Academic journal article Academy of Accounting and Financial Studies Journal

What Does Book-to-Market Proxy: Risk or Investor Sentiment?

Academic journal article Academy of Accounting and Financial Studies Journal

What Does Book-to-Market Proxy: Risk or Investor Sentiment?

Article excerpt


Debate persists over the long run impacts of systematic risks versus investor sentiment on asset returns (Griffin & Lemmon, 2002; Vassalou & Xing, 2004). At the center of this controversy is the book-to-market ratio, which our study decomposes into three parts. This approach allows us to better investigate whether book-to-market is a proxy for risk factors or investor sentiment. Time-series regression analysis is applied to ten industries over 1934-2003. In contrast to prior research on investor over-/under-reaction, we find the component of book-to-market correlated with investor sentiment has only marginal explanatory power. The component of book-to-market correlated with systematic risks better explains the time variation of industry portfolio returns.


There is abundant research documenting the robustness of book-to-market values of equity in explaining stock returns. However, while the explanatory power of the book-to-market 'factor' is generally accepted, there is significant debate over whether the factor is a risk proxy or its significance is the product of mispricing. Risk proxy arguments generally focus on the risk of financial distress, which is presumably greater in small firms and firms with market values of equity that are low relative to their book values. Mispricing arguments usually focus on investor irrationality as captured by over-reaction to past information or extrapolation of past trends too far into the future, both of which produce over- or under-valuations that are ultimately corrected as new information is revealed. Resolving this debate is important because if firm size and book-to-market are indeed proxies of systematic risks, rather than due to mispricing, then using the popular Fama and French (1993) model as an asset pricing model is justified.

Because evidence exists to support both explanations, clearly separating the two potential sources of significance is difficult. However, attempts have been made to test which source dominates. Daniel and Titman (1 997) form portfolios based first on book-to-market (B/M) and then on Fama and French's three factors. They find a stronger relation exists between expected return and B/M than between expected return and the three factors. They conclude that B/M has more explanatory power than the factor loadings. Their results support a characteristics-based model, and are consistent with a mispricing story. In contrast to the Daniel and Titman results, Lewellen (1999) shows with a conditional time-series regression analysis that B/M provides no incremental information beyond that which is provided by the Fama-French factors. In other words, Lewellen's approach assumes the Fama-French factors are risk factors and any portion of B/M not correlated to them could be considered something related to investor over-reaction.

Our study builds upon Lewellen's framework by examining the role of investor sentiment vis-à-vis risk factors to explain long run stock returns to industry portfolios. We decompose B/M into three parts - the part correlated with risk factors, the part correlated with investor sentiment, and the part not correlated with either. While Lewellen implicitly assumes that the part of B/M not correlated with risk factors reflects investor sentiment, we investigate whether B/M can be decomposed into three different components as well as the relative importance of each component. We find that B/M' s correlation with risks most effectively explains industry portfolio returns over time.

The next section provides an overview of relevant literature. Section three describes the empirical methodology. Section four presents the results. Section five concludes.


Fama and French (1992) report that both firm size and book-to-market values of equity dominate other variables in explaining the average returns on U.S. stocks. More specifically, leverage and price-to-earnings ratios alone were found to be significant in explaining average monthly returns. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.