By Brad Case, Ph.D., CFA, CAIA
This is the second in a series of articles focusing on the strengths of different indices that are published regularly and may be appropriate for benchmarking, risk assessment, and other real estate investment purposes. (The first article focused on two similar index families, the Moody’s/RCA Commercial Property Price Index (CPPI) and the CoStar Commercial Repeat-Sales Index (CCRSI), both of which measure monthly capital appreciation at the property level.) In this article I will focus on the NCREIF Property Index (NPI) published by the National Council of Real Estate Investment Fiduciaries.
The NPI measures average capital appreciation, average gross income, and average gross total return at the property level. It is published quarterly based on reports submitted by NCREIF data contributing members for more than 7,000 properties worth nearly $400 billion owned at least in part by tax-exempt institutional investors such as pension funds. It is available for the aggregate U.S. commercial property market as well as for several important market segments.
What It Measures:
- The NPI measures returns at the property level, meaning that it does not measure the effects of leverage, which increases returns (provided that returns are greater than the cost of debt) and also increases volatility and other forms of risk. For properties purchased with debt, or held by investment managers who use debt, the NPI measures returns on the asset rather than on the equity invested. The relationship between returns on the asset and returns on the equity invested is: %RoA = [%RoE + %WACD * %L/(1-%L)] / [1 + %L/(1-%L)] where %WACD is the weighted average cost of debt and %L is leverage measured as debt divided by asset value.
- The NPI measures returns for operating properties, defined as existing properties that are not undergoing redevelopment, plus newly developed and redeveloped properties that have achieved occupancy of at least 60%.
- Along with total return, the NPI provides measures of both capital appreciation and income.
o Capital appreciation is NET of capital expenditures on the property: that is, it is equal to the change in the value of the property minus capital expenditures. This is normal in real estate, but users need to be careful not to confuse capital appreciation with property appreciation. NCREIF also produces a measure of price change (appreciation in property value), but does not publish it with the NPI.
o Income is GROSS of capital expenditures. Again, users need to be careful not to confuse income as defined in the NPI with income from stocks, bonds, and other investments, which is likely to be NET of capital expenditures. NCREIF also produces a measure of cash flow (income net of capital expenditures), but does not publish it with the NPI.
Reporting Schedule and Access: The NPI is published roughly four weeks following the end of each quarter. It is available free to NCREIF data contributing members, while others who cannot qualify as data contributing members can purchase access to the data (http://ncreif.org/public_files/NCREIF_Data_and_Products_Guide.pdf).
Geography: The NPI is available for a rich set of geographic areas, with data starting in 1978q1 for the nationwide aggregate, four regions (East Midwest, South, and West), and eight divisions (two in each region); some states have indices starting as early as 1978q1, while others do not yet have enough data for their own indices:
- Mideast: North Carolina (78q1), Maryland (78q3), South Caroline (78q3), Virginia (82q2), District of Columbia (86q2), Kentucky (88q2), Delaware (13q1), and West Virginia
- Northeast: New York (78q1), Massachusetts (78q4), Pennsylvania (79q1), New Jersey (80q1), Connecticut (80q3), New Hampshire (00q2), Rhode Island (04q4), Maine, and Vermont
- East North Central: Illinois (78q1), Ohio (78q1), Wisconsin (78q3), Indiana (81q3), and Michigan (82q2)
- West North Central: Missouri (78q1), Minnesota (78q2), Kansas (80q1), Nebraska (08q1), Iowa (12q1), North Dakota, and South Dakota
- Southeast: Florida (78q1), Georgia (78q1), Tennessee (78q1), Alabama (89q4), and Mississippi (06q2)
- Southwest: Texas (78q1), Louisiana (04q3), Oklahoma (05q3), and Arkansas (11q4)
- Mountain West: Arizona (78q1), Colorado (78q1), New Mexico (79q4), Utah (84q2), Nevada (88q1), Idaho, Montana, and Wyoming
- Pacific: California (78q1), Washington (79q1), Oregon (81q1), Hawaii (98q2), and Alaska (08q2)
In addition, the NPI provides indices for 104 metropolitan areas, with data going back all the way to 1978q1 for Atlanta, Chicago, Dallas, Denver, Houston, Jacksonville, Los Angeles, Memphis, Oakland, Phoenix, St. Louis, San Jose, and Santa Ana.
