Academic journal article The Journal of Real Estate Research

Investor Rationality: An Analysis of NCREIF Commercial Property Data

Academic journal article The Journal of Real Estate Research

Investor Rationality: An Analysis of NCREIF Commercial Property Data

Article excerpt


National Council of Real Estate Fiduciaries multiple listing service level cash flows and panels of capitalization rates for industrial, office and retail properties over the last two decades are examined in this study. Real NOI 5-year future growth is shown to be negatively related to deviations of current real NOI from trend. Given this trend reversion in real cash flows, investor rationality requires that income multipliers be low (capitalization rates be high) when real cash flows are above trend and visa versa. In the panel estimates, the opposite is seen to be the case. Whether this is due to questionable data or irrational behavior is uncertain.


Capitalization rates should be linked to expected real cash flow growth. The higher is expected real growth, the more investors should be willing to pay for a current dollar of cash flow and thus the lower should the capitalization rate be. To illustrate, when real rents are expected to rebound from a cycle bottom, investors should be willing to pay relatively more for a dollar of current cash flow-cap rates should be low-than when only normal growth is expected.

Real property cash flows per unit space have been shown to be mean reverting in the United States (Wheaton and Torto, 1994), as well as in many other countries.1 As a result, when real cash flows are highest relative to trend, investors should expect negative (or at least below average) real cash flow growth and cap rates should therefore be relatively high.2 However, based on panel estimation of NCREIF data, both Sivitanides, Southard, Torto and Wheaton (SSTW, 2001) and Chen, Hudson-Wilson and Nordby (CHN, 2004) report negative relationships between cap rates and the ratio of market to long run equilibrium real rents. SSTW conclude that U.S. investors have not built the "obvious" mean reversion of real rents into their forecasts of real rental growth and thus have overvalued property at rental cyclical peaks (used too low cap rates) and undervalued it at cyclical troughs.3

These results contrast markedly with those of Hendershott and MacGregor (2005), who find that investors in the office and retail markets in the United Kingdom have appropriately built mean reversion into U.K. cap rates. The contrasting results could be due to differences in modeling, in the quality of data, or in differences in U.K. and U.S. investor behavior. Regarding modeling, Hendershott and MacGregor found a strong link between property cap rates and the stock market (both the dividend price ratio and proxies for the expected growth rate in real dividends), a link not explored in the SSTW and CHN studies. As for data, two weaknesses in the U.S. data analyses are noteworthy. First, the NCREIF cap rate series are based, in part, on stale appraisals and, at certain times and places, on very few properties. second, the real net rental data employed in constructing expected rental growth proxies are based on market rent (new leases) data rather than actual property cash flows.

The present paper takes another look at the NCREIF data, building different data panels and using real NOI cash flows. Stale appraisals are excluded in the computation of cap rates, as are periods and/or MSAs from the panel estimation when there are too few properties to give plausible NOI growth rates. Both underlying NCREIF NOI data and market rents are used to compute proxies for expected real NOI growth. Unfortunately, attempts to establish links to equity market variables similar to those found for the U.K. market have been unsuccessful and thus are not reported.

section two describes the modeling framework and reviews the SSTW and CHN methodologies and empirical results. section three describes the NCREIF cap rate and real NOI data used. There are serious concerns with the quality of the data. The capitalization rates are extremely volatile, and numerous outlandishly large (in absolute value) quarterly real NOI growth rates are observed. …

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.