The value of stock analysts’ recommendations has been the subject of debate in the academics as well as the popular media. Although the press doesn’t seem to be much favorable to stock analysts in the midst of recent law suits against major brokerages, empirical evidence from academic literature documents that analysts’ recommendations do affect individual stock prices. This is generally taken as evidence against market efficiency, although the magnitude of the mispricing detected by analysts may not be large enough to provide arbitrage opportunities after trading costs.
The evidence in the stock market naturally leads us to question the extent to which higher moments of the stock price distribution such as volatility might be affected by analyst recommendations. There are two main ways of measuring volatility; using information on historical stock returns or backing out ‘implied’ volatilities from equity option market. Traditional view has been that if markets are perfect, in terms of information and frictions, and the prices for the underlying stocks follow a constant or deterministic volatility process, then options are redundant assets whose payoffs can be replicated exactly by other fundamental securities. Under such circumstances, option prices are completely characterized by underlying stock prices and the effect of analyst recommendations on options would be determined only through changes in the stock prices.
However, a growing body of literature recognizes the possibility that options are not just simple redundant securities. The theoretical background for the possible non-redundancy has been focused on information asymmetry, suggesting that informed traders may prefer to trade in the options due to various reasons such as leverage effect, first suggested by Black (1975). Subsequent research has attempted to identify indications of informed trading in the option market, mainly through option volume. Empirical evidence is still somewhat mixed as to whether option or stock market leads information trading, although more recent works on this subject seem to provide evidence in favor of information trading in the option market. A more radical behavioral approach in support of non redundancy is that stock markets and option markets are segmented in terms of the type of investors that participate in each market. Generally, option market participants are assumed to be more sophisticated than stock market participants in that they are less affected by investor sentiment. If so, option market should exhibit less overreaction or underreaction to a specific event, compared to the underlying stock market.
In this paper, I investigate potential underreaction or overreaction and the presence of information trading in the options market by comparing the changes in implied volatility against realized volatility around analyst recommendation revisions. Analyst recommendations are distinct from other major corporate announcements such as earnings announcements or stock splits. Unlike earnings announcements which are generally scheduled in advance, analyst recommendations are unanticipated. And unlike stock splits which seldom occur, recommendation revisions are frequently observed across a wide range of stocks.
Download
Analyst recommendations and option market reactions
