-year periods.

Yes, and Shiller today says not to put too much stock into the CAPE, but you've never listened to him, you've only read about CAPE in university.

A parsimonious set of macroeconomic variables can account for much of the movement in the CAPE ratio over the past five decades. These variables include the “natural” real rate of interest, the growth rate of potential GDP, and the core inflation rate. Reasonable projections for these same macroeconomic variables over the next 10 years can provide a prediction about future movements in the CAPE ratio, which, in turn, will influence the magnitude of future stock returns.

#1, You can't project shit without including political inputs.

#2, r-star is something you made up. It doesn't exist in the real world. Nobody knows the natural rate of interest. And you might want to look under the hood to see what r-star is derived from, as I'll show later....

#3, Whose r-star are you using? The USA's? The entire world invests in US-listed equities, not just the USA. And there's no reason to expect r-star of each nation to converge.

#4, 5 decades is a heterogenous dataset. Regressing on it will be misleading.

A simple regression model can be used to assess how well macroeconomic variables explain movements in the CAPE ratio. Specifically, the model regresses the quarterly average CAPE ratio on a constant, the Laubach-Williams (LW) two-sided estimate of r-star, the CBO four-quarter growth rate of potential GDP, the 20-quarter change in the LW r-star estimate, and the four-quarter core PCE inflation rate. The explanatory variables are each lagged by one quarter in the regression equation to help account for real-time data availability issues.

Figure 4 plots fitted values of the CAPE ratio from the regression model through 2017:Q3. The fit is quite good, accounting for 70% of the variance in the actual CAPE ratio over the past five decades.

So you're finding co-movement between the CAPE and the following:

a) GDP growth and core PCE inflation, which are highly correlated with earnings.
b) Laubach-Williams r-star, which is a made up thing, and probably uses as its inputs things that are highly correlated with earnings, GDP growth and core PCE.

No wonder your R-squared is 70%! If you want to get it higher, maybe you can try dropping Laubach-Williams r-star entirely!

Absent further changes in the ratio, stock prices can rise only as fast as earnings. Over the past 30 years, real earnings for the S&P 500 have grown at an average compound rate of about 4% per year. A decline in the CAPE ratio from current levels would imply that stock prices must grow slower than earnings.

No, it wouldn't imply that. Because despite your graduating with an MA in Finance, Kevin, you've forgotten your basic portfolio theory: stock prices are dependent on supply and demand, and wealth is growing faster than growth in available investments.

Again, here's the simple question that I always ask: where else are people going to put their trillions? Bonds? Real estate? Cash?

No comments:

Post a Comment