Scientific Investing IV: Putting it all together
Updated: Jul 29, 2022
If we put all of the insights from the last three “Scientific Investing” blog posts together, the reader arrives at a point of academically clean, evidence-based investing. This is reflected in portfolio structures which adhere to all the insights of Modern Portfolio Theory (MPT).
The obvious question an investor should ask herself is whether the premiums of the dimensions of expected returns are in fact sustainable and what they all added up to over 10-year rolling periods from 1928 until today.
As described in a previous blog article, Professors Eugene Fama and Kenneth French developed the 3-factor model, which has since been extended to the following 4 equity dimensions of expected returns through the insights of MPT:
Let’s look at sustainability scores for each of these dimensions. The Equity Premium (also called Equity Risk Premium = ERP) is measured as the return difference between the equity market and the short-term treasury bills (called T-Bills for short). The Company Size Dimension in the picture above is the return difference between small-capitalization-company equity returns minus large-capitalization-company equity returns. The Relative Price Dimension in the picture above is quite easily determined by the equity return in value companies minus the equity returns in growth companies (where value stocks are those with lower price-to-book ratios compared to growth stocks). Finally, the Profitability Dimension shown above reflects the difference in equity returns of highly profitable companies minus the equity returns of companies with low profitability (where profitability itself is a measure of current profitability based on current information of individual companies’ income statements).
Here is a snapshot of the last 3 Dimensions of expected returns, where we have the longest data since 1928 for the US markets, while we have data since 1970 for developed markets ex US and since 1989 for Emerging Markets:
The quite obvious observation is that all of these 3 dimensions give positive expected returns. Since these individual dimensions are not statistically independent of each other in any investor’s portfolio, it would be a mistake to just add them up to arrive at a sustainability score for the dimensions of expected return. In addition, each of these dimensions of expected returns fluctuates depending on the observation period, with a small subset of periods leading to marginal negative results for an individual dimension observed.
A much better way to judge the sustainability of these dimensions is to look at the rolling 10-year returns for each individual dimension, combined with the Equity Risk Premium discussed above. For the US markets we have data going back to 1937, i.e the 10-year period ending in 1937:
The picture for the same 10-year rolling periods for Developed Markets ex-US – where we have data since 1984 – is depicted below:
Finally, for Emerging Markets, where we have data on these 4 dimensions since 1998, a similar picture emerges:
Looking at the 3 pictures above clarifies the sustainability score for each of these 4 equity dimensions of expected returns. While each individual dimension itself does not guarantee immediate positive 10-year rolling returns (except for the Profitability Dimension which has always been positive across all markets and all rolling 10-year periods), adding up all 4 of the dimensions has consistently rewarded investors with positive returns. The only exception to this observation is the period from 1927 to 1939, i.e three 10-year rolling periods. These 3 individual 10-year rolling periods all had one thing in common: their results suffered from the pandemonium meted out by the Great Depression from 1929 to 1933, where the capitalistic market economy came to a grinding halt with soup kitchens springing up everywhere and unemployment rates of 25% in the US and Western Europe.
In summary, following the academic way of investing laid out in the blog posts series titled “Scientific Investing I –IV” did indeed lead to investor rewards. This is not only substantiated by the diverse 10-year rolling periods depicted above for the four dimensions of expected returns discussed above. More importantly, these insights came from Nobel Laureates and academics who founded Modern Portfolio Theory. Ultimately, an investor should be able to invest and relax in the knowledge that Nobel Laureates are in his retirement boat rowing on the investor’s behalf!
Sources: Data provided by DFA, Vanguard, CompuStat, Bloomberg and CRSP