International Journal of Hospitality Management 28 (2009) 439–445

Contents lists available at ScienceDirect

International Journal of Hospitality Management

journal homepage: www.elsevier.com/locate/ijhosman

EVA, refined EVA, MVA, or traditional performance measures for the hospitality
industry?
Seoki Lee a,1, Woo Gon Kim b,*
a
School of Tourism and Hospitality Management, Temple University, 1700 N. Broad Street, Suite201-F, Philadelphia, PA 19122, United States
b
Dedman School of Hospitality, The College of Business, Florida State University, 288 Champions Way, UCB 4116, Tallahassee, FL 32306, United States

A R T I C L E I N F O A B S T R A C T

Keywords: Stewart (1991) proposed economic value added (EVA) as a true measurement of a firm’s performance
Economic value added (EVA) and an executive’s evaluation tool because EVA reflects only incremental values added to a firm after
Refined EVA considering cost of capital. Kim [Kim, W.G., 2006. EVA and traditional accounting measures: which
Market value added (MVA) metric is a better predictor of market value of hospitality companies? Journal of Hospitality & Tourism
Traditional accounting measures
Research 30(1), 34–39] examined EVA in the hospitality setting and concluded that EVA is not superior to
Hospitality industry
other available measurements for accounting. However, this study contributes several improvements to
Kim’s (2006) study and compares the incremental explanatory power of six firm performance measures
including EVA, refined EVA (REVA), market value added (MVA), and three traditional accounting
performance measures for market adjusted returns. According to the findings, REVA and MVA are,
apparently, valuable performance measures for evaluating hospitality firms.
ß 2009 Elsevier Ltd. All rights reserved.

1. Introduction interested in the topic and EVA studies somewhat diminished in

In the 1990s, the concept of economic value added (EVA)
became a topic of considerable interest resulting in wide, financial
economics literature research from various perspectives. Stewart
(1991) proposed EVA as a firm’s performance measurement and as
executives’ performance evaluation tool by arguing that EVA
represents a firm’s true performance because EVA reflects only
incremental values added to a firm after considering cost of capital.
Fortune published a cover story that discussed the benefits of EVA
and a long list of major companies that adopted EVA as an
evaluation tool (Tully, 1993). To verify such support for EVA, a
major part of the EVA literature attempted to examine the
superiority of EVA to traditional performance measurements (for
example, Brossy and Balkcom, 1994; Chen and Dodd, 1997; Garvey
and Milbourn, 2000; Lehn and Makhija, 1996) and the comparison
also extended to a modified EVA (Bacidore et al., 1997). However,
several studies found that EVA was not superior to other traditional
firm performance measurements (for example, Biddle et al., 1997;
Chen and Dodd, 2001; de Villiers, 1997; Kim, 2006; Weissenrieder,
1998; Zimmerman, 1997). Researchers, therefore, became less
* Corresponding author. Tel.: +1 850 644 8242; fax: +1 850 644 5565.
E-mail addresses: seokilee@temple.edu (S. Lee), wkim@cob.fsu.edu,
woogon@yahoo.com (W.G. Kim).
1
Tel.: +1 215 204 0543; fax: +1 215 204 8705.
the literature.
However, EVA has found itself continuously used and examined
in more variety of disciplines including not only finance and
accounting, but also strategic management, marketing, and human
resources (for example, Ezzamel and Burns, 2005; Fletcher and
Smith, 2004; Mir and Seboui, 2006; Zinkhan and Verbrugge, 2000;
Zinkin, 2006). Moreover, EVA research in the hospitality industry
has also received various attentions from industry practitioners as
well as academics. Kefgen and Mahoney (1996) showed the
application of EVA as a new performance measure for incentive pay
by illustrating Walt Disney’s case. Before implementing EVA-based
incentive plan, the hospitality consultants suggested hospitality
firms develop a system that is easy to understand and is widely
accepted by stakeholders in the company. Chow et al. (2003)
conducted a survey of hotel managers and the respondents
confirmed that adopting value-based performance measure such
as EVA is more effective than traditional financial performance
measures. They also provided useful guideline before implement-
ing EVA as a hotel performance measure. For the restaurant
industry, Aliouche and Schlentrich (2005) compared franchising
and non-franchising restaurant companies in terms of their firm
performance, estimated by EVA while Madanoglu et al. (2004) used
EVA in illustrating value destruction phenomenon for airline
companies in 1990s.
However, these studies did not explore the validity of EVA
concept in a comparative and empirical manner, especially with

