Investments, tenth edition

  Anomalies over Time   

We pointed out earlier that while no market can be perfectly 

efficient, in well-functioning markets, anomalies ought to be self-destructing. As market 

participants learn of profitable trading strategies, their attempts to exploit them should 

move prices to levels at which abnormal profits are no longer available. Chordia, Sub-

ramanyam, and Tong  

42

   look for this dynamic in the pattern of many of the anomalies 

discussed in this chapter. They focus on abnormal returns associated with several char-

acteristics including size, book-to-market ratio, momentum, and turnover (which may 

be inversely related to the neglected firm effect). They break their sample at 1993 and 

show that the abnormal returns associated with these characteristics in the pre-1993 period 

largely disappear in the post-1993 period (with the notable exception of the book-to-

market effect). Their interpretation is that the market has become more efficient as knowl-

edge about these anomalies percolated through the investment community. Interestingly, 

they find that that the attenuation of alphas is greatest in the most liquid stocks, where 

trading activity is least costly.

  

 McLean and Pontiff   

43

   provide further insight into this phenomenon. They identify more 

than 80 characteristics recognized in the academic literature as associated with abnormal 

returns. Rather than using a common break point for all characteristics, they carefully track 

both the publication date of each finding as well as the date the paper was first posted to the 

  

42

 T. Chordia, A. Subrahmanyam, and Q. Tong, “Trends in the Cross-Section of Expected Stock Returns” (May 2, 

2012). Available at SSRN:   http://ssrn.com/abstract=2029057   or   http://dx.doi.org/10.2139/ssrn.2029057.     

  

43

 David R. McLean and Jeffrey E. Pontiff, “Does Academic Research Destroy Stock Return Predictability?” 

(October 3, 2012). AFFI/EUROFIDAI, Paris, December 2012 Finance Meetings Paper. Available at SSRN: 

  http://ssrn.com/abstract=2156623   or   http://dx.doi.org/10.2139/ssrn.2156623.     

bod61671_ch11_349-387.indd   373

bod61671_ch11_349-387.indd   373

7/17/13   3:41 PM

7/17/13   3:41 PM

Final PDF to printer

---

374 

P A R T   I I I

  Equilibrium in Capital Markets

Social Science Research Network. This allows them to break the sample for each finding 

at dates corresponding to when that particular finding became public. They conclude that 

the post-publication decay in abnormal return is about 35% (e.g., a 5% abnormal return 

pre-publication falls on average to 3.25% after publication).  

44

   They show that trading vol-

ume and variance in stocks identified with anomalies increases, as does short interest in 

“overpriced” stocks. These patterns are consistent with informed participants attempting to 

exploit newly recognized mispricing. Moreover, the decay in alpha is most pronounced for 

stocks that are larger, more liquid, and with low idiosyncratic risk. These are precisely the 

stocks for which trading activity in pursuit of reliable abnormal returns is most feasible. 

Thus, while abnormal returns do not fully disappear, these results are consistent with a 

market groping its way toward greater efficiency over time.