The TBI was first created at a time when there were no other sources of indices based on property transactions. The methodology adjusts the appraised value of all the properties in the NPI based on the ratio of the sale price to appraised value of just the sold properties. This has become increasingly problematic as the number of properties in the NPI has increased relative to the number of transactions. For example, in the 2nd quarter of 2020 there are only 30 sold properties at $1.7 billion compared to the NPI with 8,652 properties and a market value of $695.7 billion. Thus, the TBI would be using the sale to appraised value ratio on only a handful of properties to infer what all the properties in the NPI would have sold for compared to their appraised value. Furthermore, the motivations for sale of those 30 properties may not be typical, especially in the current COVID-19 environment. Therefore, the NCREIF Board has decided it is best to discontinue the TBI as it is currently produced.
The NPI is a value weighted index composed of a capital appreciation return and an income return. It is calculated using appraised values. The NTBI is an equal-weighted transaction and appraisal index. Neither one is right or wrong. They are two different methods of creating indices to track the real estate market. A value weighted index treats the NPI as a population or portfolio of all properties held by members whereas an equal weighted index views the properties in the NPI as a sample from a broader population of commercial real estate.The NPI includes hotels. There is no NTBI for the hotel property type, given the smaller population of properties and transactions in the NCREIF database. This may change in the future, but the current index series only includes apartment, industrial, office and retail property type indexes.
We use the equal-weighted version in the NTBI because we view the NTBI as a statistical estimation, based on a sample of properties (the ones that sold) inferring from that sample the change in value in the market represented by the entire NCREIF population of properties. We view each sold property as an equally-valid representative of that population. Hence, it makes sense for us to use the equal-weighted version of the NPI for comparison purposes. Also, the equal-weighted NPI tends to have less volatility (in effect it is more diversified, less dominated by a few large properties). Over time there should be no material difference in the trends, but there may be quarterly fluctuations that cause divergence depending on which properties transact.
The NTBI is not intended to be a benchmark. Only a small subset of properties in the NPI sells each quarter and thus do not represent the larger commercial real estate universe. Therefore, using the NTBI as a benchmarking index would not be appropriate. Furthermore, managers typically calculate the returns on their own funds using appraised values so they should compare these returns with an appraisal based index.