Academic journal article The Journal of Real Estate Research

Heterogeneous Information and Appraisal Smoothing

Academic journal article The Journal of Real Estate Research

Heterogeneous Information and Appraisal Smoothing

Article excerpt

Abstract This study examines the heterogeneous appraiser behavior and its implication on traditional appraisal smoothing theory. The findings demonstrate that the partial adjustment model is consistent with the traditional appraisal smoothing argument only when all appraisers choose the same smoothing technique. However, if appraiser behavior is heterogeneous and exhibits cross-sectional variation due to the difference in their access to, and interpretation of information, the model actually leads to a mixed outcome: The variance of the transaction-based returns can be higher or lower than the variance of comparable-based return depending on the degree of such heterogeneity. Contrary to what the traditional appraisal smoothing theory would predict, appraisal-based indices may not suffer any ''smoothing'' bias. These findings suggest that the traditional appraisal smoothing theory, which fails to consider the heterogeneity of appraiser behaviors, exaggerates the effect of appraisal smoothing.

(ProQuest: ... denotes formulae omitted.)

One of the most extensively documented observations in the real estate literature is that appraisal-based return series (as exemplified by the NCREIF Property Indices) exhibit comparable returns to common stocks, but with much lower volatility. That is, the risk-adjusted returns of commercial real estate are significantly higher than those of common stocks. This is often referred to as the ''real estate risk premium puzzle'' (Lusht, 1988; Shilling, 2003) as it suggests that real estate rewards investors with additional premium for the same risk. The consistently repeated observation and the desire to solve the ''puzzle'' have prompted one of the leading theories in real estate: the appraisal smoothing theory.

As articulated by Geltner (1989, 1991) and many others, the foundation of appraisal smoothing theory is that the seemingly ''superior'' performance of real estate is essentially caused by the rational behavior that appraisers exhibit in their practices. Particularly, it is believed that, in forming an appraisal estimate on a given property, the appraiser tends to not only analyze the current market information, but also gives some consideration to the past market information of the same or similar properties. This is called partial adjustment behavior, as the current market information is not given 100% consideration. This behavior effectively implies that appraisers tend to weight-average the past and present information to form their appraisal opinion. An appraisal-based return series such as the NCREIF Property Index (NPI), therefore, is essentially some sort of ''moving average'' of the underlying asset performance. As such, it is expected to be ''smoothed'' and exhibit downward-biased volatility over time. Correcting such bias, it has been argued, will lead to increased variance and reduced riskadjusted returns for real estate to the levels comparable to that of common stocks, thus solving the ''real estate risk premium puzzle.''

Based on the assumed appraiser behavior, Geltner (1989, 1991) extended the work of Blundell and Ward (1987) to propose a formal model known as the partial adjustment model, in which the current appraisal return is expressed as a weighted average of the past true property returns (or true property appreciation rates). This model formalizes the smoothing argument and provides the normative foundation for the theory. Over the past twenty years, the partial adjustment model has gained wide popularity among academics, and is often taken as the theoretical launch pad for studies that want to quantify the smoothing effect or develop de-smoothing methodologies.

Despite repeated and often sophisticated elaborations by many studies from various angles, it is worth noting that the whole idea of appraisal smoothing theory is almost entirely motivated to explain one simple fact: the exceptionally low volatility in the only industry standard benchmark of commercial real estate at the time, the NPI. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.