Factor Valuation and Passive Investing

During the 3rd quarter, Morningstar reported that over $95 billion flowed out of active U.S. Equity strategies and into passive and semi-passive investment portfolios. Net funds have flowed out of active and into passive vehicles in 68 of the past 69 months, and as the chart below depicts, since 2008, over $3 trillion has flowed out of actively managed strategies and into passive vehicles.

In August of this year, the percentage of U.S. Equity funds managed passively exceeded 50% for the first time.

Source: Morningstar.

While the full implications of this surge of assets into mostly automated and price-insensitive investment vehicles cannot now be fully foreseen, we can observe some coincident structural developments that are creating strains and imbalances across the investment landscape.

Most passive vehicles are managed using a capitalization-weighted construction process. This practice has the advantage of being mechanically simple and mostly self-rebalancing, but on the downside, cap-weighted portfolios are not as diversified as their names often imply. For example, currently only 63 companies make up 50% of the weight of the Russell 1000 Index. The largest 25 stocks, which represent only 2.5% of the count of companies, comprise slightly over one-third of weight of the entire benchmark. For every dollar that flows into either index mutual funds or ETFs based on this index, 33 cents are invested in these largest 25 stocks.

Besides overstating the amount of diversification in these funds, purchases and sales are transacted with no regard to price or valuation. We believe one of the consequences of billions of dollars of inflows is that the largest stocks in U.S. markets have performed at a level that has far exceeded historical norms, and as a result, have become exceedingly expensive.

We measure the sector-neutralized returns for many security characteristics and risk factors. Below is a chart of the annual returns of a theoretical portfolio that invests long in the 10% of the largest stocks in the Russell 1000 Index, on a sector-neutralized basis, and then invests an equal dollar amount short in the 10% of the smallest stocks in the same index. Transaction costs and fees have been ignored in this analysis.

Source: S&P Global, 361 Capital Research.

Since 2000, the smallest decile of stocks in the Russell 1000 Index has outperformed the largest decile by about 3.1% per year. This finding is consistent with long-term research suggesting that smaller stocks should provide a return premium relative to larger stocks due to their higher volatility, less diverse operations, and higher stock-specific risk. However, since 2014, the largest decile of the Russell 1000 has outperformed the smallest decile by approximately 400 basis points per year. 2016 was the only year of the past six where smaller stocks outperformed their larger contemporaries.

This large stock premium phenomenon is not consistent with long-term risk and return expectations and suggests that a ‘free lunch’ has been available by purchasing large mega-cap stocks (higher returns with lower risks). This flocking by many investors into passive and index-related strategies is not without consequence, however.

Valuation distortions are normally the result of abnormal asset flows. Over the past quarters and years, we have begun to observe potential valuation factor bubbles that would be consistent with immense asset flows into passive vehicles noted above.

For example, just like we can measure the valuation characteristics of individual stocks and broad market indexes, we can measure the relative valuation of risk factors and portfolio characteristics. The chart below measures the difference between the median price-to-sales ratio for the largest quintile of Russell 1000 stocks and the smallest quintile of Russell 1000 stocks (again measured on a sector-neutralized basis).

Source: S&P Global, 361 Capital Research.

Prior to 2014, the spread between price-to-sales ratios for the largest and smallest companies averaged +0.7x and the relatively stable spread only widened beyond 1.0x in 12 of the 145 months between 2000 and 2014 (8% of months). Since 2014, the average spread has been 1.2x and the value recorded has been above 1.0x in 54 of 69 months (78% of months). As of the end of the 3rd quarter, the spread in median price-to-sales ratios has risen to 2.1x or more than 180% greater than the pre-2014 average. The same relationship is generally seen if we use price-to-cash flow or price-to-forward earnings estimates.

Clearly, something significant has shifted in the valuation of capitalization. Large companies have performed exceedingly well, and their valuation has reached historic and extreme levels. We believe this previously unseen level of valuation has been driven by the hundreds of billions of investor dollars that have flowed into relatively unmanaged and price-insensitive passive vehicles. Ironically, this flood of dollars into passive vehicles improves the performance of index-related portfolios, creates stronger performance headwinds for the remaining active managers, and in-turn, draws more investors into passive investing.

At some point, the valuation imbalances being created by millions of investors blindly herding into unmanaged index funds will be corrected by market forces. We only hope that the old adage proclaiming that the market can remain irrational longer than (active) investors can remain solvent does not apply in this case.

Read more in our latest blog, Don’t Diversify This Decade: The Best Advice You Could’ve Received at the End of 2009 >