It has been well documented that rents for multi-family properties have “spiked” sharply as a result of limited new product (supply) and continued growth (demand) in the past 2-3 years.  In recent projects, RCDH has confronted an unusual dilemma:  unit value results via the Income Approach are or can be radically different from unit values indicated by sales data.  This is especially true in downtown Washington or in other markets (e.g., Arlington; Tysons Corner; Montgomery County) where high-rise product exists.  How is “the gap” reconciled?  Leaving aside the issue of “cost” for discussion purposes, this question will be answered in the following paragraphs.

RCDH has conducted an extensive search for comparable hi-rise apartment complexes recently sold in Washington DC Metro and nationwide. In certain local cases, the application of the Income Approach indicates market values of +/-$300,000/DU.  In other more “suburban” cases, market values can equate to +-$200,000/DU. In both examples, all assumptions and data (rent levels, vacancies, expenses, etc.) are well supported by market evidence.  In fact, they (assumptions), in some cases, can be deemed conservative.  Yet no local sales have traded at unit values even close to these pricing levels.  The most recent sales are only in the +-$110-$150,000/DU range, with 1 sale at $175,000/DU. How is the investor or analyst to reconcile such a wide discrepancy?

Issues/Solutions

Table 1 illustrates a summary of the most recent “high end” apartment sales in DC metro.  Here, unit values range from +-$110-$150,000/DU, with one sale at $177,000/DU.

Insert Table 1.

Can these “comp sales” support unit value (via the Income Approach) well above those indicated by the sale data?  Answer:  Yes and No.

Step 1.

To the extent that the cap/yield rates provide consistent results, they can be and, in fact, are very comparable.  Should cap and yield rates be divergent, however, they are not as useful, as ordinary adjustments are not likely to yield defensible conclusions. In this case, net adjustments of over 50% are needed to arrive at Income Approach conclusions.  In the example at hand, the sale data abstracted cap and yield rates are at consistent levels.  Sales of quality high-rise product generally are in the 7-8.5% range, with most currently in the 7-8% range.  Thus, cap rate data can be applied to actual or projected cash flows on an individual property with a good level of confidence. The issue is that the resulting unit value via the Income Approach is now well over $200,000/DU is certain local cases.

Step 2.

Given this situation, RCDH conducted a national sale search recently constructed apartment projects.  The results of this study are shown in Table 2.

Insert Table 2

The number of recent sales of high-end apartment complexes built from 1995-2001 during the past 3 years, with a market value of more than $200,000/DU is about 20 in the entire US (as of 4Q/00).

Again, the data indicates a fairly consistent unit value conclusion given the wide range of locations (Boston to San Diego).  In addition, cap and yield rates are also surprisingly uniform.  From that perspective, again, the sale data is, in fact, comparable despite wide locational variances.

Interviews indicate that buyers will only pay such high unit values under the following circumstances:

1. Local supply constraint is well known, long-standing and a difficult process.
2. Rent growth is supported by local economic and effective household income growth.  Said differently, population growth alone is not enough. Tenants must have the ability to pay luxury rents (effective demand) on a reliable basis (good job security; broad market depth).  This limits pricing to markets with high income and education levels where market fluctuations do not materially alter the tenant base.
3. Since this pricing category is limited to the most recently delivered product, verifiable cost figures are a must.  Application of these parameters to older product is not indicated.
4. Clearly, as one approaches $300,000/DU, sale volume thins very quickly.

Given such a rapid shift in unit value ($110-$150,000/DU) over the past 12 –18 months to figures now over $200,000/DU, the uniformity of buyer criteria and rates of return is a strong indication of defensible market value conclusion via the Income Approach.  This is in spite of the absence of local sales at these much higher unit values.  Said differently, local investors are long-term holders who rarely “mark to market” (i.e., absence of sale data locally).

Step 3.

In the event the national sale database is not acceptable or pertinent (lender underwriting rules, etc), an analyst can employ the “condo conversion” methodology.

Basically, in this scenario, an investor acquires the property on a “wholesale” basis (as an apartment complex) and then re-sells individual units to tenants at a “retail” price. In fact, several of the “national” sales appear to have this as a “buyer motivation”.  Thus, it would appear that this analytical tool also is grounded in actual market activity.  In the event this analysis is used, market based assumptions relative to condo price, absorption (sellout), conversion costs, “insider discounts”, yield rates, carrying expenses, etc, must all be supported by market evidence.  In the data in Table 2, a fairly uniform ratio of “wholesale” to “retail” pricing is reported in the cases where this factor is known.  Further, with proper analysis, the resulting “wholesale” conclusion should indicate a reliable barometer of value.

Conclusion

Based on the preceding, Washington metro appears to lag the national market in terms of unit pricing for some aspects (luxury; Class A; high-rise; close-in) of the multi-family market.  It is, therefore, recommended that a complete analysis of all aspects of a particular investment consider these three steps as a minimum.  Otherwise, the underwriting of such properties may be incorrectly evaluated.

Dennis Duffy, MAI

Principal of RCDH & Co.

Published in DCBIA, July 2001.  Pipeline, Washington, DC: DCBIA.

Posted in Publications