We analyze the determinants of expansions and contractions of shopping centers using a unique dataset of property level data for shopping centers in eleven metropolitan areas over the period from 1995 through 2005. We find that shopping centers with large operating costs are less likely to expand and are more likely to contract. Higher expected revenue per square foot increases the likelihood of expansion and decreases the likelihood of contraction. For small shopping centers the decision to change gross leasable area (GLA) is largely driven by potential revenue, while the decision to change the number of stores is largely a function of cost. We find some support for Grenadier’s theory that a larger number of competitors reduces the value of option to wait and increases the likelihood of both expansion and contraction. The market share of competitors reduces the likelihood of increasing the number of stores as suggested by the theory of strategic positioning. Our hypotheses best explain contraction decisions of large shopping centers and expansion decisions of small shopping centers. We find that both expansions and contractions of GLA are less likely for large shopping centers in MSAs with greater uncertainty about real estate prices, indicating that the option to delay has value. Moreover, small centers are significantly less sensitive to cost and revenue; since small centers are likely to have greater idiosyncratic risk than large; this provides indirect evidence for a significant delay option.