Working Papers
Working Papers
In a multi-period investment setting with a time-varying investment opportunity set, an asset's risk premium is determined by its covariance with the market and the investment opportunity set. This paper develops a measure of the investment opportunity set. The measure is a linear function of characteristic-sorted portfolio excess returns with time-varying coefficients which are estimated using an ICAPM implication, using past data and without using macro variables. I validate this measure by investigating its relationship with macro variables, return predictability, unconditional and conditional pricing power in the universe of characteristic-sorted portfolios, U.S. common stocks and U.S. industry portfolios.
I develop an asset pricing model with agents who have heterogeneous investment horizons. Differently from short-term investors, long-term investors hedge against reinvestment risk. In the model, risky investment opportunity is characterized as a scaled expected return of the tangency portfolio. I measure the unobservable risky investment opportunity by approximating the risky return space by Fama-French 5 factors and momentum. I construct long-short portfolios that are exposed to reinvestment risk premiums and find that they have significant positive average and risk-adjusted returns. I also test the portfolio choice implication of the model using 13F data by measuring an institution's investment horizon by its portfolio turnover rate and type. I find that long-term institutions overweight hedging assets relative to the market portfolio, which suggests that (i) investment opportunities are well estimated by the method developed in this paper and (ii) long-term investors' intertemporal hedging demand is significantly priced among assets.
(with Matthew Richardson)Â
(Revise & Resubmit, Journal of Financial Economics)
Greenwood, Shleifer and You (2018, GSY) investigate Eugene Fama's claim that stock price bubbles are difficult to identify. GSY (2018) document a series of stylized facts that put into question Fama's view on bubbles. We reinterpret these empirical facts and find much more support for Fama than implied by GSY (2018). In particular, we show theoretically and then document empirically that many of GSY's (2018) stylized "bubble" facts are expected in a world without bubbles.