In the modern information economy, an increasing amount of information about capital market participants is available. Transparency regulation has facilitated this type of information. For example, short positions in Europe require unmasking the identity of the trader. Although prior research has studied the impact of revealing trader identities on capital markets, little is known about the information content of the identity itself. In our study, we consider whether traders consider the identity of other traders in their trading decisions.
We motivate our empirical study based on the idea that a trader’s identity, or reputation, could either inhibit or stimulate subsequent trading activity. A reputation for being informed should alert other investors to the existence of private information in the market. Private information makes markets less liquid, which in turn should suppress the inclination of others to trade (“illiquidity effect”). However, a reputation for being informed might lead other investors to place more weight on price movements resulting from informed trading, and this, in turn, might motivate more trading (“information effect”). We study which of these two countervailing forces dominates when a trader’s identity is revealed.
Because current U.S. regulations do not require traders to reveal their identity in the short window around a trade date, we consider the unique regulatory environment for short-selling in Europe. Beginning in 2012, all short positions in Europe above the 0.5% of share capital threshold, along with changes in these positions of greater than 0.1%, are required to be disclosed within a day of the change in short position. The intent of the regulation was to discourage abusive short-selling. This setting offers several advantages for our study. First, the disclosure is made shortly after a short seller executes a trade. Second, the disclosure reveals the short seller’s identity, which allows us to exploit variation in perceptions about the extent to which the short seller is informed, based on reputation. Third, these disclosures encompass many dates for many countries, short sellers, and firms, which allows us to design tests that address critiques that our findings can be explained solely by time-series or cross-sectional characteristics.
We use an event-study design to isolate the role of the short-seller identity from all other information in the short-position disclosure. Our ideal experiment compares information-based trades where traders’ identities are disclosed, to trades where identities are not revealed. However, in the pre-regulation period where there was no disclosure of traders’ identities, individual trades were also not disclosed. To control for the existence of a short-position disclosure, our research design holds constant the existence of a disclosure and compares disclosures made by better-informed short sellers to those made by less-informed short sellers. Thus, our design is equivalent to a difference-in-differences (DiD) design; we assess how changes in several outcome variables at the time of disclosure are associated with the reputation of the short seller whose identity is disclosed.
Our main finding is that, despite increased illiquidity associated with a trader’s identity revealing she is well-informed, reputation is associated with increased subsequent trade from other traders. Other traders glean information from trader identity and place more weight on price adjustments.
Intuition
The intuition for our study derives from how illiquidity arises in imperfectly competitive markets (e.g., Kyle, 1985). Prior to revealing a trader’s identity, the degree to which markets are illiquid should depend upon the extent to which traders who participate in the market are perceived to be privately informed on average. Post-disclosure, illiquidity should adjust to reflect a revised perception about the extent to which traders are privately informed. This revision is based on a specific trader’s reputation. For example, if a trader has a reputation for being well-informed (uninformed), illiquidity should increase (decline). We test whether the change in bid-ask spreads once a short seller’s identity is revealed increases with a proxy for the short seller’s informedness. This proxy is based on several measures: a local (same-country trader) indicator, higher returns from previous short sales, and a hedge fund or investment adviser indicator.
Revealing a trader’s identity should also affect assessments of the information content of prices. For example, if the trader has a reputation for being informed, revealing her identity should induce other investors to place more weight on prices in their expectations about the value of the asset being traded. We refer to this phenomenon as other investors perceiving those prices to have greater information content. Thus, the perception that a short seller is well-informed unleashes countervailing effects on subsequent trading volume. This perception should heighten adverse selection and transaction costs, which in turn should inhibit subsequent trading (“illiquidity effect”). That said, this perception should also induce other investors to place more weight on prices, and thereby motivate more subsequent trade (“information effect”). To assess which of the two effects dominates, we test whether the short seller’s reputation for being informed is associated with subsequent changes in stock price and trading activity. A positive association between the short seller’s reputation and subsequent changes in stock price and trading activity would imply that the information effect dominates the illiquidity effect.
It is possible that subsequent trading activity after disclosure of the short seller’s trade may not vary with the identity of the short seller. For example, a prior study finds that short sellers often hide their information in this setting by choosing to keep their position below the disclosure threshold. In addition, another prior study finds that privately informed individuals add noise to their trades when required to disclose their identity.
Findings
Despite the possibility of a null result raised by these prior studies, our findings support the notion that the identity of the short seller explains variation in bid-ask spreads. Although we cannot examine pre-2012 disclosures of trader identity or individual trades, as they do not exist, we are able to examine pre/post-disclosure regime short selling activity more broadly. We find that stocks with short sellers who had a reputation of being well-informed became more expensive to borrow after the mandatory identity-disclosure regime became effective. Institutions are less willing to lend shares to short sellers and therefore trading costs increase. Thus, in conjunction with our results showing the illiquidity effect, these results show that a reputation of being well-informed indeed makes short selling more costly.
After studying the illiquidity effect, we turn to assessing the relative magnitudes of the illiquidity effect and the information effect. We find evidence that the information effect dominates the illiquidity effect. The magnitude of returns and abnormal subsequent trading volume both increase as the trader is perceived to be better-informed.
We also find that the information gleaned from trader identity is associated with improved price informativeness. Firm-years with better-informed short sellers disclosing their identities tend to have returns that incorporate more information about future earnings.
Contributions
Our study makes several contributions. First, we contribute to a better understanding of the interaction among disclosure, adverse selection, and subsequent trading volume. Investor-identity disclosure has the interesting feature of unleashing countervailing effects on subsequent trading: the “illiquidity effect” and the “information effect.” This allows us to study which of the effects dominates—we find that the information effect dominates.
Second, we contribute to the literature on capital market effects that arise from informed trade. Traders can benefit from voluntarily revealing their identity to encourage other investors to move prices toward fundamental value. However, requiring a trader to reveal her identity could also reduce the profits she expects to earn, due to concerns about competitors reverse engineering trading strategies (we label foregone trading opportunities from the requirement that a trader reveal her identity “privacy costs”). We show that, conditional on a trade having been executed, the reputation revealed by a trader’s identity motivates subsequent trades.
Finally, we complement prior studies on the economic consequences of transparency about investor identity. Prior studies study the impacts of revealing short-positions, whereas we study whether other capital market participants consider the specific information in trader identity. Unlike in Europe, current regulations in the U.S. do not require that short sellers disclose either their trades or their identities. Nonetheless, the SEC has considered its feasibility. Changes in disclosure policy might arise in conjunction with ongoing amendments to U.S. regulations governing the Consolidated Audit Trail, which tracks and consolidates U.S. trading activity. Furthermore, the SEC has recently proposed increasing the reporting threshold for long positions on Form 13-F, which would mask the identity of several institutions. Our results suggest that a trader’s reputation motivates subsequent trades, despite the increased illiquidity associated with trader reputation. Although we do not claim to measure the trade-offs between stimulating more trading activity and a trader’s privacy costs from revealing identity, our results should be informative to both academics and regulators.
Christina Zhu is an Assistant Professor of Accounting at the The Wharton School, the University of Pennsylvania, Accounting Department.
Robert E. Verrecchia is the Elizabeth F. Putzel Professor Emeritus of Accounting at The Wharton School, the University of Pennsylvania.
This post is adapted from their paper, “Liquidity, Trade, and Investor-Identity Disclosure,” available on SSRN.
The views expressed in this post are those of the authors and do not represent the views of the Global Financial Markets Center or Duke Law.
