Real estate performance measures

The commercial real estate market is increasingly dominated by institutional investors who require periodic performance measures of their investment portfolios. This presents a challenge to private real estate investments because individual properties are not bought and sold on a regular basis like stocks and bonds. Therefore, investors cannot measure gains or losses in their real estate portfolio based on actual transaction prices. They must use alternative measures of real estate performance.

This article will provide an overview on how performance measures for real estate investments are constructed and used by institutional investors. In particular, the article will focus on an index of real estate performance published by the National Council of Real Estate Investment Fiduciaries (NCREIF). The index includes over 1,500 properties whose market value in 1995 exceeds $20 billion dollars of commercial real estate.(l) This article will show how the index is used both to gauge the performance of real estate compared to other asset classes and to measure the performance of different property types and geographic areas so institutional investors can make diversification decisions. Understanding these indices helps real estate professionals understand the motivations of institutional investors when purchasing, valuing and selling real estate income property. These indices can also be used to track trends in real estate values and capitalization rates for different property types and geographic areas.

Construction Of Performance Measures

Indices of real estate performance measure change in the rate of return over time, e.g., annual or quarterly rates of return. As indicated earlier, this is easy for publicly traded investments like stocks and bonds because transaction prices are available to measure changes in value. In the case of real estate, it is necessary to rely on appraised values as a proxy for transaction prices. Institutional investors which hold property for pension funds are required to mark to market, i.e., report the market value of their holdings. Thus, appraisals are done on a periodic basis, usually once per year by an outside appraiser with quarterly updates by inside appraisers.

Performance measures essentially calculate an internal rate of return (IRR) for each time period, e.g., each year, based on the appraised value of the property at the beginning and end of the period as well as the net operating income (NOI) received during the period. This is shown in the following equation:

Equation 1

Total Return = NOI + (Sale price - Purchase price)/ Purchase Price (1)

Equation 1 calculates an internal rate of return for a single period of time. The return is called the total return because it includes return from net operating income and change in value (sale price -purchase price). This return assumes that the property could be sold each year at its appraised value.(2) In the case of stocks and bonds, dividend and interest income respectively would be used in place of NOI and actual transaction prices would be used for the sale price and purchase price.

Income And Capital Return

The total return already discussed can be broken down into an income return and a capital return. The income return is equal to

Income Return = NOI / Purchase Price (2)

The income return is analogous to an overall capitalization rate (cap rate) for a property. There may be differences, however, because the NOI used by the institutional investor may differ from the NOI used by an appraiser. For example, the institutional investor may not include any replacement allowance in the operating expenses which, compared to what the appraiser might estimate, may tend to understate the NOI. Alternatively, the institutional investor might include expenses for remodeling, leasing commissions, tenant improvements, etc. which, compared to what an appraiser would use as a stabilized expense, would overstate expenses.

Aside from these differences, trends in the income return can provide valuable insight into trends for capitalization rates. Thus, performance measures can provide a valuable complement to other sources of capitalization rates used by appraisers and counselors.

The capital return measures the effect of any appreciation or depreciation on the rate of return. It is calculated as follows: Capital Return = (Sale price - Purchase price) / Purchase Price (3)

The capital return assumes gain or loss which is recognized each period, e.g. the property is sold and repurchased. The capital return is also referred to as the appreciation return. It is possible, of course, that the appreciation be negative, i.e., depreciation in capital value.

Index

An index is often calculated from the return measures. The index indicates how much wealth the investor would have accumulated if he invested in the property and held it over time. It reflects cumulative rates of return over time. For example, if the total return was 5 percent in year 1 and 6 percent in year 2, n, an investment of $100 would increase to ($100 x 1.05 x 1.06) or $111.30 by the the end of the second year.

Example

Exhibit I is an example of the calculation on the income return, appreciation return and total return. The table also indicates how an index could then be calculated with 100 as the starting point.

Exhibit II illustrates the returns for the example in the graph. Note that the income return is relatively constant when capitalization rates do not vary significantly over time. Most of the variability is in the appreciation return. The average total return is 8.91 percent and the standard deviation of the total return is 3.51 percent.(3)

NCREIF Property Index

Institutional investors who are members of NCREIF report the information necessary to calculate income and capital returns on properties they hold. The formula used by NCREIF is a slight variation of the formula already discussed that includes the impact of any capital improvements and partial sales on the return. The differences are not material and do not affect the interpretation of the index.(4)

The NCREIF index is calculated by averaging the returns for all the individual properties reported to the organization by its members. The average can either be equal weighted or value weighted. Value weighting places more weight on the return from properties that have greater appraised values. This is the manner in which the traditional NCREIF Index is calculated. Exhibit III shows the total returns over time based on both equal and value weighted returns.(5)

Exhibit IV shows a breakdown if quarterly income and appreciation returns for the NCREIF Index. Note that income returns (cap rates) have been relatively steady over time and most of the volatility has been in the capital (appreciation) return. Because the returns are quarterly, they have to be multiplied by 4 for an estimate of the annual return.

Performance Of Different Property Types

NCREIF also reports performance measures for the different property types held by its members. Exhibit V shows a breakdown of quarterly returns by property type. Note that the relative performance of different property types varies over time. This is important because it indicates that institutional investors can receive diversification benefits by including different types of properties in a portfolio.

Performance Measures For Different Geographic Regions

NCREIF also breaks down performance measures by geographic regions. Exhibit VI shows the results. Again, note that performance differs over time for different regions which suggests diversification benefits for holding properties in different regions of the country.

Indices For Different Metropolitan AreasOffice Buildings

It is also possible to request performance measures from NCREIF for different metropolitan areas where a sufficient number of properties are held by its members. This provides a more refined analysis on the performance of different property types and trends by metropolitan area for different property types. More sophisticated institutional investors make diversification decisions by evaluating returns for different metropolitan areas. Exhibit VII shows the performance of office buildings in different metropolitan areas.(6)

Performance Of Private And Public Markets

With the tremendous growth of the REIT industry in recent years, public markets have become an alternative vehicle for institutional investors to hold real estate rather than make investments in the private market. As discussed, the NCREIF Index is based on appraised values of individual properties held directly by institutional investors. Performance measures also exist for REITs such as the NAREIT and Wilshire Indices. An important issue to consider is whether we can directly compare the performance of public and private market indices.

Two problems have been identified in making these comparisons. First, public market indices have much more volatility. Perhaps this is because they are based on actual transaction prices rather than appraised values. Another explanation is that trading by Wall Street introduces more volatility because REITs behave more like stocks, and thus are subject to the whims of the market. In either case, the differences in volatility must be considered. Second, transaction prices in public markets seem to lead appraised values in private markets. This may occur because public markets are better at anticipating and are quicker to reflect changes in market conditions. It may also be because the appraiser is less likely to put as much weight on the most recent transaction when estimating market value. We know that transaction prices for individual properties may not be indicative of market value. Furthermore, comparable sales are, by nature, historical despite attempts to make adjustments in market conditions.

To better understand the difference in performance of public and private markets, Exhibit VIII compares the NCREIF and NAREIT Indices. The differences are handled by 1. using different scales for the performance of the NCREIF versus the NAREIT Index to adjust for differences in volatility, and 2. lagging the return for the NAREIT Index by one year since it tends to lead the NCREIF Index. The performance over the period shown in the graph, 1978 to 1992, is quite informative. Note that the NAREIT Index rises relative to the NCREIF Index in 1991. This is the result of the 1981 tax act which increased the tax benefits to private market investors and produced a repricing of real estate. The returns in the exhibit are before tax. Thus, lower before tax returns were needed on private market investments relative to public market investments after 1981. These tax benefits were essentially eliminated in 1986, and we see the NAREIT Index decreasing as we move into 1986. This is followed by a significant decline in the NAREIT Index during the late 1980s as public markets appear to anticipate the real estate recession much faster than private markets. The NCREIF Index was slow to respond to declining real estate prices and the effect this had on performance. The opposite may have occurred in the early 1990s when public markets anticipated a recovery and prices rose for REITs faster than appraised values used for the NCREIF Index. The author expects these two markets will get closer in the future as institutional investors move funds between these markets and attempt to take advantage of any perceived arbitrage opportunities.

Conclusion

Indices of real estate performance can provide valuable insights into trends in real estate values and cap rates for different property types in different geographic areas. Although indices for private real estate markets are difficult to compare with their public market counterparts based on stocks (including REITs) and bonds, trends in the performance of REITs can provide insight to trends in private market performance. Real estate counselors and other industry professionals can use these performance measures to supplement more traditional sources of cap rate and return information for commercial real estate.

NOTES

1. Information about the index can be obtained by contacting NCREIF at Two Prudential Plaza, 180 N. Stetson Avenue, Suite 2515, Chicago, IL 60601. Telephone (312) 819-5890.

2. Although this seems quite reasonable, there have been several studies that suggest appraised values lag changes in transaction prices. See for example "On the Reliability of Commercial Ap praisals: An Analysis of Properties Sold from the RussellNCREIF Index (1978-1992)", R. Brian Webb, Real Estate Finance, Spring, 1994.

3. The standard deviation is a measure of the variability of the cash flows as defined in any standard statistics book. It is often used as a measure of the riskiness of an investment's returns.

4. For further discussion see Giliberto, S.F. "The Inside Story on

Rates of Return," Real Estate Finance, Spring 1994.

5. The equal weighted returns were calculated by the author with permission of NCREIF and first appeared in Fisher, Jeffrey D. "Alternative Measures of Real Estate Performance: Exploring the Russell-NCREIF Data Base," Real Estate Finance, Fall 1994.

6. This exhibit first appeared in Real Estate Portfolio Management and Strategy," a seminar prepared by Jeffrey D. Fisher and sponsored by the National Council of Real Estate Investment Fiduciaries (NCREIF) in Chicago, August 1994. The reader should note that metropolitan indices are based on relatively small sample sizes and may not always be representative of returns for the entire metropolitan area.

Jeffrey D. Fisher, Ph.D. is director of the Center for Real Estate Studies and associate professor of finance and real estate at the Indiana University School of Business. Currently he serves on the Board of Directors of the National Council of Real Estate Investment Fiduciaries (NCREIF).