0278-4319/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhm.2009.01.004
440 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445
other traditional firm performance measures until Kim (2006).
According to Kim (2006), EVA did not explain market capitalization
better than the other two traditional accounting measures: net
operating profit after tax (NOPAT) and free cash flow (FCF). In
general, this finding was somewhat consistent with previous EVA
literature and might have reduced hospitality researchers’ interest
in examining EVA further in that particular context.
However, this study raises two issues regarding the EVA concept
for the hospitality industry. The first issue is applying a refined EVA
(REVA) concept developed by Bacidore et al. (1997) to the hospitality
industry and the second issue is proposing improvements to Kim’s
(2006) research. Bacidore et al. (1997) developed REVA by arguing
that assessing the capital charge on the market value of the firm
(REVA) rather than on the economic book value of its assets (EVA).
They regressed abnormal returns on EVA and REVA and found that
REVA presented a stronger coefficient than EVA. Moreover, whereas
REVA showed a positive sign, EVA demonstrated a negative sign.
However, REVA, somehow, did not receive much attention in the
literature possibly because the previous great interest in EVA quickly
disappeared in the late 1990s, and REVA has never been applied to
the hospitality industry. Therefore, this study adopts and examines
REVA, particularly in comparison with EVA, and specifically to the
hospitality sector.
While Kim (2006) made a contribution to the hospitality
literature by introducing the EVA concept, several issues confound
the study. For example, as the study acknowledged, it encom-
passed only one sample group, the entire hospitality industry,
without further sub-sectors such as hotels, restaurants and
casinos; the sample period was relatively short (1995–2001),
and the data was not provided by Stern Stewart & Co., but
estimated by the researcher. Stern Steward & Co. developed EVA
and has been consistently producing the EVA data set. Given these
issues, this study attempts to make several improvements on Kim
(2006) by, for example, examining sub-sectors (i.e., hotels,
restaurants and casinos) of the hospitality industry and using
the data set from Stern Stewart & Co. for 1985–2004 (along with
others discussed further in the following sections).
The study purpose, therefore, is to introduce refined EVA to the
hospitality literature and compare it to EVA, market value added
(MVA) and other traditional accounting measures (i.e., cash flow
from operations [CFO], return on assets [ROA], and return on equity
[ROE]) on market adjusted returns from each of three hospitality
sectors (i.e., hotel, restaurant, and casino) and the total (all three
hospitality sectors). When considering the hospitality industry’s
specificity and the argument, ‘‘EVA is better suited for industries
such as food and beverages, which have lower volatility’’
(Milunovich and Tsuie, 1996, p. 112), this study has the potential
to contribute meaningfully to the hospitality industry’s financial
literature. Next is a review of relevant literature followed by data
and methodology sections. Findings and implications appear prior
to a discussion of limitations and suggested future research which
conclude the study.
2. Literature review
Stewart (1991) proposed EVA as the financial performance
measure by arguing that EVA presents a company’s true profit. A key
component of EVA is to consider cost of capital in estimating the
performance measurement; only when a company generates returns
(i.e., NOPAT in EVA estimation) exceeding cost of capital (including
both equity and debt), does a company’s value become enhanced
(Stewart, 1991, 1994). Ever since the introduction of EVA to the
literature, many studies investigated the measurement by often
comparing it to other traditionally used performance measurements,
and naturally two groups of researchers emerged in the EVA
literature: (1) proponents of EVA and (2) opponents to EVA.
In the early years of the EVA literature, the first group
(proponents of EVA) seemed to prevail in the literature. Several
articles from prestigious business trade magazines such as Fortune
and Forbes praised the use of EVA and provided a list of top
performers based on EVA data from Stern Stewart & Co. (Fisher,
1995; Lieber, 1996; Rutledge, 1993; Teitelbaum, 1997; Tully, 1993,
1994; Walbert, 1994). Through a series of trade magazine articles,
EVA began receiving more attention from academicians. Several
researchers presented findings regarding a relationship between
EVA and MVA, and this relationship indicates: EVA explains a
significant portion of MVA (Grant, 1996); EVA explains more
variations of MVA when compared to some traditional accounting
measures such as EPS, ROE, ROA and net income (Uyemura et al.,
