Value investing: the book-to-market effect, accounting information, and stock returns

Abstract

Although the book-to-market (B/M) effect is vastly studied, the majority of the conclusions in prior analysis is only applicable to U.S. firms. In this work, we evaluate the performance of portfolios selected using three modified versions of B/M strategy applied to stocks listed in Euronext markets (Paris, Amsterdam, Brussels, and Lisbon) between 1993 and 2003. From the analysis of 4,715 firms across 11 years, 943 firms were elected as reference for portfolio formation. The modified B/M strategies use accounting information to segregate good from troubled firms. The first strategy follows Piotroski's (2000) nine signals to measure three areas of the companies' financial situation and enabling to select firms from the high B/M quintile. The second strategy creates a portfolio from the intersection of high B/M portfolio with low accruals portfolios, following Bartov and Kim (2004) research design. The last strategy combines high B/M and low probability of bankruptcy, using the methodology described in Airman (1968) and Hillegeist et al. (2004). This study shows that the average annual return observed by the high B/M portfolio is increased by 9.2% using the strategy developed by Piotroski (2000). Furthermore, there is clear evidence that the entire high B/M firms return distribution is shifted to the right when the score screen is applied. By opposition, other suggested alternative techniques pointed out in the literature using similar accounting and market data failed to prove as being a more efficient investment strategy