Abstract

Investment managers are becoming increasingly concerned by style investing, as demonstrated by recent studies. This article begins by giving an overview of what style investing is and then describes how this type of investing, when massively followed by investment managers, can influence stock returns.

Style investing consists of constructing portfolios by choosing assets according to the style they belong to. The mutual funds industry is now offering funds that follow the different styles. Typically, a strategy based on style creates arbitrage between a main style and a competing style, the latter having opposing characteristics to the main style: for example, growth versus value or small capitalization versus large capitalization. Depending on different time periods, the styles involved in each pair encounter more or less favourable results. Investors seeking to achieve better performance will tend to rebalance their portfolio towards the style that is favoured by the market. These coordinated demands for assets of a same style are likely to have an impact on stock returns. Indeed, if the volume of transactions becomes too high, this may prevent prices from being fully corrected by arbitrageurs.

Previous asset studies have documented the impact of institutional block trades on stock prices. More recently, several studies have also documented the effect of investor sentiment on returns. This article contributes to the subject by considering the special case of style investing and how the practice of style-switching by individual investors influences stock returns. Two recent papers related to the subject of style-based investing can be cited, Teo and Woo (2001) and Barberis, Shleifer, and Wurgler (2001), although their focus is quite distinct. Teo and Woo have studied the profitability of style-level momentum and contrarian trading strategies. Barberis, Shleifer and Wurgler (2001) examine changes in comovement of stocks added to or deleted from the S&P 500 index.

Kumar has conducted a study using monthly data from more than 60,000 individual investors over a period of six years (from 1991 to 1996). The database contains, in particular, all transactions carried out by the investors. According to the author, the sample of investors used for the study gives a good representation of US households. Its study is divided into three parts. Firstly, it tests whether investors formulate their demand shifts at a style level. Then it looks for the main determinants of investors’ style switching behaviour. Finally, it investigates whether style-switching induced demand-shocks influence stock returns.

The author follows the portfolio characteristic-based approach of Grinblatt and Titman (1989) to construct style portfolios and investigates whether individual investors in the sample base their fund re-allocation on style. According to the results, there is strong evidence of style-based investing among individual investors. Considering simple stock characteristics, such as market capitalisation or book-to-market ratio, investors group stocks into styles and reallocate funds at a style level.

Kumar then investigated the main determinants of investors’ style-switching behavior. He found that they base their decisions concerning the allocation of funds to style on the past performance of the different styles, although this may be controversial since past results do not predict future results. Investors also use analysts' recommendations published in newsletters, generating a relatively higher demand for the style favoured during the previous month, but they appear to be only weakly influenced by innovations in macro-economic variables.

Finally, with the help of a multi-factor model, the author highlights the existence of a relation between style returns and investors’ demand shifts. The model uses the three factors of Fama and French (1992, 1993), the momentum factor of Carhart and a fifth factor which is the demand-shift differential variable. These results suggest that investors’ style switching behavior is part of the stock returns generation process. These results are consistent with those of Barberis and Shleifer (2001). They also provide justification, in relation to investors’ behavior, for the choice of size and book-to-market ratio as explanatory factors in asset pricing models. An additional consequence of these observations is that past style returns and demand shifts enable future style shifts to be predicted.

References:

Barberis N., Shleifer A., "Style Investing", Working Paper, University of Chicago, November 2001.

Barberis N., Shleifer A., Wurgler J., "Comovement", Working Paper, Department of Economics, Harvard University, November 2001.

Fama E. F., French K. R., "The cross-section of expected stock returns", Journal of Finance, vol. 47, p. 427-465.

Fama E. F., French K. R., "Common Risk factors in Returns on Stocks and Bonds", Journal of Financial Economics, vol. 33, p. 3-56.

Grinblatt M., Titman S., "Mutual Fund Performance: An Analysis of Quarterly Portfolio Holdings", Journal of Business, vol. 62, p. 393-416.

Teo M., Woo S.-J., "Style Effects", Working Paper, Department of Economics, Harvard University, November 2001.