1996), when compared to NOPAT (O’Byrne, 1996), and when
compared to EPS growth and free cash flow (Milunovich and Tsuie,
1996).
Another group of researchers investigated a relationship
between EVA and stock returns. While Lehn and Makhija (1996)
argued that EVA explains most of the variations in stock returns
among its peer performance indicators such as MVA, ROE, ROA,
and ROS, Chen and Dodd (1997) found that although EVA
explains stock returns better than other accounting measures,
the accounting measures such as earnings per share, return on
assets and return on equity still provide significantly valuable
information.
The second group, those voicing opposition in EVA literature,
used a similar methodology, but found opposite results. Biddle
et al. (1997) examined an explanatory power of EVA, accrual
earnings, residual income and cash flow from operations on market
adjusted returns, and revealed that accrual earnings showed a
higher explanatory power compared to other performance
measures including EVA. Chen and Dodd (2001) also compared
an explanatory power of EVA for stock returns relationship to
operating income (OI) and residual income (RI), and found that OI
explained the most and EVA the least. Clinton and Chen (1998) also
examined a relationship between performance measures including
EVA, residual cash flow (RCF) and ROI-based measures, and found
that RCF was the best predictor. In a context that examines a
relationship between performance measures including EVA and
MVA, Kramer and Pushner (1997) found that NOPAT explains more
of the variations of MVA than EVA.
Contrasting these two groups, Bacidore et al. (1997) proposed
the use of a refined EVA (REVA) in which market value of the firm’s
assets should be used in EVA calculations, rather than book value of
assets because the capital charge for the firm is based on its
market-based weighted average cost of capital (WACC). They
compared REVA to EVA in terms of explanatory power for
abnormal returns and found REVA to have better explanatory
power than EVA. Following Bacidore et al. (1997), this study
includes REVA in the comparison analysis with EVA and other
accounting measures.
After many EVA studies in the mainstream literature, Kim
(2006) compared EVA to other accounting performance measures
(NOPAT and free cash flow [FCF]) in the hospitality setting. Kim
(2006) examined the relationships among the three firm perfor-
mance measures (EVA, NOPAT, and FCF) and market capitalization
(MKTCAP) to see which performance measure explains the most
about MKTCAP. From a multiple regression analysis, although EVA
revealed a positive and statistically significant coefficient, the
significance was less than NOPAT and FCF, and therefore, does not
support EVA’s superiority.
However, as briefly mentioned in Section 1, this study’s focus is
to reconcile and improve the issues others have with the results of
Kim’s study. First, this study introduces REVA proposed by
Bacidore et al. (1997) as an additional performance measure to
see if REVA is, in fact, superior in terms of explanatory power when
 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445
compared to EVA. Second, this study uses market adjusted returns
as a dependent variable following Biddle et al. (1997) because this
dependent variable is expected to be a better representation of
shareholders’ value than market capitalization. Third, this study
examines three sub-sectors of the hospitality industry (hotel,
restaurant and casino). Last, more recent and extensive sample
data (1985–2004) compared to Kim’s study (sample data from
1995 to 2001) is the data source for the current study. In addition,
this study uses the data provided by Stern Stewart & Co., not by
independent, unofficial estimates that may create inconsistent and
inaccurate measures.
3. Data
This study retrieved data from three main sources: (1) Stern
Stewart & Co.’s EVA database, (2) Compustat, and (3) the Center for
Research in Security Prices (CRSP). All annual EVA related data
analyzed in the study were from Stern Stewart & Co.’s EVA
database. The study collected annual financial data such as
operating income before depreciation, shareholders’ equity, debt,
and sales from Compustat, and stock price data from CRSP. The
sample period is 1985–2004 according to the EVA data availability,
and the total sample size for the analysis is 353 including three
sub-sectors of the hospitality industry: hotels (46), restaurants
(134) and casinos (135). The other 38 observations are from cruise
lines and theme parks. Since the sample size of each of these two
sectors is too small to perform a statistically meaningful analysis,
the study examines only the three sub-sectors of hotels,
restaurants and casinos.
After collecting the initial data set of 353 observations in total,
the study calculated Mahalanobis distance scores in all four
regression analyses (i.e., analyses for hospitality, hotel, restaurant,
