Portal de Periódicos da CAPES

The ‘Dogs of the Dow’ strategy revisited: Finnish evidence

2011; Taylor & Francis; Volume: 17; Issue: 5-6 Linguagem: Inglês

10.1080/1351847x.2010.544951

1466-4364

Eemeli Rinne, Sami Vähämaa,

Auditing, Earnings Management, Governance

Abstract This paper re-examines the performance of the 'Dogs of the Dow' (DoD) investment strategy in a different market setting and over a different time period. In particular, we use Finnish data over the period 1988–2008 to examine whether the DoD strategy can be successfully replicated in different types of markets and in different market conditions. Our empirical findings suggest that the DoD investment strategy is profitable in the Finnish stock market. The DoD strategy outperforms the market index with an average (median) annual abnormal return of 4.5% (7.5%). The outperformance of the DoD strategy appears particularly pronounced in stock market downturns. Furthermore, our results indicate that the DoD strategy outperforms the market index even after most risk adjustments and thereby suggest that the outperformance of the strategy is not merely a compensation for higher risk. Nevertheless, we also document that the superior returns of the DoD strategy may be largely attributed to the winner–loser effect. Keywords: Dogs of the Dow strategyinvestment strategyportfolio management Acknowledgements We wish to thank two anonymous referees for valuable comments and suggestions. Part of this paper was written while Vähämaa was visiting the Kiel Institute for the World Economy, and he would like to thank the Institute, Harmen Lehment, and Leonardo Morales-Arias for their kind hospitality during the visit. Notes Nevertheless, it should be noted that a considerable academic literature exists on the association between dividend yields and stock returns. In general, empirical evidence suggests that dividend yields and changes in dividends may contain useful information for investors (see, for example, Fama and French 1988 Fama, E. F. and French, K. 1988. Dividend yields and expected stock returns. Journal of Financial Economics, 22(1): 3–26. [Crossref], [Web of Science ®] , [Google Scholar]; Martikainen, Rothovius, and Yli-Olli 1993 Martikainen, T., Rothovius, T. and Yli-Olli, P. 1993. On the individual and incremental information content of accrual earnings, cash flows and cash dividends in the Finnish stock market. European Journal of Operational Research, 68(3): 318–33. [Crossref], [Web of Science ®] , [Google Scholar]; Grant 1995 Grant, J. 1995. A yield effect in common stock returns. Journal of Portfolio Management, 21(2): 35–40. [Crossref], [Web of Science ®] , [Google Scholar]). At least two distinct features of the Finnish stock market should be noted in this context. First, due to the very small number of publicly traded firms in Finland, the Finnish DoD strategy involves investing in about 10% of the listed firms. Therefore, the DoD strategy may not necessarily be as effective in the Finnish stock market as in markets with a much larger number of listed firms. Second, in contrast to the USA, dividends are treated more favorably than capital gains in Finland. Since a large proportion of the total return of the DoD strategy comes in the form of dividends, tax considerations may make the DoD strategy particularly attractive in the Finnish stock market. Martikainen (2000) Martikainen, T. 2000. "Security market anomalies in Finland". In Security market imperfections in world wide equity markets, Edited by: Keim, D. and Ziemba, W. 390–415. Cambridge, UK: Cambridge University Press. [Google Scholar] provides a detailed description of the main features of the Finnish stock market and reviews the empirical evidence on stock market anomalies in Finland. A single company, Nokia Oyj, accounts for a vast proportion of the market capitalization of the standard value-weighted OMX Helsinki index. Hence, this value-weighted market index is not considered an appropriate benchmark for the Finnish stock market. It should be noted that there are numerous possible variations of the original DoD-10 strategy. Besides altering the number of 'dog' stocks included in the portfolio, the strategy can be implemented, for instance, with alternative holding periods and portfolio formation months or with a value-weighted portfolio. We also examined the performance of five 'dog' stock (DoD-5) investment strategy using the same methodology. The results for this alternative DoD strategy are broadly consistent with the reported ones. Alternative strategies would be to invest the proceeds in the remaining nine DoD stocks or to move to the stock with the 11th highest dividend yield. Due to the 36-month estimation window and data unavailability, the Fama–French adjusted returns are used in the analysis only over the period 1991–2008. Furthermore, because of insufficient data, the model is estimated in the reduced form without the SMB factor for years 1991 and 1992. In interpreting the Fama–French adjusted returns, it is necessary to consider several limitations of our analysis. First, given the small number of publicly traded firms in Finland, the portfolios underlying the Fama–French factors are very thin. Thus, the constructed SMB and HML factors may not adequately capture the risks that they are designed to capture. Moreover, it should be noted that combining the thin factor portfolios with a relatively short 36-month estimation period obviously reduces the precision of the estimated factor loadings. A further limitation is that the SMB factor portfolios exclude dividend payments and may thereby overestimate the small firm premium if dividend yield is positively correlated with firm size. Consequently, the Fama–French adjusted returns should be interpreted with some caution. An interesting extension of the analysis would be to examine whether the performance of the DoD strategy can be attributed more to the poor past returns or to lagged dividend yields.