Richard W. Sias

Herding and Feedback Trading by Institutional and Individual Investors

We document strong positive correlation between changes in institutional ownership and returns measured over the same period. The result suggests that either institutional investors positive-feedback trade more than individual investors or institutional herding impacts prices more than herding by individual investors. We find evidence that both factors play a role in explaining the relation. We find no evidence, however, of return mean-reversion in the year following large changes in institutional ownership—stocks institutional investors purchase subsequently outperform those they sell. Moreover, institutional herding is positively correlated with lag returns and appears to be related to stock return momentum. HERDING AND FEEDBACK TRADING HAVE THE POTENTIAL to explain a number of financial phenomena, such as excess volatility, momentum, and reversals in stock prices. Herding is a group of investors trading in the same direction over a period of time; feedback trading involves correlation between herding and lag returns. 1 Although a recent growing body of literature is devoted to investor herding and feedback trading, extant studies take divergent paths. One path depicts individual investors as engaging in herding as a result of irrational, but systematic, responses to fads or sentiment. A second path depicts institutional investors engaging in herding as a result of agency problems, security characteristics, fads, or the manner in which information is impounded in the market.

Voting with their feet: institutional ownership changes around forced CEO turnover

Abstract We investigate whether institutional investors “vote with their feet” when dissatisfied with a firms management by examining changes in equity ownership around forced CEO turnover. We find that aggregate institutional ownership and the number of institutional investors decline in the year prior to forced CEO turnover. However, selling by institutions is far from universal. Overall, there is an increase in shareholdings of individual investors and a decrease in holdings of institutional investors who are more concerned with holding prudent securities, are better informed, or are engaged in momentum trading. Measures of institutional ownership changes are negatively related to the likelihoods of forced CEO turnover and that an executive from outside the firm is appointed CEO.

Return autocorrelation and institutional investors

Abstract We propose and test the hypothesis that trading by institutional investors contributes to serial correlation in daily returns. Our results demonstrate that NYSE particles and individual security daily return autocorrelationsare an increasing function of the level of institutional ownership. Moreover, the results are consistent with the hypothesis that institutional trading reflects information and increases the speed of price adjustment. The relation between autocorrelation and institutional holdings does not, however, apparent to be driven by market frictions or rational time-varying required rates of return. We conclude that institutional investors correlated trading patterns contribute to axial correlation in daily returns.

Changes in Institutional Ownership and Stock Returns: Assessment and Methodology

Although the relation between quarterly changes in institutional investor ownership and contemporaneous stock returns is well documented, the source of the relation remains unclear because institutional ownership data are unavailable at higher frequencies. In this study, we develop a method to generate estimates of higher frequency covariances when one variable is observed at lower frequencies (e.g., quarterly changes in institutional ownership and monthly stock returns). Our method provides evidence that institutional trading has both temporary and permanent price effects and that the latter is associated with information effects.

Causes and Seasonality of Momentum Profits

With Januaries (a month in which lagged “losers” typically outperform lagged “winners”) excluded, the average monthly return to a momentum strategy for U.S. stocks was found to be 59 bps for non-quarter-ending months but 310 bps for quarter-ending months. The pattern was stronger for stocks with high levels of institutional trading and was particularly strong in December. The results suggest that window dressing by institutional investors and tax-loss selling contribute to stock return momentum. Investors using a momentum strategy should focus on quarter-ending months and securities with high levels of institutional trading. Stocks exhibit return momentum: Lagged “winners” (i.e., securities in the top performance decile based on returns over the previous six months) tend to subsequently outperform lagged “losers” (i.e., securities in the bottom lagged six-month performance decile). Both tax-loss selling in December and window dressing by institutional investors in quarter-ending months may contribute to stock return momentum. In the case of tax-loss selling, investors (both individual investors and some institutional investors) may favor selling lagged losers in December to realize taxable losses and may avoid selling lagged winners in December to forestall recognizing taxable gains. This behavior may contribute to return momentum in December. In the case of institutional investors, their window-dressing behavior also may contribute to momentum-profit seasonality. At quarter-end, and especially year-end, institutional investors may want to abandon lagged losers to avoid reporting “embarrassing” stocks in their end-of-quarter or end-of-year holdings. Similarly, managers may buy lagged winners to appear as if they held respectable or “winning” stocks throughout the period. This study found that the profitability of momentum strategies in the past 20 years arose primarily from the last month of each quarter, which is consistent with the hypothesis that year-end tax-motivated trading and institutional window dressing contribute to stock return momentum. Momentum profits were, on average, negative when quarter-ending months (March, June, September, and December) were excluded from the sample. January, a month when lagged losers typically outperform lagged winners, explains part of the negative momentum profits for non-quarter-ending months. Even after excluding January from the sample, however, momentum profits from quarter-ending months averaged more than five times the momentum profits from non-quarter-ending months. The seasonal pattern was particularly strong in stocks with high levels of institutional trading and in December. The momentum-profit seasonality and the relationship between this seasonality and institutional trading suggest that tax-loss selling and institutional window dressing play substantial roles in driving stock return momentum. Investors attempting to exploit return momentum should thus focus their efforts on quarter-ending months and on securities with high levels of institutional trading.

