It’s no secret that a key aspect of Loring Ward’s portfolio management effort is capturing value, small cap, and profitability factor premiums for our clients. It’s also no secret that some of these premiums haven’t offered much to capture in recent years. In fact, the value factor premium in the U.S. has been negative for more than a decade.1

The prolonged absence of some premiums has caused some advisors to question whether these factor premiums still exist. There are lots of possible theories as to why a specific factor premium may have permanently disappeared. However plausible these theories may sound, historical data tells us factors have been positive over the long run, despite having disappeared for seasons in the past.

When we look at the historical returns of factor premiums, we see positive averages, although ones that are highly volatile (see Table 1 below).

As we can see, while the long-term average return for each of the factor premiums is positive, the monthly standard deviation (i.e., the risk of receiving the factor premium) for each is very high relative to the monthly average return. This high standard deviation means that it is very possible investors can see multi-year periods of negative factor premiums.

But how long do we have to live with negative factor premiums before we lose faith in them? To help answer this question we looked at historical data and used Monte Carlo simulation techniques to see what the probability of a multi-year disappearance of any one of these factor premiums would be. What we found was that prolonged factor premium disappearances are normal and should be expected.

We ran an extensive series of Monte Carlo simulations on the three factor premiums we target to test how often we would expect to see extended periods of negative premiums in each of those factors, given what we’ve seen from each of them historically. We simulated multi-year time periods ranging from 5 years to 30 years and performed 10,000 simulations for each period on each factor. We then calculated the percentage of time each factor return was cumulatively negative across the tested period. Table 2 below shows our results.

When we look at these probabilities, we see a number of important insights. Take, for example, the value factor over a 10-year period. The probability of observing a negative value factor premium over a 10-year period is roughly 25%, based on our simulation results. We could also state this as the probability of observing a positive value factor premium over a 10-year period is roughly 75%. The probability of earning the value premium is fairly high; this is why we tilt our portfolios toward a value factor exposure. But as investment professionals we must also recognize and respect risk, and the risk here is a 25% chance that the value factor could be cumulatively negative for a 10-year period. The same logic applies to the other factors we simulated across all other time periods.

Gene Fama and Ken French came to a similar conclusion in their recently published paper, “Volatility Lessons.” They used a bootstrapping method based on realized historical data and historical data with some added variance around the average return. Their conclusions were similar to ours, and they summed up their findings as follows:

The high volatility of monthly stocks returns and premiums means that for the three- and five-year periods used by many professional investors to evaluate asset allocations, the probabilities that premiums are negative on a purely chance basis are substantial, and they are nontrivial even for 10- and 20-year periods.”

Based on our work and the independent work of Fama and French cited above, we believe that the occurrence of factor premiums over long horizons is probable, but the high volatility of these factor premiums means that they are not guaranteed, and it may take a number of years to receive the premiums. In fact, the current period of factor underperformance is well within the statistical probabilities and is to be expected.

We also believe that the occurrence of a long-horizon negative premium is not justification in and of itself to abandon our strategy of tilting towards these factor premiums. It just indicates that factor tilts — investment choices we believe are worth making — face risks of underperformance (as do all investment strategies). It just means we must stick with the strategy for an extended investment horizon to increase the probability of capturing positive factor returns. And that really shouldn’t be a problem for long-term investors.

Special thanks to Bob Bannon for his help in running factor premium simulations

1 Measured as the Russell 1000 Growth Index minus the Russell 1000 Value Index.

Past performance is no guarantee of future results. Information is not indicative of any Loring Ward product or strategy.