Em busca de um índice alternativo à relação Book to Market para a construção de carteiras mais rentáveis

Abstract

Many studies have tried to predict the potential return on stock portfolios, aiming to get better return on invested capital. Several modeling have been used, and the more popular are those that relate the risk with the return. In this area, stand out the Portfolio Theory proposed by Markowitz and the CAPM proposed by Sharpe. Through these theories can be understood the influence of the covariance of returns, and for a best performance of a portfolio, is not enough to assess each individual asset. On the other hand, many criticism of the CAPM, have generating additional studies in search of other variables to improve the methods of selection of assets. Fama and French (1993) conducted a study with additional variables in relation to the CAPM beta, using the size and the relationship Book to Market, achieving better results than the traditional CAPM. This study considers the issue of the generated profit reinvestment, and using the model of Gordon proposes a classification variable for growth companies and value companies, which are concepts already used in finance literature. Based on this variable are set up stock portfolios between the years 2005 and 2012 and it is observed that it is possible to get earnings with the logic proposed. Over the period could be obtained with the selected portfolios up to 107.85% gains against the returns of 55.58% of the complete portfolio with all assets. We organize the same stocks from the perspective of the relationship Book to Market which had a total return of 90.42% on the whole period. Although observed a clear change of behavior, where only the first four years of the study portfolios with value companies are superior and the last four years portfolios of growth companies are the best. These results are consistent with the results of Braga and Leal (2000), and Mescolin Martinelli Braga and Costa Jr. (1997), watching a better performance for value companies.