This article develops a theoretical framework and formulates a unified risk metric that integrates both real estate price risk and uncertainty of time on market (TOM). We demonstrate that real estate sellers with different degrees of financial distress face not only different marketing period risks, but also receive different return distributions upon successful sales. The major findings of this article can be summarized as follows. First, we show that real estate return and risk, which account for both price and TOM risk, are investor specific, varying over investors with different financial circumstances and holding periods. Second, the traditional valuation of real estate return and risk, which is based solely on the return distribution of a successful sale without considering the uncertainty of TOM and the investor's financial circumstances, underestimates real estate risk and exaggerates real estate return. Third, our empirical applications in both residential and commercial real estate markets show that the Sharpe ratio estimated by the traditional approach is seriously overstated-to the largest extent for investors with high financial distress. In addition, we find that, given the typical 5- to 7-year holding period for real estate, the Sharpe ratios estimated by integrating both price and TOM risk are much in line with the performance of financial assets. These findings can help to explain the apparent "risk-premium puzzle" in real estate. Copyright 2008 American Real Estate and Urban Economics Association