| Background
Until approximately 25 years ago, investment-grade real estate was bought and sold on the basis of a property's capitalization rate (income divided by value). Although the internal rate of return--IRR--(the total return on an investment, including both annual cash flow and residual proceeds at the end of the holding period) could theoretically be calculated, the IRR was not used as an analytic tool when making the initial investment decision because of the perceived vast number of undeterminable variables, as noted in the following excerpt from an article written in January 1974 by the late L.W. Ellwood, noted real estate valuation expert:
"The most significant factors for the determination of yield on the equity investment are usually the duration of the holding period and the net reversion to equity. Obviously, these are imponderables at the time of appraisal. They will be the result of management decisions, which cannot be made until specific opportunities to sell occur at various specific times in the future. Because the number of combinations is infinite at the time of appraisal, it follows that the appraiser has no reliable way to predict them."
Despite the fact that Elwood's assessment generally remains true today, the IRR has gained wider acceptance over the past few decades, often supplanting capitalization rates as the most revealing indicator of real estate investment performance. When did this happen and why? More importantly, will IRRs continue to be the Rosetta Stone of real estate returns?
An Industry Milestone
The first major office building to be sold on the basis of an IRR analysis was the Pan Am (now known as the MetLife) Building at 200 Park Avenue situated above Grand Central Terminal, arguably the 100 percent location in Manhattan. An agreement in principal was reached in July 1980 between the seller, Pan Am World Airways, and the buyer, the Metropolitan Life Insurance Company. Most market pundits were initially shocked at the sale price of roughly $400 million, a then record of $177 per square foot, and were even more amazed considering that the architecturally superior Seagram Building a few blocks north at 375 Park Avenue had just sold in February of that year for a record $135 per square foot. (Incidentally, a partial interest in 375 Park Avenue recently sold for $179 million, or the equivalent of about $660 per square foot.)
Estimated market rents per square foot within the Pan Am Building in 1980
| Lobby |
Mezzanine |
Fls. 3-9 |
Fls. 10-19 |
Fls. 20-29 |
Fls. 30-39 |
Fls. 40-49 |
Fls. 50-56 |
| $15-40 |
$15-25 |
$16 |
$25 |
$27 |
$29 |
$30 |
$35 |
Even more astounding were the measly initial year returns for the Pan Am property: a capitalization rate (net operating income divided by purchase price or value) of 3.4 percent, and a cash-on-cash return (cash flow after debt service divided by equity investment) of 2.4 percent. In contrast, the initial year return on the Seagram Building transaction, acquired by Teachers Insurance and Annuity Association from Joseph E. Seagram & Sons Inc., in an all-cash transaction, was 7.1 percent.
So, what did Landauer Associates, Inc., the seller's agent for the Pan Am Building, recognize that certain others didn't? For one thing, that the IRR was a better measure of value than the capitalization rate; and for another, that the building was well positioned to capitalize on anticipated future increases in Midtown Manhattan's office rents.
This was a period of skyrocketing rents in Manhattan. According to monthly market reports prepared by a prominent local brokerage firm, average Midtown Manhattan rents soared by 178 percent from $14.37 per square foot in December 1978 to $40.00 per square foot in December 1981. And Downtown, average rents increased by 139 percent during the same period, from $11.64 per square foot to $27.86 per square foot. Moreover, more than half of the square footage at the Pan Am Building, or about 1,150,000 square feet, was set to expire in 1983 and 1984, enabling the purchaser of the property to substantially increase rents, and therefore the return on their investment.
Technology also assisted in the landmark Pan Am Building transaction. Landauer was able to convince at least one member of the universe of potential purchasers that property values were going to increase by use of a lease-by-lease computer program that enabled it to forecast the impact of changing market conditions on the building's cash flow. Viewed this way, the expected internal rate of return on MetLife's acquisition was roughly 10 percent assuming a ten-year holding period, and 12 percent based on a 15-year hold. At the same time, Landauer estimated the IRR on the Seagram Building sale to be 12.7 percent, very much in line with that of the Pan Am Building, despite the wide disparity in first year returns.
So, the confluence of inflation, anticipated changes in property income levels, and the availability of affordable computers in the early 1980s gave birth to the use of IRRs in real estate investment analysis. Unfortunately, over the ensuing years, the application of the discounted cash flow technique (DCF), of which IRRs are a component, sometimes resulted in unrealistic cash flow projections, with forecasted increases in market rents continually outpacing increases in operating expenses, or as one prominent investment banker described it, trees growing to the sky.
Today, many of the problems with the application of a DCF analysis and the uncertainty of forecasting IRRs remain. For one, there is only one certainty associated with even the most carefully prepared cash flow forecast--and that is that it's wrong! Simply put, neither market rents nor most operating expenses will increase by exactly 2.5 percent or 3.0 percent per year for the next ten years, as is typically forecast. Appraisers and investors justify use of this approach by rightfully conceding that that they cannot guarantee the peaks and valleys in their forecasts, and use the expected average rate of change over their forecast periods.
Similarly, the IRR on a particular investment cannot be known with certainty until the asset is sold. Candid real estate investors will concede that they simply don't know what their IRR will be at the time of acquisition. Their anticipated IRR can better be expressed as the minimum rate of return hoped for on their investment. Importantly, widely referenced investor surveys convey the same type of information, but keep in mind that they are minimum rate of return expectations.
