A helpful summary on value strategies from Tweedy, Browne

9 Nov 2014 by Jim Fickett.

Tweedy, Browne Company, LLC is a small firm with a long history in value investing:

The Firm’s 94-year history is grounded in undervalued securities, first as a market maker, then as an investor and investment advisor. The Firm’s investment approach derives from the work of the late Benjamin Graham, co-author of the first textbook on investment research, Security Analysis (1934) and author of The Intelligent Investor (1949). Graham, through his investment firm Graham-Newman Corp., was one of the Firm’s primary brokerage clients in the 1930s, 1940s, and 1950s. It was through Graham that the original partners of the Firm developed brokerage relationships with investment legends Walter Schloss and Warren Buffett …

In 1959, the partners of then Tweedy, Browne & Knapp pooled their capital in a partnership investment vehicle. In 1968, the firm accepted its first outside money management clients as limited partners of this vehicle. In 1975, Tweedy, Browne registered as an investment advisor and began managing separate accounts for individuals and institutions. As of September 30, 2014, the firm managed approximately $21.4 billion for individuals, institutions, partnerships, off-shore funds and four mutual funds

Tweedy, Browne have made available a number of research reports, one of which is entitled “What Has Worked In Investing, Studies of Investment Approaches and Characteristics Associated With Exceptional Returns”, or “What has worked” for short.

What Tweedy, Browne have done is read about 50 published reports, each of which evaluates the results of following a particular value-oriented strategy in a particular market over a particular period. They have grouped these reports by strategy and market, and summarized the results. Although the general idea of “buy low, sell high” is obvious enough, there are many ways to implement it. The studies covered in this review evaluate cheapness relative to current assets, to book value, to profits, to cash flow, and to dividends. They also look at other indicators sometimes related to cheapness, such as insider buying and market capitalization.

The three most important conclusions are that (1) almost any form of “buy cheap” does give better returns than investing in an index covering the whole market, (2) this better performance does NOT come at the price of higher risk – value stocks decline less than others in a bear market, and (3) the definition of “cheap” giving the best results overall is “the stock price is cheap compared to book value”. The latter conclusion comes from a study by Fama and French:

the authors, through a regression analysis, examined the power of the following characteristics to predict future investment returns: market beta, market capitalization, price/earnings ratio, leverage and price-to-book value percentage. Their conclusion: price-to-book value “is consistently the most powerful for explaining the cross-section of average stock returns.”

The fact that “buy cheap” works is not a surprise, though it is useful to see how many studies confirm that it works in various forms. There are some surprises in the report. For example, one study concludes that among undervalued stocks, the ones that perform best are those for which the company is losing money. I would have expected the opposite – that positive income would be one more indication that the market was being irrational, while a company making a loss would be, in fact, expected to be a so-called “value trap”.

After reading “What has worked”, one realizes that there are some gaps in the current literature on value investing. Although it does seem intuitively that the basic strategy of “buy cheap” should work universally, formal studies cover only a very narrow scope. There are exceptions, but most of the studies cover American stocks, and most cover time periods in the last three decades of the twentieth century. This is, of course, a result of data that are more easily accessible. But it would be nice to see more studies extending to other domains.

The studies also leave open the value of active management. It is interesting that most of the studies show that some mechanical value strategy produces excess returns (over the whole market) of several percent annually. Tweedy, Browne's excess returns for the Global Value Fund, since inception in 1993, are about 4%. The possibility that a rigid, mechanical strategy might do just as well as an active manager is probably irrelevant for the small investor, because investing in a portfolio of a large number of stocks, with frequent turnover, would incur excessive transaction costs. But it does raise the question of whether it would be valuable for one of the big fund managers to introduce a low-cost fund that just invests mechanically in those stocks with the lowest price to book ratio.