Illiquidity and Pricing Biases in the Real Estate Market

This article addresses the micro‐analytic foundations of illiquidity and price dynamics in the real estate market by integrating modern portfolio theory with models describing the real estate transaction process. Based on the notion that real estate is a heterogeneous good that is traded in decentralized markets and that transactions in these markets are often characterized by costly searches, we argue that the most important aspects defining real estate illiquidity in both residential and commercial markets are the time required for sale and the uncertainty of the marketing period. These aspects provide two sources of bias in the commonly adopted methods of real estate valuation, which are based solely on the prices of sold properties and implicitly assume immediate execution. We demonstrate that estimated returns must be biased upward and risks downward. These biases can be significant, especially when the marketing period is highly uncertain relative to the holding period. We also find that real estate risk is closely related to investors' time horizons, specifically that real estate risk decreases when the holding period increases. These results are consistent with the conventional wisdom that real estate is more favorable to long‐term investors than to short‐term investors. They also provide a theoretical foundation for the recent econometric literature, which finds evidence of smoothing of real estate returns. Our findings help explain the apparent risk‐premium puzzle in real estate—that is, that ex post returns appear too high, given their apparent low volatility—and can lead to the formal derivation of adjustments that can define real estate's proper role in the mixed‐asset portfolio.