Charles M. Jones
Uris Hall, Room 801
New York, NY 10027
Tel: 212 854 4109
Institutional Affiliation: Columbia University
NBER Working Papers and Publications
|October 2004||Do Stock Prices Really Reflect Fundamental Values? The Case of REITs|
with , : w10850
Real estate investment trust (REIT) stock prices deviate substantially from net asset values (NAV). Using REIT data since 1990, we find large positive excess returns to a strategy of buying stocks that trade at a discount to NAV, and shorting stocks trading at a premium to NAV. Estimated alphas from this strategy are between 0.9% and 1.8% per month, with little risk. Trading costs and short-sale constraints are not prohibitive and the results strengthen when we control for differences in liquidity or the extent of institutional ownership. We find that some variation in P/NAV makes sense, as premiums are positively related to recent and future NAV growth. However, there appears to be too much volatility in P/NAV, giving rise to potential profits from short-term mean reversion. The closed-en...
|October 2001||Short Sale Constraints and Stock Returns|
with : w8494
Stocks can be overpriced when short sale constraints bind. We study the costs of short selling equities, 1926-1933, using the publicly observable market for borrowing stock. Some stocks are sometimes expensive to short, and it appears that stocks enter the borrowing market when shorting demand is high. We find that stocks that are expensive to short or which enter the borrowing market have high valuations and low subsequent returns, consistent with the overpricing hypothesis. Size-adjusted returns are one to two percent lower per month for new entrants, and despite high costs it is profitable to short them.
Published: Jones, Charles M. and Owen A. Lamont. "Short-Sale Constraints and Stock Returns." Journal of Financial Economics 66, 2-3 (November-December 2002): 207-39. citation courtesy of
|January 1996||Public Information and the Persistence of Bond Market Volatility|
with , : w5446
We examine the reaction of daily bond prices to the release of government macroeconomic news. These news releases are of interest because they are released on periodic, preannounced dates and because they cause substantial bond market volatility. The news component of volatility is not positively autocorrelated on these dates, since the news is released at a specific moment in time. We find that (1) expected returns on the short end of the bond market are significantly higher on these announcement dates, and (2) the persistence pattern of daily volatility is quite different around these days.
Published: Journal of Financial Economics, Vol. 47, no. 3 (March 1998): 315-337.