Daily Academic Alpha: Warren Buffett Market Predictions By Wesley R. Gray, Ph.D., Alpha Architect, Quantitative Value: A Practitioner’s Guide to Automating Intelligent Investment and Eliminating Behavioral Errors.
Last week we had a fairly long post on a valuation based asset allocation strategy that might actually work. This post followed a couple of other research projects on the issue, which showed limited evidence for simple valuation-based timing strategies.
Now there is a new paper on Warren Buffett’s favorite timing mechanism, Market Cap/ GNP (or GDP). We’ve discussed this metric in 2013, which suggested one get out of the market, and the out-of-sample results were horrific–the market has been on a tear!
Below is a link to the new paper and abstract (h.t. CXO advisory).
Select quotes from the paper:
Our analysis shows that Warren Buffett’s market value of all publicly traded securities as a percentage of GNP (MV/GNP), and its parent the lorgarithm of the market value of all publicly traded securities as a percentage of GNP (lnMV/GNP), can be a statistically significant predictors of future market downturns. However, for these measures to work, we need to use time-varying confidence-based thresholds rather than fixed thresholds.
The authors continue:
This conclusion dispels a common myth about the MV/GNP ratio: that absolute level matters. This myth has led market commentators and investment practitioners to suggest that the level of the MV/GNP is the harbinger of an impeding market meltdown.
The authors examine a bevy of measures:
- MV/GNP with a fixed threshold at a 120% level;
- MV/GNP with a threshold computed using a standard 95% one-tail confidence interval based on a Normal distribution;
- MV/GNP with a threshold computed using Cantelli’s inequality;
- logMV/GNP with a fixed threshold at a 120% level;
- logMV/GNP with a threshold computed using a standard 95% one-tail confidence interval based on a Normal distribution;
- logMV/GNP with a threshold computed using Cantelli’s inequality;
And here are the core results, highlighting that absolute measures stink, but time-varying metrics may be promising:
Conclusion: The results are mixed (absolute doesn’t work while time-varying works) when it comes to tactically allocating assets based on market valuations. This result is in line with much of our own research.