Real estate share prices have been implicitly forecasting property declines that are yet to materialise – what’s going on?
The most noticeable impact of the EU referendum in June 2016 upon the UK real estate sector has been the sharp contrast in demand (and as a result valuations) between the listed sector and the underlying assets themselves.
The share prices of the leading UK REITs traded down 20–25% immediately on the announcement of the referendum result. While commercial property values (with the exception of Q3 2016) improved, and occupational demand for industrial and London office space remained robust, the sector languished. It still trades at discounts to underlying value (NAV) of 30% for office-based REITs and >50% for retail-based portfolios.
This contrast in fortunes cannot be completely explained away by the decline in the value of sterling (making UK property assets more attractive to overseas investors) and the timing difference between forward-looking (ex-ante) share prices and backward-looking (ex-post) property valuations.
Typically (as last witnessed in the global financial crisis of 2007–09) real estate share prices lead the change in movement of property values (both up and down) by six to nine months. Historically there have been longer lags of 12, even 18, months. This time round, however, share prices have been (implicitly) forecasting property value declines of 10–15% – which, for over two years, have failed to materialise.
Does this signal the end of the connection between the public and private markets, and the usefulness of REITs as an implicit forecasting tool for real-estate valuations in the UK? I believe that the relationship still exists, but there are reasons why the lead/lag relationship has been both longer and more divergent.
Why was it different this time?
There are five factors in particular that added noise to the pricing signals from the listed market and distorted the accuracy of future property value predictions. These are: