M. Nimalendran [*]
Despite their widespread use as benchmarks of U.S. commercial real estate returns, indexes produced by the National Council of Real Estate Investment Fiduciaries (NCREIF) are subject to measurement problems that severely impair their ability to capture the true risk--return characteristics--especially volatility--of privately held commercial real estate. We utilize latent-variable statistical methods to estimate an alternative index of privately held (unsecuritized) commercial real estate returns. Latent-variable methods have been extensively applied in the behavioral sciences and, more recently, in finance and economics. Unlike factor analysis or other unconditional statistical approaches, latent variable models allow us to extract interpretable common information about unobserved private real estate returns using the information contained in various competing measures of returns that are measured with error. We find that our latent-variable real estate return series is approximately twice as volatile as the aggregate NCREIF total return index, but less than half as volatile as the NAREIT equity index. Overall, our results strongly support the use of latent-variable statistical models in the construction of return series for commercial real estate.
The return performance of both publicly traded (securitized) and privately held (unsecuritized) commercial real estate, and their proper roles in mixed-asset portfolios, have received considerable attention over the past decade. Critical inputs into the evaluation of return performance and the specification of input parameters in mean--variance asset allocation decisions include historical means and variances, as well as covariances among the returns from the various asset classes potentially includable in the mixed-asset portfolio. For publicly traded stocks, including real estate investment trusts (REITs), historical return measurement poses few problems, as an abundance of performance data are available for both individual assets and portfolios. However, less than 5% of investable U.S. commercial real estate is publicly traded. And, unfortunately, transaction-based return information on privately held commercial properties is virtually non-existent, due to the relative infrequency of sales of the same prop erty. In the absence of transaction data, the private market uses periodic property appraisals to assess changes in values. This process creates smoothing and inertia in unsecuritized valuation series, a problem that has been extensively documented and discussed in the professional and academic literature (see, for instance, Giliberto 1988 and Geltner 1991).
In this paper, we use latent variable statistical methods to estimate an alternative index of returns on privately held commercial real estate. Latent-variable methods have been extensively applied in the behavioral sciences, where they have allowed researchers to analyze such abstract concepts as intelligence and motivation from proxies that are measured with error (see, for example, Joreskog and Sorbom 1979). These same methods also have recently been applied in finance and economics to measure unobservable macroeconomic risk factors, the determinants of corporate capital-structure choice, and the effects of earnings surprises on stock prices.  Unlike factor analysis or other unconditional statistical approaches, latent-variable models allow us to extract interpretable common information about unobserved private real estate returns using the information contained in various competing measures of returns that are measured with error.
The results from applying latent-variable techniques to estimate a return series for unsecuritized commercial real estate are encouraging. Using quarterly data from 1978-1997, we estimate a latent real estate return series that is about twice as volatile as the aggregate NCREIF total return index, but less than half as …