We discuss the Price to Book values of various stocks quite often on this site, but how useful a metric is it? From a logical standpoint, as a purchaser of a business (which is how we view all our stock purchases), a prudent buyer ensures - barring certain exceptional circumstances - that he does not pay too much more for a company than the value of its assets. In this way, he receives downside protection to a certain extent. Though book value is not a perfect measure of the value of a company's net assets, it does provide at least some level of a proxy for it.
That's all well and good in theory, but does this practice actually hold up against empirical data? Do stocks with low P/B values indeed outperform the market? There have been several studies that suggest that historically, stocks with lower P/B values have in fact outperformed, however, there are certain caveats to keep in mind.
One study that has gained industry credence was carried out by Bauman, Conover, and Miller. The authors used an international sample of stocks and divided them into quartiles based on P/B. They observed over the ten-year period of their study that the quartile of the lowest P/B stocks had mean returns of 18.1%, while those of the highest P/B had mean returns of 12.4%, representing an annual spread of 5.7%. This is consistent with what other studies have found, including those conducted by David Dreman, discussed here.
Based on this data, buying a basket of low P/B stocks may get you outstanding returns, but you may do even better if you can determine which of the low P/B stocks are worth purchasing and which are about to go bankrupt: the standard deviation of the returns for the lowest P/B stocks was 70 as compared to 57 for the highest P/B quartile, suggesting the low P/B space contained some big winners along with some big losers.
This is why looking for companies with low debt and good liquidity among issues trading at discounts to their book values can present great investment opportunities, some of which we discuss here.