Tracking Style and Style Drift
In the previous post, we highlighted the many challenges that today’s active managers face. Chief among those challenges is standing out among the crowd. Today, passive funds are both plentiful and inexpensive, so if an active manager is going to attract capital they need to be offering something that is not easily available elsewhere. It is hard to quantify the insight and judgement of a professional investor, but with improving quantitative methods it is easier to do this now than ever before. One of the ways that this can be done is by applying factor analysis to investment portfolios. Factor analysis can provide insight into an investor’s general style, communicate that style to their investor and show both the manager and their investors if there is any change in that style.A factor is a characteristic that relates to a group of securities which helps explain their risk and return. The Fama - French three factor model popularized this approach, when they used the explanatory power of value and size factors to help explain why some stock returns were better than others. Academics and investors have been finding more factors ever since to better explain and predict stock returns.
There is a negative stereotype that factor analysis is just for funds that apply too much leverage to eke out uncorrelated returns for their investors until they blow up in a flash crash. But it can actually be a very useful tool. In order to understand how factor analysis helps with both the communication of the style and to make sure that style drift is not occurring we will apply it to the portfolio of the world’s most famous value investor, Warren Buffett. His public market long portfolio is available for everyone to peruse via 13F filings and we can analyze his factors to get a sense of Berkshire Hathaway’s investment style*.
Buffett is known for buying more stable and seemingly boring long term stocks. One way to approximate this stability is to look at the volatility factor, and as we expect Berkshire Hathaway has been buying stocks with volatility in the lower deciles than the broad market, which would have a score of 5.5. (5.5 represents the 50th percentile stock, as 5.5 is exactly in the middle of the lowest decile, the 1st, and the highest decile, the 10th.)
Berkshire has also been holding stocks that have had better dividend growth than their peers over the past few years.
Berkshire stocks also have a higher return on equity.
The interesting thing that all of these charts have in common is that they all show the stocks exposed to factors that might be used to approximate classic value styles of investing, and all of the factors have been moving away from the value style towards a more moderate style. If Berkshire Hathaway were a hedge fund that only traded public stocks, its investors should be questioning what is changing at the company. And in fact a related question was the first question asked at the 2016 annual meeting. And when we look at the deciles of more classic value factors such as price to book or PE ratios, Berkshire’s public companies do not look significantly cheaper than the broader US market represented here as the “Short” category.
Not all style shifts are bad or unwanted. Sometimes market environments change and portfolio managers are making a purposeful decision to react. That is probably the case with Berkshire’s view towards public companies owning debt. Warren Buffett has historically been against public market companies owning too much debt. But over the past few years their portfolio companies have been taking on more debt relative to their public market comparisons.
This is likely due to Berkshire’s switch towards buying more capital intensive businesses. Charlie Munger and Warren Buffett addressed the falling return on equity pretty directly by acknowledging that they were not seeing high ROE types of business opportunities in this market. Charlie Munger called these opportunities Plan B. This change is related to the fact that they just have too much capital to allocate compared to when they were able to significantly outperform the S&P 500 in the past.
Active managers can benefit from using factor analysis to understand their own style, exposures to core and unintended risk and communicate style and risk exposure to their investors. The investors putting money with active managers also have reason to apply factor analysis to active managers. First, they make sure that the story they are being pitched reflects the portfolio the fund actually holds. After their money is invested, tracking factor exposure is also tracking style drift and may give the investor the opportunity to question managers about what is changing at the fund before they see unexpected changes in the underlying returns.
There are other advantages to tracking factor exposures, which we will cover in upcoming posts.
Regards,
Omer
