Characteristics of a Full-Disclosure, Transaction-Based Index of Commercial Real Estate
Downs, David H., Slade, Barrett A., Journal of Real Estate Portfolio Management
Executive Summary. This study addresses the characteristics of a transaction-based index of commercial real estate as reported in a full-disclosure market. Prior research on transaction-based indices has enumerated various shortcomings for commercial real estate markets. This study circumvents many of these problems by using a large data set of commercial property transactions obtained for the Phoenix, Arizona metropolitan statistical area. The empirical analysis demonstrates that investors stand to gain considerable insight by comparing fulldisclosure, transaction-based indices with voluntarydisclosure, appraisal-based indices. The results suggest that full-disclosure indices avoid some of the institutional biases associated with other benchmarks of commercial real estate performance. In addition, some public policy issues emerge on the role of state mandated disclosure rules.
The weaknesses of appraisal-based indices have led practitioners and academicians to pursue transaction-based indices. This pursuit however, has not been without its obstacles. Perhaps the most significant problem has been the lack of adequate data. As a case in point, Miles, Hartzell, Guilkey and Shears (1991) attempt to formulate a transaction-based index by obtaining transaction data of industrial properties in the Charlotte, North Carolina market. However, the authors report that "the basic elements needed to construct the index are ... not publicly available information."1 Consequently, the researchers attempted to obtain transaction data from proprietary sources such as brokers. This effort also met with limited success. Due to the difficulty of obtaining reliable data, Miles, et.al. question whether a reliable transaction-based index can be constructed in other major markets.2
In order to address this and related issues, our study employs a data set obtained from a fulldisclosure market to overcome some of the problems encountered with appraisal-based and proprietary transaction-based indices. Full-disclosure is a term coined by real estate appraisers to indicate a high level of public disclosure of transaction data. The laws pertaining to public reporting of real estate transaction data vary considerably by state.
In states that do not require public disclosure, real estate professionals rely upon informal networks to obtain data. The time and cost to obtain data can be significant and the reliability often remains questionable. However, for states that require full public disclosure, the situation is different. Arizona, for instance, requires disclosure of all transaction data. Both buyer and seller are required by law to complete an affidavit disclosing the purchase price, financing type, arm's length nature of the parties transacting, property type, exchange, legal description and any non-realty items included in the sale.
A study by the Appraisal Subcommittee of the Federal Financial Institutions Examination Council documents the level of public disclosure of transaction data for all states.3 Exhibit 1 illustrates disclosure practices for real estate transactions by state.
The objective of this study was to determine whether the reliability of commercial appraisals resulting from the lack of public disclosure contributed to the banking crisis of the late 1980s and early 1990s. The study provides evidence that the lack of public disclosure did contribute to the unreliability of commercial appraisals. Furthermore, in a separate study conducted by the American Institute of Real Estate Appraisers (Appraisal Institute), only three states (Arizona, Colorado and Massachusetts) and the District of Columbia provided sufficient public disclosure data to complete an appraisal assignment as envisioned by FIRREA.4
The availability of public data in Arizona has allowed us to compile a large data set of office property transactions for the Phoenix metropolitan area. The data set consists of 935 transactions from the fourth quarter of 1987 through the third quarter of 1996. …