Why Company-Specific Risk Changes over Time

Company-specific risk climbed steadily between 1962 and 1999 in the U.S. market but fell sharply between 2000 and 2003. This article explores the hypothesis that three factors are primarily responsible for observed changes in company-specific risk: changes in the market weights of “riskier” industries, changes in the relative role of small-capitalization stocks in the market, and measurement error associated with changes in within-industry concentration. Empirical tests reveal that each factor contributes to changes in company-specific risk over time and that, combined, these three factors largely explain changes in company-specific risk over the past 40 years. Recent studies have demonstrated that company-specific risk steadily increased between the early 1960s and the late 1990s. Since the market peak in 2000, however, company-specific risk has exhibited a secular decline. These changes are important to active managers because they have an impact on the effectiveness of portfolio diversification, tracking error, return dispersion across managers, and the ability of traders to exploit mispriced securities and because some recent research suggests company-specific risk may be rewarded (i.e., priced). A number of potential explanations have been offered that suggest fundamental changes in the economy and/or markets as the explanation for the rise in company-specific risk over time. Suggestions include decreases in operational diversification as companies narrowed their product/market focus, increased use of stock options as compensation, a rise in the volatility of returns on equity or growth opportunities, a decline in financial reporting quality, an increase in the role of institutional investors in the market and their tendency to herd, increases over time in levels of informed trading, an increase in capital market openness, and an increase in competition between companies. In the study reported here, we proposed and tested an alternative explanation for changes in aggregate company-specific risk: Carrying out tests for the August 1962–December 2003 U.S. market, we found that company-specific risk changes over time not as a result of fundamental changes in the market but, rather, as a result of changes in the composition of the securities that make up the market. Specifically, we proposed that three key changes in the composition of the market explain changes in company-specific risk over time: changes in the relative importance of industries, changes in the relative importance of small companies in the market, and changes in measurement error induced by changing within-industry concentration. Distinguishing between these explanations is important in practice because our explanation suggests that changes in company-specific risk faced by a given manager will be a function of the changes in that manager’s portfolio. For example, if a manager does not have great exposure to small-capitalization stocks, the rise in aggregate company-specific risk attributed to the growth of small-cap stocks in the market will not affect that manager. Empirical tests support each of our three hypotheses. Changes in within-industry concentration and the relative roles of small-cap stocks and riskier industries largely explain the patterns in company-specific risk over time. We conclude that these three factors are the primary culprits behind the long rise and recent decline of company-specific volatility over the past 40 years.

Hedge Fund Crowds and Mispricing

Recent models and the popular press suggest that large groups of hedge funds follow similar strategies resulting in crowded equity positions that destabilize markets. Inconsistent with this assertion, we find that hedge fund equity portfolios are remarkably independent. Moreover, when hedge funds do buy and sell the same stocks, their demand shocks are, on average, positively related to subsequent raw and risk-adjusted returns. Even in periods of extreme market stress, we find no evidence that hedge fund demand shocks are inversely related to subsequent returns. Our results have important implications for the ongoing debate regarding hedge fund regulation.