Property Types: In addition to the all-property indices the NPI is published for five property types and 13 sub-types:
- Apartment (78q1): Garden (88q2), Highrise (90q4), and Lowrise (95q1)
- Industrial (78q1): R&D (78q1), Warehouse (78q1), Flex (87q1), Other (98q4), Office Showroom (05q2), and Manufacturing (06q2)
- Office (78q1): CBD (78q1) and Suburban (78q1)
- Retail (78q1): Community Center (78q1), Neighborhood Center (78q1), Regional Mall (83q1), Super-Regional Mall (83q1), Power Center (94q4), Single-Tenant (95q2) , Fashion/Specialty Center (97q2), and Theme/Festival Center (12q4)
- Hotel (82q3)
Geography/Type Combinations: There are a huge number of geography/type combinations available through the NPI, with several of them going back to 1978q1.
Weighting: The NPI is value-weighted.
Advantages: The great advantages of the NPI are the length of the available historical period and the number of geographic areas, property types, and combinations available. The other advantage of the NPI, for institutional investors such as pension funds, endowments, and foundations—is that it includes specifically properties owned (or co-owned) by other institutional investors, which tend to be concentrated in the higher-quality and more-expensive segments of the market.
Disadvantages: The great disadvantage of the NPI is the fact that it is based on appraisals rather than transaction prices.
- Because it takes a long time to complete an appraisal, the NPI suffers from appraisal lag. This is especially true because appraisals are based on information from completed transactions of comparable properties, which is already outdated because the transactions themselves take so long to complete (illiquidity lag).
- Because “independent” appraisals are expensive, typically they’re done infrequently—once a year or even once every three years—with internal updating between external appraisals. The use of stale data makes the NPI suffer from non-appraisal lag.
- The combined effects of illiquidity lag, appraisal lag, and non-appraisal lag mean that returns measured by the NPI tend to lag behind actual returns by about four quarters on average. This lag makes it appear that property returns (as measured by the NPI) have a very low correlation with returns on other assets, when that is not really true.
- Appraisals tend to be radically smoothed relative to actual changes in property values, owing at least in part to the behavioral bias known as “anchoring”—the fact that appraisers tend to put too much weight on the known values of previous appraisals and not enough weight on the unknown current value of the property.
- The combination of appraisal smoothing and illiquidity smoothing makes it appear that the volatility of property returns (as measured by the NPI) is very low, when that is not really true.
- Because property values are updated infrequently, the NPI fails to measure property price movements or returns for any period less than a quarter.
- Because the process for collecting and reporting property values and income is time-consuming, the NPI is significantly delayed relative to actual movements in property values and returns, with publication coming about four weeks after each quarter has ended.
Notwithstanding these disadvantages, the NPI can be very valuable for benchmarking, risk assessment, and other investment purposes, particularly when it is important to measure long-term average returns but not important to measure volatility (quarter-to-quarter variation) or correlations with other assets.
Brad Case is senior vice president, research & industry information for the National Association of Real Estate Investment Trusts (NAREIT). Dr. Case has researched residential and commercial real estate markets, domestically and globally, for more than 25 years. His research encompasses investment return characteristics including returns, volatilities, and correlations with other assets; measuring appreciation in property values; inflation protection; use of DCC-GARCH and Markov regime switching models to measure and predict investment characteristics; the length of the real estate market cycle; and the role of the investment horizon. He holds patents as the co-inventor of the FTSE NAREIT PureProperty(r) index methodology and the backward-forward trading contract. Dr. Case earned his Ph.D. in Economics at Yale University, where he worked with Robert Shiller and William Goetzmann, and holds the Certified Alternative Investment Analyst (CAIA) designation.