Search for a Common Factor in Public and Private Real Estate Returns

Abstract

We introduce a methodology to estimate common real estate returns and cycles across public and private real estate markets. We first place REIT indices and direct real estate-NCREIF appraisal-based and transaction-based indices (NPI and NTBI)-on a comparable basis by adjusting for leverage and sector. We extract a common real estate factor, which is allowed to be persistent, from all these markets. Individual real estate indices load on this common factor and they also are driven by persistent, idiosyncratic shocks. The common real estate factor is procyclical and has low correlations with standard systematic factors. Short-run idiosyncratic deviations from the common real estate factor load on several capital market factors for REITs and on liquidity factors for direct real estate.

International Real Estate Review

“…Yunus et al (2012) examine the international markets and discover that in addition to the U.S. and U.K. markets, Australia and Netherlands also exhibit long-run relationships between the public and private real estate markets. Ang et al (2013) study the long-run commonality between the two markets and find that the common real estate factor is highly autocorrelated, thus reflecting the cyclical nature of real estate. Moreover, in the long-run, both public and private real estate vehicles exhibit similar characteristics.…”

“…On the other hand, public securitized real estate markets, such as real estate investment trusts (REITs), offer investors a low-cost alternative to invest into the real estate market, yet the prices of real estate securities may be affected by noise traders or investor sentiment and may deviate from the fundamentals (e.g., Clayton and MacKinnon 2001a;Gentry et al 2004;Lee et al 2013). Although studies have shown that a short-run correlation between public and private real estate markets is relatively low (e.g., Sagalyn 1990;Mueller and Mueller 2003;Brounen and Eichholtz 2003) and that the public real estate market more resembles the general stock market (e.g., Goetzmann and Ibbotson 1990;Wang et al 1995;Ling and Naranjo 1999), the long-run co-integration between these two markets is particularly strong (e.g., Geltner and Kluger 1998;Liow 2003;Liow and Li 2006;Ang et al 2013). With regard to the price determination in the public and private real estate markets, earlier studies (e.g., Gyourko and Keim 1992;Barkham and Geltner 1995;Eichholtz and Hartzell 1996;Chau et al 2001;Geltner et al 2003) that have examined the long-run lead-lag relation between these two markets have predominantly agreed that the public market leads the private market, thus implying that new information is first incorporated into securitized real estate prices before it is reflected in the private market.…”