The paper investigates the strategic behavior of hedge fund families. It focuses on decisions to start and liquidate family-member funds. Hedge fund families tend to liquidate funds that underperform compared to other member funds, and to replace them by new ones. By choosing a launch time after a short period of superior performance by their member funds, families extend the spillover to new funds. Hedge fund families seem to be more experienced in promoting their funds and attracting fund inflow than in generating superior performance. This results in higher dollar compensation earned by managers within multi-fund families than in stand-alone funds.

We find no evidence of stock price manipulation by hedge funds from 2011 to 2019, despite confirming the portfolio-pumping pattern documented between 2000 and 2010. In the more recent period, the magnitude, frequency, and persistence of manipulation by hedge funds appear to have declined. This decrease is linked to reduced rewards, as fund flows no longer react positively to the end-of-quarter returns of hedge fund portfolios. Proactive regulatory actions, measured by SEC litigation cases involving hedge funds, and increased press attention to hedge fund fraud also contribute to reduced manipulation during both periods.

Constructing portfolios based on second-order stochastic dominance (SSD) is theoretically attractive since all risk-averse investors would prefer a dominating portfolio. However, choosing among SSD efficient portfolios is a challenge without an obvious ranking metric. We explore particular choices based on Kuosmanen (2004) plus Kopa and Post (2011), comparing their performance to other SSD-related strategies and to standard portfolio choice approaches. These SSD-related choices outperform portfolios chosen based on their Sharpe ratio, information ratio, or using equal weights. Portfolios based on minimum variance that also match the benchmark’s mean return perform on a par with the SSD-related choices.

Numerous hedge funds stop reporting each year to commercial databases, wreaking havoc with analyzing investment strategies that incur the unobserved delisting return. We use estimated portfolio holdings for funds-of-funds to back out estimated hedge-fund delisting returns. For all exiting funds, the estimated mean delisting return is insignificantly different from the average monthly return for live hedge funds. However, funds with poor prior performance and no clearly stated delisting reason had a significantly negative estimated mean delisting return of -5.97%, suggesting that a shock to their returns “tips them over the edge” and leads to delisting.

This paper investigates the information content of aggregate hedge fund flow and its predictive power with respect to bond yields. Using a sample of 9725 hedge funds from 1994 to 2012, we find that fund flow is negatively related to the changes in 10-year Treasury and Moody’s Baa bond yields one month ahead. This relation is still pronounced after controlling for other determinants of yield changes, including the amount of arbitrage capital available in the economy, suggesting a non-trivial effect of flow-induced hedge fund trading on bond yields. Flow impact on corporate bonds is further amplified during periods of decreasing market liquidity, consistent with a fire-sale hypothesis. Hedge fund flow also predicts convergence between constant maturity swap rate and constant maturity Treasury rate, as well as between the TIPS and Treasury bond yields, suggesting that hedge funds exploit arbitrage opportunities in these fixed-income markets.