The Social Media Risk Premium (with Amin Hoseini, Gergana Jostova, and Robert Savickas)

Abstract
Using novel corporate Twitter data on all U.S. public firms, we show that firms with a Twitter account earn 50 basis points per month higher returns than similar firms without a Twitter account. This `Twitter premium' is higher among smaller firms and firms with higher fundamentals uncertainty, and is not explained by existing risk-factor models. Having a Twitter account presents opportunities for value creation but also raises social media risks. We show that a social media risk factor is priced in the cross-section of U.S. stock returns and carries a premium of 30 to 75 basis points per month controlling for other risk factors.

Analyst Bias and Mispricing (with Mark Grinblatt and Gergana Jostova)

Abstract
Predictable biases in analyst forecasts, both conservative and optimistic, distort share prices, but only for firms with hard-to-forecast earnings---those with extreme past returns, credit risk, idiosyncratic volatility, and other attributes linked to 14 popular anomalies. The prevalence of analyst optimism (and the rarity of analyst conservatism) among these firms emerges as a likely explanation for their overpricing and subsequent negative alphas. The profitability of anomaly strategies disappears once we account for analyst bias.

Shorting Fees, Private Information, and Equity Mispricing (with Brian Henderson and Gergana Jostova)

Abstract
Decomposing lending fees into predicted (fair) and residual (premium or discount) fees reveals overpricing among a third of hard-to-borrow stocks: those for which borrowers pay a premium. Despite paying the highest fees, they are the only profitable shorters. Their net annualized profits of 5% reveal informed shorting. We also document smart lending. Lenders appear attuned to lending-market conditions, discounting fees on stocks with elastic shorting demand, thereby increasing revenues. Stocks with the most discounted fees attract the highest short interest, yet are predominantly easy-to-borrow. Their short sellers do not generate net shorting profits or appear motivated by private information.

Back to top