While valuation has been considered as important to factor investing, the right way to do it isn't so clear
While adoption of factor strategies is still in its early stages, they are becoming increasingly popular among institutional and wholesale investors alike. With that rising popularity, some academics and practitioners are beginning to ask whether factor premiums have been arbitraged away, and are turning to relative valuation measures of factors for answers.
Unfortunately, determining relative valuations for investing factors can be fraught with complication. “Despite the importance attached to valuation in general and its specific importance to factor investing, there is no consensus on the appropriate valuation metric or approach to follow,” FTSE Russel said in a new report.
To illustrate, it examined the valuation of a set of single-factor US indexes (Size, Value, Low volatility, Quality, and Momentum) over the period from September 2000 to January 2019. Aside from active Value factor exposure, the analysis looked at two relative valuation measures — sales-to-price and book-to-price ratios relative to the underlying US index — for each single-factor US index.
“Relative valuation is calculated as the ratio of the single factor index and the US market capitalization weighted index valuation metrics multiplied by 100,” the report explained. Readings above 100 indicate that the factor index is cheaper than the underlying index; conversely, readings below 100 signal that the factor index is more expensive.
Focusing on the Quality US index, the report showed that its relative sales-to-price valuation was range-bound, fluctuating between 88 and 102. But looking at the relative book-to-price ratio for the same factor index, it found the valuation drifting steadily downward from 89 to 76, suggesting that the Quality index has become more expensive over time from a book-to-price ratio perspective.
The analysis also found differences between sales-to-price and book-to-price ratios relative to recent history. As a particularly stark example, it noted that the relative book-to-price ratio in April 2018 was 1.2 standard deviations above the rolling mean, indicating relative cheapness; for the same period, the sales-to-price ratio was close to 0.8 standard deviations below the rolling mean, suggesting that Quality was relatively expensive.
“Valuation differences are more striking when alternative index construction approaches to the creation of single factor indexes are used,” the paper continued. In Selection and Weighting (S&W), one common alternative approach, a proportion of some initial universe of constituents is selected after being sorted based on the desired factor characteristic. The index is then formed by a simple diversified weighting scheme.
FTSE Russell analysts created an S&W index by selecting the top 50% of stocks ranked by the Quality score and then equally weighting the constituents. After determining relative valuations as of January 18, 2019, they found that the S&W Quality Index appeared notably cheaper than the Standard Quality Index based on both book-to-price (83.9 vs. 75.5) and sales-to-price (121.3 vs. 94.1) measures.