Here's an intriguing question for value investors: If you had to choose between the two, which stock would you pick – Fairfax Financial Holdings Ltd. or Berkshire Hathaway Inc.?
Both companies are in the insurance industry, led by legendary value investors Prem Watsa and Warren Buffett, respectively. They are both capable insurance underwriters who have wisely invested their float (that is, the difference between premiums collected and claims paid) over the years of operation, thus profiting from both sides of their balance sheet and earning unparalleled returns for their shareholders. But this is the past; the question is, going forward, which one makes a better long-term investment?
Both companies subscribe to the value-investing philosophy of making investment decisions following a three-step process. First, they search for stocks with desirable characteristics; second, they value such stocks to determine their intrinsic value; and third, they make an investment decision to buy only if a stock meets the desired margin-of-safety requirement. However, while both companies follow similar steps in stock selection, they differ in what they view as desirable stocks – the first step of the process – deserving to be considered for the second step.
Fairfax follows a Ben Graham approach, which involves looking for obscure and unattractive stocks (that is, companies in financial distress, with a low price-to-earnings ratio and low expected growth). Berkshire Hathaway, on the other hand, looks for stocks that have barriers to entry and a sustainable competitive advantage. The former puts less emphasis on quality investing, while the latter makes quality investing a key ingredient of their investment process. It all then boils down to which approach can produce better long-term investing results. Put simply, do lower-quality value stocks produce better long-term performance than higher-quality value stocks?
I looked at this question both from a qualitative and quantitative point of view. From a qualitative point of view, we can make the following observations with regards to the investment-performance potential of Fairfax versus Berkshire. First of all, we know from Berkshire's performance over the years that its highest returns were earned when Mr. Buffett was following the Ben Graham approach, and that its more recent performance, after Mr. Buffett changed to a quality investor, while satisfactory, lags Berkshire's earlier performance. If we split the 1965 to 2016 period of Berkshire's existence into two equal sub-periods, we see that in the first period it had an average rate of return of 24 per cent, whereas in the second 25-year period it averaged just less than 16 per cent. This confirms the slowing down of Berkshire's returns since Mr. Buffett started to emphasize high-quality stocks.
Second, Berkshire is now the sixth-largest company in the U.S. by market cap. It is hard to compound returns at the same rate as it did in the past, when it was smaller. Fairfax is a much smaller company. Third, Berkshire is focused on North America, which, as a region, is bound to have fewer growth opportunities than markets in other parts of the world, particularly emerging economies where Fairfax has been heavily investing in recent years. Fourth, Mr. Buffett is at an advanced age and, in my opinion, is no longer making Berkshire's investment decisions; I have no doubt that Berkshire's recent investments in airlines and Apple were not made by Mr. Buffett. Mr. Watsa is much younger and still heavily involved in investment decisions. Finally, Fairfax invests in smaller, lesser-known companies than Berkshire, which, by definition, have better opportunities to compound returns than the larger companies in which Berkshire invests.
How about examining the potential future performance of Fairfax and Berkshire by looking at the performance of different quality investing styles in a more quantitative, generic and objective fashion? My co-author Vasiliki Athanasakou and I tried to answer this question in a recent study, as yet unpublished.
The study, which uses all AMEX-, NYSE- and Nasdaq-listed U.S. companies, covers the period from 1982 to 2015. We arrive at a measure of earnings quality as follows: We first divide earnings before extraordinary items by total assets and then take the standard deviation of this ratio over a five-year period. The resulting standard deviation is our proxy for earnings quality. For our purposes, the higher the standard deviation – a measure of historical volatility – the lower the earnings quality.
We find that while value beats growth (that is, the value premium is confirmed) in the total sample (on average, by 430 basis points annually – a basis point is 1/100th of a percentage point), the value premium is actually driven by the firms with the poorest earnings quality. The average value premium for the best earnings quality firms (the top quartile) is 60 basis points, whereas the corresponding value premium for the poorest earnings quality firms (the bottom quartile) is 960 basis points. That is, an investment strategy that emphasizes lower-quality value stocks will improve the long-term performance of a value portfolio. True, such approach has higher short-term volatility, but as value investors, we do not consider volatility as a measure of risk.
So, Fairfax or Berkshire Hathaway? Based on the above qualitative and quantitative evidence, I dare to predict that, going forward, Fairfax's stock may have a better long-term performance than Berkshire Hathaway's, given Fairfax's orientation toward the Ben Graham (lower-quality) type of value stocks.
George Athanassakos is a professor of finance and holds the Ben Graham Chair in Value Investing at the Richard Ivey School of Business, University of Western Ontario.