Lest one think that investing in real estate is substantially more speculative than investing in equities, consider the thought process that a stock market investor undertakes before investing. He or she may begin with an examination of the dividend and the company's record of paying dividends (akin to a real estate investor's preparation of a cash flow forecast). Then, future earnings are projected and a multiple applied to derive the anticipated future price of the stock in question on the anticipated sale date; this is the same procedure that the real estate analyst follows, estimating future income and applying a residual capitalization rate to estimate future value. Still, neither the IRR on the real estate investment nor the stock purchase will be known until the asset is sold.
Yet, assembling comprehensive market analytics can go a long way to significantly improving one's chance for success. The reader will recall that the key to the successful sale and purchase of the Pan Am Building was Landauer's astute analysis of the Manhattan office market and the convincing case it was able to make that the market was poised for an upswing. In fact, research techniques have become more sophisticated (significantly exceeding those available to MetLife when it acquired the Pan Am Building), enabling institutional investors to better assess a project's value and chances of a successful acquisition or sale, even among a wide variety of property types and geographic locations, albeit often at a substantial cost.
In October, publicly traded real estate investment trust (REIT) Vornado Realty Trust announced an agreement to sell its 965,000-square-foot Class B office building in Manhattan known as Two Park Avenue. The proposed sale to the German-led investor group, SEB Immobiliem-Investment GMBH, was for a price of $292 million, or about $303 per square foot. Vornado acquired the building in December 1986 for $151.5 million or $157 per square foot. This works out to a very impressive average annual rate of appreciation of some 9.2 percent per year, as well as an apparent gain of more than $140 million for Vornado, and a very substantial IRR. (Vornado then used part of the proceeds to acquire New Jersey's 1-million-square-foot Bergen Mall from Simon Property Group for $145 million in a Section 1031 tax-free like-kind exchange. Major tenants at the mall, which is expected to undergo a makeover and repositioning, include Saks Fifth Avenue, Macy's, Marshall's, and Value City.)
According to Reis's third quarter 2003 report, generally declining vacancy rates as well as gradually increasing rents are anticipated in Midtown Manhattan--the old adage about never going broke by taking profits appears to be at work. Furthermore, our analysis of the recent offering materials for the Two Park Avenue property indicates an anticipated IRR for the buyer of between 10 and 11 percent.
Another case in point is Biltmore Fashion Park in Phoenix, Arizona, a 567,000-square-foot regional shopping center with an upscale tenant roster and anchors that include Saks Fifth Avenue and Macy's. In late 1994, the center was sold for $115 million to Taubman Centers. An analysis of the Phoenix Retail Market prepared by Reis indicates that vacancy rates for several categories of retail properties were generally declining in 1994, with market rents expected to increase, indicating that, at least by this measure, it was an opportune time for investors to consider acquiring retail assets there.
An agreement was recently reached to sell the property to the Macerich Co., with a closing scheduled for the end of 2003. The reported sale price is $158.5 million, an average annual increase of 3.6 percent. Although average annual cash flows during the holding period were not available, it is very likely that this translated into a double-digit IRR. Equally important, Reis's analysis indicates that retail vacancy rates for general merchandise-oriented assets are likely to increase in Phoenix over the short term, loosely corroborating Taubman's decision to sell now.
It is important to note that there are of course many other considerations associated with a decision to acquire or dispose of real estate. Biltmore Fashion Park, although an open-air mall, is arguably a trophy property (as of the end of August 2003, annual tenant sales were an impressive $479 per square foot), and Reis forecasts that despite generally increasing vacancy levels, market rents for open-air general merchandise-oriented centers are still expected to increase. The point is that, just as real estate investment decisions have evolved from analyzing a single year's income stream through use of capitalization rates to an IRR/DCF analysis, market analysis tools have evolved as well, and are generally more reliable than they were years ago.
In yet one last example, AvalonBay Communities, Inc., a residential developer and publicly traded REIT, recently announced the sale of a 272-apartment home community in Orange County, California, resulting in an unleveraged IRR of 16.6 percent. "Market" IRRs have become increasingly transparent to members of the real estate community.
In fact, even the textbook definition of the IRR has gotten more complex:
The annualized yield rate or rate of return on capital that is generated or capable of being generated within an investment or portfolio over a period of ownership. The IRR is the rate of discount that makes the net present value of the investment equal to zero. The IRR discounts all returns from the investment, including returns from its reversion, to equal the original capital outlay. This rate is similar to the equity yield rate. As a measure of investment performance, the IRR is the rate of discount that produces a profitability index of one and a net present value of zero. It may be used to measure profitability after income taxes, i.e., the after-tax equity yield rate.1
Outlook
So, as market analysis techniques have improved, and financial data such as IRRs have become more transparent, has the risk of investing in real estate been reduced? A resounding "maybe." While investors can certainly make more informed decisions and calculate theoretical IRRs to an almost infinite number of decimal places, investing is still a risk-versus-reward decision.
In a "back to the future" type way, capitalization rates remain a significant, and oft-quoted measure of real estate values. Although investors have undoubtedly become more IRR-driven since the Pan Am Building sale, implicit in their targeted IRR is a minimum first year (capitalization rate) requirement. An institutional real estate investment with a 10.0 percent IRR and 8.0 percent initial year capitalization rate is far more likely than one with the same IRR and a 3.0 or 4.0 percent initial year capitalization rate.
Of course, despite all of the analytical techniques in the world, there is still something to be said for going out and "kicking the bricks" of the asset in question.
1The Dictionary of Real Estate Appraisal, Fourth Edition, 2002, Appraisal Institute, Chicago, page 150.
Howie Gelbtuch is a principal with New York-based Greenwich Realty Advisors Incorporated, an internationally oriented real estate valuation and consulting firm. |
