The underperformance of value: a tale of unintended consequences

Many value investors have struggled over the decade since the financial crisis, and their underperformance has in many cases been even more apparent over recent months.

Taking the performance of the MSCI All Country World Index (ACWI) Value as representative, we apply our proprietary factor attribution methodology to decompose the active returns of the index into underlying factor and signal drivers.

Two key features have driven the underperformance of MSCI ACWI Value, and by extension other value strategies. 

Bigger negative impact from momentum and quality

First, as with most single factor approaches, actual performance is not only due to positive exposure to the desired factor, but also by negative and generally uncontrolled exposures to other factors. 

In this case, ACWI Value is not just a value strategy; it is also an anti-momentum and anti-quality strategy.

When we broke down the index into five major style factors we found that the value factor accounted for only 0.2% points of annualised underperformance between January 2010 and August 2018. The bigger impact came from momentum and particularly quality. Together, these accounted for around 85% of underperformance over the period.

Greater exposure to underperforming measures of value

Second, as with most factors, there are large performance differences between different measures of “value”.  ACWI Value embodies the most widely used measures of value (book-to-price and dividend yield). Unfortunately, and perhaps not coincidentally, these have performed the worst of any of our value measures over the past decade.

The family of signals focused on cashflow measures (operating cashflow yield, free cashflow yield, earnings yield, and buyback yield) have all performed better than the traditional measures and are the measures to which the index is least exposed.

There are at least two possible forces that may have driven the relative performance of these metrics. First, returns to the widely-used value signals may have been arbitraged away over time. Second, in a world in which manufacturing companies are less dominant than in the past, it is possible that a measure such as book-to-price fails to convey meaningful information about expected future cashflows.

We conclude that investors would do well to choose their value factor carefully and to control their exposure to extraneous factors. Investors are likely to make the most of factor investing by forming integrated portfolios of non-commoditised factors.

We explore this subject in greater detail in our full research paper below.