- Courses and Faculty
- 5x5x5 Student Portfolio
- Get Involved
The Basics and Two Classic Papers
The value premium is the empiral observation that stocks with high book-to-market have on average high average returns whereas stocks with low book-to-market have on average low returns. A stock has a high book-to-market whenever the accounting value of the equity is much larger than the corresponding market value.
There are two classic references in the field: One is a remarkable piece by Barr Rosenberg, Kenneth Reid and Ronald Lanstein entitled "Persuasive Evidence of Market Inefficiency" in the Journal of Portfolio Management (1985: 11, 9-17); the other is a well known 1992 classic paper by Gene Fama and Ken French ("The Cross Section of Expected Stock Returns," Journal of Finance, 1992: 47, 427-465). There is an earlier reference which is a difficult to find paper by Dennis Stattman ("Book Values and Stock Returns," The Chicago MBA: A Journal of Selected Papers, 1980:4, 25-45.) The piece by Fama and French is one of the most heavily quoted papers in academic finance and for good reason: It is a remarkable piece not only of clarification of the empirical regularities in the cross section of stock returns but also a methodological contribution that has had a lasting effect on the subsequent literature.
It is perhaps worth it to review briefly this methodology, which is not that dissimilar to what many casual investors do as well. Indeed many investors form "lists" of attractive investments according to their favorite metric. Fama and French do something similar but with the entire universe of publicly traded stocks: They form portfolios of stocks according to book-to-market. What does this mean? They first construct a list of the universe of stocks starting with the ones with the lowest book-to-market. Then they bundle these stocks into ten portfolios (it's more like 12, but that need not concern us here.) The first portfolio then is the one with the stocks with the lowest book-to-market and the tenth of the stocks with the highest book-to-market. Academics refer to the first portfolio as the extreme growth portfolio and the last one as the extreme value portfolio. They do this at the end of June of every year using accounting information as of December of the previous year. This may seem strange at first but it is simply a conservative way of making sure that indeed the information is available to the investor when forming these portfolios, that is, that the strategy is implementable. The portfolios can be either equal or value weighted. We carry these portfolios for a year, calculating the monthly returns for an entire year. June of the next year we liquidate the portfolios, resort the stocks and start all over again. The result is a time series of monthly returns for these ten portfolios where the first portfolio always contains the stocks that had the lowest book-to-market when the portfolio was formed in June of a particular year and portfolio ten has the stocks with the highest book-to-market at that time. There are many technical details in the formation of these portfolios but the essence of the procedure is as simple as this.
What are the results? The table (below) shows the value and equal weighted monthly average returns (in percentages) as well as the average book-to-market (BE/ME, for book equity divided by market equity) for ten portfolios of stocks sorted according to book-to-market. The Extreme Growth portfolio represents the portfolio of stocks with the lowest book-to-market, D2 represents the portfolio of stocks with the next lowest book-to-market and so until the Extreme Value portfolio which represents the portfolio of stocks with the highest book-to-market. Indeed the Extreme growth portfolio has a value weighted average (annual) book-to-market of .25 versus 4.88 of the Extreme Portfolio.
The value weighted returns go from a low of .86% for the Extreme Growth portfolio to the 1.36% of the Extreme Value portfolio, that is, a monthly spread of 50 basic points. The spread is an impressive 1.22% per month when we consider equal weighted returns. What is the source of the difference between equal and value weighted returns? Obviously that the value premium is a small-to-medium firm effect, a message that is important to keep in mind when thinking about this the most salient fact about the cross section of stock returns.
|Value||Extreme Growth||D2||D3||D4||D5||D6||D7||D8||D9||Extreme Value|
|Value Weighted Average BE/ME||0.25||0.44||0.59||0.73||0.88||1.06||1.28||1.60||2.21||4.88|