Analysts’ Stock Ownership and Stock Recommendations

Yong Yu is Professor at the McCombs School of Business at the University of Texas at Austin. This post is based on a recent article, forthcoming in the Journal of Accounting and Economics, by Professor Yu; Jesse Chan, Ph.D. student at the McCombs School of Business at the University of Texas at Austin; Steve W.J. Lin, Knight Ridder Eminent Scholar Chair in Accounting at Florida International University; and Wuyang Zhao, Assistant Professor at the McCombs School of Business at the University of Texas at Austin.

Financial analysts’ stock ownership in the companies they cover has long been controversial. Regulators and the financial press have repeatedly pointed to analysts’ stock ownership as a potential source of conflicts of interest and have publicly warned investors that analysts’ stock ownership can impair the objectivity of their recommendations. According to this view, analysts’ stock ownership directly links analysts’ personal wealth to their own recommendations, creating a conflict of interest between their responsibility to provide clients with unbiased recommendations and their incentive to maximize the value of their own investments. However, proponents of analysts’ stock ownership stress that analysts’ stock ownership can enhance the credibility of their recommendations by putting their money where their mouth is. In line with these competing views, investors’ opinions differ greatly. Brokers also vary substantially in their policies, with some banning analysts’ stock ownership altogether while others allow or even encourage it.

In our article, Analysts’ Stock Ownership and Stock Recommendations, forthcoming in the Journal of Accounting and Economics, we explore a large, hand-collected sample of research reports issued by analysts who own stock in the firm to shed light on this ongoing debate. New regulations on sell-side research (NASD Rule 2711, 2008 and NYSE Rule 472, 2007) in 2002 mandate that analysts disclose in research reports whether they, or any household members, own any shares in the firms they cover. This mandated ownership disclosure allows us to identify reports issued by analysts who own shares in the firm (though we cannot observe the magnitude of analysts’ holdings). We check disclosures pertaining to analysts’ stock ownership in reports over 2003–2012 from 74 brokers covered by both ThomsonOne and IBES databases and identify 29,108 reports containing a disclosure that analysts or their family members own stock in the firm (i.e., ownership reports).

In an average sample year, 13% of analysts own stock in covered firms, and among these ownership analysts, they invest in 18% of all covered firms. Additional data collection reveals that 12% of analysts own stock in covered firms in 2016, indicating that analysts’ stock ownership continues to be prevalent today. Analysts’ stock ownership is associated with more favorable recommendations. 66% of ownership reports contain a favorable (buy) recommendation, compared to 44% for all recommendations from the same broker-periods. Analysts’ stock ownership tends to be long-term. The average ownership string (i.e., a series of consecutive ownership reports) contains 13 reports and spans about 16 months. Fourth, among the 74 brokers in our sample, 60 allowed analysts’ ownership. In contrast, the remaining 14 had virtually no ownership reports and thus likely banned it. Compared with analysts without ownership, analysts owning stock are more experienced, cover more firms and more industries, provide fewer types of forecasts, have higher prior forecast frequency and accuracy, and are more likely to work at brokers that are smaller and have no underwriting relationship with the firm. They tend to invest in firms that are larger, more mature, have higher growth, better financial performance, greater analyst following, and higher institutional ownership.

Examining stock market reactions to ownership versus non-ownership recommendations, we find that favorable ownership recommendations trigger significantly stronger market reactions than favorable non-ownership recommendations. This incremental reaction associated with analysts’ stock ownership becomes even larger for coverage initiations and upgrades. These results suggest that ownership buy recommendations are more informative, consistent with analysts’ ownership enhancing the credibility of their recommendations. We then examine analysts’ effort as a potential mechanism through which analysts’ stock ownership can lead to superior information. Using both forecast frequency and timeliness to capture effort, we find that analysts’ stock ownership is associated with more effort in covering the firm. This result is consistent with the notion that having “skin in the game” motivates analysts to exert more effort in covering the company, which helps them obtain superior information.

We also analyze the trading behavior of analysts who own stock in covered firms, based on over-time changes in analysts’ ownership status (the presence or absence of stock ownership). We identify 822 initial share purchases (initiations) and 763 final share sales (terminations) to test two commonly voiced concerns on analysts’ stock trading. First, we examine whether analysts trade against their own recommendations—a trading behavior explicitly banned by the regulations on analyst research. We find that 3% of analysts initiated ownership while having a sell recommendation outstanding, and 56% terminated ownership while having a buy recommendation outstanding. This finding is surprising, suggesting a potentially widespread violation of the regulations, though we cannot rule out the possibility of unanticipated financial hardships—the only exception allowed by the regulations. Further tests provide some weak evidence that experienced and all-star analysts are more likely to be involved in such trading behavior in firms with lower institutional ownership. Second, we examine whether analysts trade upon information before releasing it to investors. If an analyst buys (sells) before releasing positive (negative) information, we expect upward (downward) revisions of recommendations or forecasts after ownership initiations (terminations). In general, we do not find clear evidence of such patterns, except for some weak evidence that analysts revise earnings forecasts downward in the 30-day period immediately after they terminate ownership. However, we caution readers that an important caveat for this test is low power, because we do not have the exact date for ownership initiations or terminations and analysts may purchase most shares after initiations and sell most shares before terminations.

Our findings contribute to the long-standing debate over analysts’ stock ownership in covered firms. They suggest that analysts’ stock ownership can have both positive and negative effects on analysts’ research. In line with the conflict-of-interest view, we find evidence that analysts’ stock ownership induces analysts to issue more upward-biased target price forecasts and to trade against their own recommendations, especially when unloading their holdings. However, consistent with the credibility enhancement view, we also find evidence that analysts’ stock ownership increases the informativeness of their stock recommendations by conveying their superior information and induces analysts to spend more effort in covering the firm. Our findings provide insights potentially relevant to investors, brokers and regulators who are concerned with the effects of analysts’ stock ownership on analysts’ research.

The complete article is available for download here.

Both comments and trackbacks are currently closed.