Academic journal article Quarterly Journal of Business and Economics

Delayed Reaction in Stocks with the Characteristics of Past Winners: Implications for Momentum, Value, and Institutional Following

Academic journal article Quarterly Journal of Business and Economics

Delayed Reaction in Stocks with the Characteristics of Past Winners: Implications for Momentum, Value, and Institutional Following

Article excerpt

Drew B. Winters [*]

The common characteristics of extreme past winners form a unique synthesis of the return-momentum, earnings-momentum, and value strategies. We find that stocks that satisfy these characteristics produce significant positive abnormal returns for up to one year in a NYSE/Amex sample and for up to two years in a NASDAQ sample. The more prolonged accumulation of abnormal returns in the NASDA Q stocks relative to the NYSE and Amex stocks and the lack of any observable return reversal support delayed reaction as an explanation for the abnormal returns. In addition, we find the institutional response is quicker in the NYSE/Amex sample than in the NASDAQ sample. We interpret this evidence as consistent with institutions contributing to the gradual correction of underreaction in these stocks. The magnitude of the abnormal returns and their association with institutional following also implies that the returns to momentum and value strategies reported in the recent literature are at least in part due to delayed reaction.

Introduction

The efficient market hypothesis (EMH) has been questioned in numerous papers that isolate one particular firm characteristic and then detect associated abnormal returns. For example, abnormal returns have been observed in stocks selected on the basis of momentum, value, institutional following, contrarian returns, and firm size.

Although that work casts doubt on the efficient market hypothesis, risk misspecification has not been rejected as an alternative explanation. In addition, it is not clear that the apparent opportunities are large enough to survive transaction costs.[1] But if the objective is to determine whether material market inefficiencies exist and, if so, the conditions under which they arise, then a shortcoming of these papers is that they presuppose where one should look. It is possible that some other characteristics, or a particular combination, could earn even higher returns.

Reinganum (1988) uses a unique approach to address this shortcoming. Instead of a priori isolating characteristics, he selects winning stocks that at least doubled in price within a calendar year and identifies their leading characteristics. He finds these winners had nine characteristics in common prior to their price appreciation. His test of their predictive power reveals that stocks that satisfy these characteristics earn a mean return that accumulates over two years to a dramatic 50.7 percent (in excess of the contemporaneous return on the S&P 500). While Reinganum's findings are widely cited in textbooks, little consideration has been given to why the strategy earns such high returns. [2]

In this paper, we replicate the strategy and document that it earns significant and material long-horizon returns that survive risk-adjustment. Given this finding, we are interested in why the strategy appears to work, as well as the implications for the market efficiency literature. In particular, is the return pattern due to overreaction or delayed reaction, and do institutions contribute to, or reduce, the inefficiencies implied in the pattern? Studying the behavior of institutions in these stocks adds to our understanding of why the return pattern exists, as well as the opportunities it may present to investors.

Why does the strategy produce high returns? Clues come from three observations regarding the original winners:

* Eight of the nine common characteristics are consistent with momentum in either past returns or earnings;

* The other characteristic is the well-known value indicator of Benjamin Graham (1973), market-to-book ratio of less than one; and

* The number of institutional investors in these winners did not increase prior to the price appreciation but more than doubled during the subsequent appreciation.

The relevance of these observations is more evident today, in light of newer findings. …

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

Oops!

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.