Can Money Flows Predict Stock Returns

“Money flow” is defined as the difference between uptick and downtick dollar trading volume. Despite little published research regarding its usefulness, the measure has become an increasingly popular technical indicator. Our analysis demonstrates that money flows are highly correlated with same-period returns. We also found strong evidence of “money flow momentum,” in that lagged money flows can be used to predict future money flows. Most important is our finding that money flows appear to predict cross-sectional variation in future returns. Their predictive ability is sensitive, however, to the method of money flow measurement (e.g., the exclusion or inclusion of block trades) and the forecast horizon. “Money flow,” defined as uptick dollar volume minus downtick dollar volume, is a technical indicator first developed in the 1970s. Although the measure has become increasingly popular in recent years, its efficacy has not yet been rigorously tested. In this research, we provide an analysis of the relationship between money flow and returns in the U.S. stock market. Our results suggest that money flow, measured properly, can assist portfolio managers in the security selection process. We began the study by computing three types of daily money flow for NYSE-listed companies: money flow based on all trades, money flow based on block trades, and money flow based on nonblock trades. We normalized each by dividing by the corresponding dollar volume of trading (e.g., normalized block money flow was calculated as block money flow divided by block volume) to create a total of six money flow measures. Although we found each of the six money flow measures to be positively correlated with same-period returns, we found normalized nonblock money flows to be most strongly related to returns. Given the strength of the relationship between contemporaneous money flows and returns, we next examined the predictability of money flows. Using cross-sectional regressions, we found that money flow displays strong persistence: Companies with high money flow in the past tend to have high money flow in the future. Specifically, we found positive relationships between cumulative money flow measured over 1-, 5-, 10-, 20-, 30-, and 40-day periods and lagged money flow measured over the past 1, 5, 10, 20, 30, and 40 days. The strength of the relationship increased with the interval length; that is, longer-term money flow exhibited greater predictability than shorter-term money flow. We show that money flow persistence can be exploited through the formation of stock portfolios that subsequently experience high money flows. We next examined whether money flows can be used to predict returns. Using cross-sectional regressions, we examined the relationships between future returns and past money flows (with combinations of past and future periods of various intervals). Our results reveal that future returns are positively related to past money flows. As when predicting money flows, the strength of the relationship increases with the length of the period over which money flows and returns are measured: Money flows measured over 40 days predict subsequent 40-day returns better than money flows measured over 5 days predict subsequent 5-day returns. Moreover, controlling for past money flows, we found that past returns contain little useful information regarding future returns but that past money flows, even after we controlled for past returns, do contain useful information for predicting future returns. As a final test of the usefulness of money flow, we formed portfolios of money flow “winners” and “losers.” We used 10-, 20-, 30-, or 40-day measurement periods to form the winners and losers and held the portfolios for the subsequent 10, 20, 30, or 40 days. Using the 16 possible combinations of measurement period plus holding period for each of the six money flow measures, we examined the outcomes of the self-financing strategy of taking long positions in money flow winners and short positions in money flow losers. The winner portfolios outperformed the loser portfolios in 74 of the 96 cases we examined. In summary, money flow contains information beyond the information contained in returns and can be a useful tool in security analysis and portfolio management.

Who are the Sentiment Traders? Evidence from the Cross-Section of Stock Returns and Demand

Recent work suggests that sentiment traders shift from safer to more speculative stocks when sentiment increases. Exploiting these cross-sectional patterns and changes in share ownership, we find that sentiment metrics capture institutional rather than individual investors’ demand shocks. We investigate the underlying economic mechanisms and find that common institutional investment styles (e.g., risk management, momentum trading) explain a significant portion of the relation between institutions and sentiment.

Style Timing with Insiders

Aggregate demand by insiders predicts time-series variation in the value premium — between 1978 and 2004, a one standard deviation increase in aggregate insider demand in the previous six months forecasts a 53 basis point decline (6.54% annualized) in the expected value premium in the month following publication of the insider trading data. Further tests suggest that insider trading forecasts the value premium because insiders trade against systematic investor sentiment-induced mispricing and growth stocks are more sensitive to changes in sentiment than value stocks, i.e., insiders sell (buy) when markets, and growth stocks especially, are overvalued (undervalued). As a result, our analysis suggests that investors can use signals from aggregate insider behavior to adjust style tilts and exploit sentiment-induced mispricing.