They Would If They Could: Assessing the Bindingness of the Property Holding Constraints for REITs

This study examines the bindingness of the property holding constraints which Real Estate Investment Trusts (REITs) face on their portfolios (the dealer rule), and illustrates how these constraints hinder REITs from exploiting opportunities to time the property market. I first simulate a set of filter‐based market timing strategies, which outperform a buy‐and‐hold strategy out of sample, and show that imposing a four‐year (or even the newer two‐year) holding constraint significantly reduces the excess returns the strategies generate. I then analyze actual holding periods of properties in REIT portfolios and find that there seems to exist a large degree of demand for short property holding periods and that the trades generated by the filter strategy generally resemble actual REIT trading activity, validating the relevance of the simulation results. A direct test for the constraint reveals that REITs' propensity to hold a property beyond the minimum period increases, the higher the profit from the transaction, consistent with the asymmetric nature in which the rule is enforced. By contrast, this effect is insignificant for Umbrella‐Partnership REITs (UPREITs), which are not as affected by the constraint. I further show that UPREITs overall achieve significantly better ex‐post market timing performance than non‐UPREITs. I thus find that overall REITs are limited by the dealer rule.