A long-standing anomaly for efficient markets has been what’s called “post-revision return drift” (PRD). Research into stock returns has found that changes in sell-side analyst recommendations for buying and selling stocks predict future long-term returns in the same direction as the change. Upgrades are followed by positive returns, and downgrades are followed by negative returns.
To explain the anomaly, it’s been hypothesized that securities analysts are better informed and more skillful than the investing public in general. In addition, PRD persists because investors typically underreact to analysts, responding only partly upon their revision announcements and slowly thereafter, perhaps taking months.
In other words, the market is inefficient, allowing for more sophisticated investors to exploit the more naive public.
Do Analysts Have An Edge?
Oya Altinkilic, Robert Hansen and Liyu Ye contribute to the literature on the PRD anomaly with their paper, “Can Analysts Pick Stocks for the Long-Run?”, which appeared in the February 2016 issue of the Journal of Financial Economics. The samples used in the study draw from revisions from the First Call Historical Database (FCHD) for the period 1997 through 2010.
The authors split the full period into two parts: a pre-sample period from 1997 through April 2003, and a post-sample period from May 2003 through 2010. The performance measure is based on the direction of analysts’ revisions and is called the “up-less-down” strategy. It seeks to capture the difference between PRD after upgrades and PRD after downgrades. This is a strategy that invests long in the upgraded stocks and short in the downgraded stocks. Following is a summary of their key findings:
- While there was a statistically significant (at the 1% level) positive PRD over the 120-day period after revisions are announced in the pre-sample period, in the post-sample period, the up-less-down PRD turned slightly negative—the anomaly disappeared. Interestingly, the long leg continued to earn positive returns. However, the short leg earned negative returns. This is inconsistent with the theory that downgrades anticipate lower prices in the future. (Note that in neither the long nor the short leg were the returns statistically significant.)
- Tests of the informed analyst hypothesis employing proxies for better-informed analysts used in prior research do not support the idea that analysts typically supply new information that correctly picks stocks for the long run.
- Transaction costs are high enough to fence PRD from profitable arbitrage-trading strategies.
- Under the transaction cost rationale, some PRD is present for stocks with relatively high transaction costs. After sorting the sample into trading-volume deciles, some statistically significant average PRD exists in the lowest decile, or 10%, of the revisions.
- Out-of-sample tests in seven developed countries (the U.S., Canada, France, Germany, Italy, Japan and the U.K.) confirm a general absence of PRD in the post-period.
The authors also observed that PRD is frequently associated with other recent news and events about the firm. They found that, after controlling for this, PRD contains little incremental information that can be credited to analysts in the post-period.
The authors concluded the “results agree with the explanation that PRD has broadly vanished due to a general decline in transaction costs, pushed down to historic lows by decimalization, the expanded use of supercomputers, and algorithmic trading. The PRD disappearance coincides with notable reductions in transaction costs that have attracted profit-taking arbitrageurs to PRD.”
These new findings contribute to research related to market efficiency because the results show that, on average, analysts’ revisions are not highly correlated with subsequent long-run returns. This indicates that analysts don’t provide new information that is relevant for the long run for typical investors.
The authors explained: “Analysts face greater competition in the market for new information in the post-period, as lower transaction costs have enabled arbitrageurs to quickly harvest more mispricing opportunities, shrinking the potential pool of neglected information.”
As Nobel Prize-winner Eugene Fama noted in his famous 1970 paper, “Efficient Capital Markets: A Review of Theory and Empirical Work”: “Markets are efficient as long as agents cannot reliably predict long run common stock returns.”
The findings from the study by Altinkilic, Hansen and Ye support the view that markets have become more efficient over time as transaction costs have declined (allowing information to be incorporated more quickly and completely into security prices), thus eliminating profit opportunities from strategies that use analysts’ revisions, as well as the predictability of long-run returns based on those revisions.
Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.