Contrarian Investment, Extrapolation, and Risk

Contrarian Investment, Extrapolation, and Risk

By: Lakonishok, Shleifer, and Vishny

            This article examines whether value strategies calling for investors to buy stocks that have low prices relative to earnings, dividends, book assets, and other measures of fundamental value have produced higher average returns than those using the strategy of choosing the large companies that are on the other end of this scale.  Lakonishok, Shleifer, and Vishny (L S and V) find that these strategies do yield higher average returns during their sample of companies in the NYSE and AMEX between 1968 and 1990.  L S and V use four main measures for choosing where companies fit on the “glamour” to “value” scale.  They use book value of equity to market (B/M), cash flow to market value of equity (C/P), and earnings to market value of equity (E/P), all of which are higher or value stocks and lower for glamour stocks.  Their last measure is preformation of portfolio 5-year growth rate of sales (GS), which is lower for value stocks and higher for glamour stocks.  Using these measures, they separate stocks into portfolios for each measure and compare average returns of each portfolio.  These results are shown in table 1 (where they use one dimensional classifications—ex. Only C/P), table 2 (where they use two dimensional classifications—ex. Both C/P and GS, and table 3 (same as table 2 but only includes the largest 50 percent of stocks).  These tables are shown below.

            Using CR₅ (compounded 5-year return assuming annual rebalancing of portfolios) in these three tables, it is easy to notice that value stocks have a much higher average return than glamour stocks using each measure of value.  This fact is further discussed in the article.  L S and V compare the strategies of naïve investors and contrarian investors.  Naïve investors will choose to invest in stocks that are doing well because they expect stocks that are doing well to continue doing well and avoid stocks who are doing poorly.  This will lead them to invest in these glamour stocks (Google, Apple), which are overpriced and will not produce a high return.  The contrarian strategy is the opposite.  Investors using this strategy will avoid these large glamour stocks that they view as overpriced based on these value measures listed above and focus on the stocks they see as undervalued based on those same measures.  The stocks they focus on are considered the value stocks in the tables above.

            The fact that value stocks are profitable compared to glamour stocks raises the question of why they are so profitable.  L S and V offer two popular explanations.  One, these strategies are fundamentally riskier, leading to a risk premium.  Two, these strategies exploit the suboptimal behavior of the typical investor.  This “suboptimal behavior” is exhibited in the actions of the naïve investor discussed earlier.  L S and V work to prove the abnormal returns are due to the behavior of investors and not simply the risk of these stocks.  To show this, they use table 6, which subtracts the return of glamour stocks from return of value stocks for three different models.  When looking at the year-by-year differences and average, it suggests the value stocks are not systematically riskier than the glamour stocks.  They also use table 7 to compare the performance of glamour and value stocks during good and bad times.  Looking at when stocks are Worst (W₂₅) and other Negative months (N₈₈), value stocks actually perform better in bad times than glamour stocks.

            To explain the behavior of naïve investors preference for glamour stocks despite their lower average return, L S and V say they could be making judgment errors and extrapolate past growth rates of glamour stocks, or just equate well-run firms with good investments, regardless of price.  These are not good reasons for investing in these glamour stocks, but L S and V may have better reasons for investing in these stocks.  They say it is easy for an institutional investor to rationalize the investing in glamour stocks because they are easy to justify to sponsors.  If they were to invest in value stocks that don’t pay off, they could be fired for investing in companies that are perceived as “prudent” investments by sponsors compared the glamour companies who have been doing well recently.  However, investing in these glamour stocks isn’t prudent at all since they have a lower expected return and are not fundamentally less risky.  They conclude by noting the implication that this reasoning could explain the inferior performance of pension fund money managers compared to the market index found in an earlier paper.