This paper investigates the role of credit risk in the relationship between short-selling activity and future stock returns. We find that the predictive power of short interest for future returns is concentrated in the worst-rated stocks. Low-grade stocks with the largest short interest decrease outperform those with the largest short interest increase by 1.09 percent in the following month. This return spread is robust to controls for cross-sectional effects and firm characteristics, and is much more pronounced during periods of high investor sentiment and low liquidity. Distressed firms with large short interest increases experience a worse performance subsequently.

We explore the use of technical indicators to forecast corporate bond returns with various machine learning models. We show that technical indicators yield statistically significant and economically meaningful results, consistently outperforming bond characteristics. Although bond characteristics possess predictive power for bond returns, they do not provide incremental value beyond technical indicators across all bonds. Additionally, machine learning models do not offer substantial improvements over the benchmark linear model. These results underscore the significance of technical indicators in the corporate bond market.

This paper investigates the impact of institutional herding on stock price formation, conditional on firms’ credit ratings, using 13F data from 1986 to 2019. In line with the current literature, we find herding intensity is driven by past returns consistent with momentum trading; however, we also find that herding is more sensitive to past returns for non-investment grade (NIG) stocks than investment grade (IG) stocks, resulting in a market bifurcation. We then examine the price impact of these trades and find that herding in NIG equities enhances price discovery. One plausible explanation is that information gradually diffuses within non-investment grade stocks, and herding behavior strengthens information discovery. Finally, we show both momentum-triggered herding and non-momentum-triggered herding contribute to price discovery among non-investment grade stocks.