Summary: |
Using both aggregate and firm level data we consider the divergence in the share prices of UK listed real estate firms relative to their Net Asset Value (NAV). Whilst it is often argued that REITs should trade close to their NAV the existing empirical evidence has often provided quite distinct results. The UK market is of interest in a number of respects. Firstly, due to the adoption of IFRS accounting standards in the UK, regular estimates of market value NAV are available. This is in contrast to markets such as the US where REITs place their real estate assets on their balance sheet at historic depreciated cost. Secondly, the UK only introduced REITs in January 2007. Prior to that date all listed real estate firms adopted a standard corporate structure, meaning that factors such as Contingent Capital Gains Tax played a major role in the premium-discount of the shares to NAV. As we have data available from 1990 this allows us to examine whether the dynamics altered following the introduction of the REIT regime. The data used is obtained from EPRA and runs from the early nineties until 2013. The methodological framework adopted in this study is based upon recent studies in the field of Purchasing Power Parity. There are a number of marked similarities between the study of reversion towards PPP with foreign exchange rates and the behaviour of public real estate firms relative to their NAV. The paper builds upon a relative small literature that has considered public real estate (e.g. Barkham & Ward, 1999; Liow, 2003; Liow & Li, 2006; Patel et al., 2008). We expand upon these existing studies in a number of respects. Firstly, we specifically utilize recent methodological advances with respect to the use of non-linear unit root tests (e.g. Sarno et al., 2004; Chortareas et al., 2008; Menkhoff & Rebitzky, 2008; Wu & Lee, 2009; Norman, 2010). Secondly, we not only empirically consider sector wide measures but also test for deviations across individual firms in a panel setting, as a number of recent PPP studies have similarly done so (e.g. Frankel & Rose, 1996; Coakley & Fuertes, 1997; Kaloncu & Kalyoncu, 2008; Chortareas & Kapetanios, 2009; Lau, 2009; Wu & Lee, 2009; Chang & Su, 2010; Koedijk et al., 2011). |