One of the most vexing issues for analysts and managers of property companies across Europe has been the existence and persistence of deviations of Net Asset Values of property companies from their market capitalisation. The issue has clear links to similar discounts and premiums in closed-end funds. The closed end fund puzzle is regarded as an important unsolved problem in financial economics undermining theories of market efficiency and the Law of One Price. Consequently, it has generated a huge body of research. Although it can be tempting to focus on the particular inefficiencies of real estate markets in attempting to explain deviations from NAV, the closed end fund discount puzzle indicates that divergences between underlying asset values and market capitalisation are not a ëpure' real estate phenomenon. When examining potential explanations, two recurring factors stand out in the closed end fund literature as often undermining the economic rationale for a discount ñ the existence of premiums and crosssectional and periodic fluctuations in the level of discount/premium. These need to be borne in mind when considering potential explanations for real estate markets. There are two approaches to investigating the discount to net asset value in closed-end funds: the ërational' approach and the ënoise trader' or ësentiment' approach. The ërational' approach hypothesizes the discount to net asset value as being the result of company specific factors relating to such factors as management quality, tax liability and the type of stocks held by the fund. Despite the intuitive appeal of the ërational' approach to closed-end fund discounts the studies have not successfully explained the variance in closed-end fund discounts or why the discount to net asset value in closed-end funds varies so much over time. The variation over time in the average sector discount is not only a feature of closed-end funds but also property companies. This paper analyses changes in the deviations from NAV for UK property companies between 1999 and 2003. The paper addresses a number of questions. i To what extent is there persistence in deviations from NAV between UK public real estate markets? ii To what extent do ërational' factors explain discount and premium in publicly traded real estate companies? iii Are trading volume and historic performance a significant explanatory variable of deviation from NAV? iv What role do market expectations play in explaining deviations from NAV?