Using Economic Value Added (EVA) as a Barometer of Hotel Investment Performance
MetadataShow full item record
Clayton, Matthew; Liu, Crocker
In this report, we show how the popular and well known economic value added (EVA) technique can be used as a barometer of investment performance for hotels and other property types. Economic value added represents the return on a project in excess of its financing cost. The EVA metric complements traditional measures which analyze the spread between the cap rate and either the ten-year U.S. Treasury bond rate (riskless debt) or the BAA bond yield (risky debt). Although both RevPAR and EVA account for revenues, the primary advantage of applying EVA analysis (versus cap rate spread) is that EVA also considers the cost of equity financing (in addition to risky debt financing), which cap rate spread and RevPAR do not capture. Unlike cap rates, EVA incorporates investors’ risk premium. To demonstrate this metric, we show that the EVA spread on hotels reached its peak in the first quarter of 2004, at the start of the boom for hotels, while it hovered near zero during 2007, when hotel values reached their apex. The EVA spread then turned negative in the second quarter of 2008 as hotel values had already started to decline, and it continues to remain negative in general, coinciding with continued economic and property sector weakness. A negative EVA spread suggests that investors are buying properties with the expectation of upside potential arising from higher cash flow growth rates typically achieved through repositioning or renovating the hotel. We find evidence to suggest that once the EVA spread is below 1 percent, which is equivalent to a year-over-year change in RevPAR of approximately 1 percent, real estate practitioners should start to check whether the canary in the coal mine is still chirping, or whether their deal has expired.
hotels; economic value added (EVA); investment performance; revenue per available room (RevPAR); risk
Required Publisher Statement: © Cornell University. This report may not be reproduced or distributed without the express permission of the publisher