and casino) to determine the existence of any outliers. According to
the method of Tabachnick and Fidell (2001), this study computes
chi-square statistics and determines appropriate cut-off values in
each regression analysis for p-value less than 0.001: hospitality
(n = 353) with x2 of 430, hotel (n = 46) with x2 of 70.7, restaurant
(n = 135) with x2 of 197, and casino (n = 135) with x2 of 197. Based
on the chi-square analysis, no observation was detected as an
outlier. Even with a more conservative level of significance
(p < 0.01), no outlier existed. Therefore, the total sample size of
353 remained for the main analysis.
4. Methodology
4.1. Model
The study performs a pooled regression analysis to compare
EVA to other performance measures (i.e., REVA, MVA, CFO, ROA and
ROE) in terms of explanatory power on market adjusted returns,
while controlling for firm size. The model is
MAR ¼ a0 þ a1 EVA þ a2 REVA þ a3 MVA þ a4 CFO þ a5 ROA
þ a6 ROE þ a7 SIZE þ e;
where MAR represents market adjusted return, estimated by
subtracting 12-month market return from a firm’s 12-month stock
return where the 12-month period ends 3 months following the
firm’s fiscal year-end; EVA represents economic value added,
retrieved from the database of the Stern Stewart Co. REVA
represents refined EVA, estimated by using total market value of
the firm’s assets in calculating total cost of capital, following
Bacidore et al. (1997). MVA represents market value added,
estimated by summing all projected EVAs in the future. CFO
represents cash flow from operations, retrieved from Compustat.
ROA is return on assets, estimated by dividing operating income
441
before depreciation (OIBD) by total assets. ROE is return on equity,
estimated by dividing OIBD by total shareholders’ equity, and SIZE
represents firm size, estimated by log of sales.
4.2. Market adjusted return
The study adopts market adjusted return (MAR) as a dependent
variable which has been widely used in information content
studies to measure a firm’s stock performance after adjusting for
market returns (Biddle et al., 1995; Biddle et al., 1997; Bowen et al.,
1989). This study calculates the difference between a firm’s stock
returns and market returns over the 12-month period, ending 3
months following the firm’s fiscal year-end. This adopted practice,
following Biddle et al. (1997), allows sufficient time for necessary
information from the firm’s annual report to be reflected in stock
market prices. All firm performance measures are expected to
present a positive explanatory power on the dependent variable
and the measurement with the highest explanatory power will be
considered a superior firm performance measure, compared to the
others.
4.3. EVA related measures
The study tests three EVA related performance measures: EVA,
REVA and MVA. EVA, as proposed by Stewart (1991, 1994) and
investigated by many other researchers, is examined in this study
to test its incremental explanatory power over the market adjusted
return when compared to other performance measures. EVA is
computed: NOPAT  (WACC  IC), where NOPAT is net operating
profit after tax; WACC is weighted average cost of capital, and IC is
invested capital. EVA basically represents a firm’s profit from
operations after consideration of the cost of capital. Therefore, a
firm’s value is evaluated to be increasing only when the firm’s
operating profit after tax is greater than its cost of capital. The
study expects to find a positive impact from EVA on MAR.
MVA is market value added, derived from the EVA concept.
MVA is a present value of all projected EVAs of a firm in the future
and presents the extent to which a firm has added value to its
capital, funded from shareholders and lenders (Milunovich and
Tsuie, 1996). The study retrieved the MVA data directly from Stern
Stewart & Co. and expects to find a positive relationship between
MVA and MAR.
The third EVA related measure is refined EVA proposed by
Bacidore et al. (1997). As previously discussed, EVA is estimated as:
NOPAT  (WACC  IC), where IC (invested capital) represents the
total economic book value of the firm’s assets. Bacidore et al. (1997)
argued that estimating IC as the economic book value of assets is
not desirable because the capital charge for the firm is based on its
market-based WACC, not book-based. Accordingly, Bacidore et al.
proposed to estimate EVA by using the market value of the firm’s
assets (i.e., the market value of the firm’s equity plus the book value
of the firm’s total debt less non-interest-bearing current liabilities)
instead. They called this new type of EVA, REVA; this study
examines REVA as one of main performance measures of firms.
4.4. Other variables
The current study analyzes three traditional accounting firm
performance measures: cash flow from operations (CFO), return on
assets (ROA) and return on equity (ROE). The study adopts CFO as
defined by Biddle et al. (1997): (1) For years after 1987, Compustat
data item D308, ‘‘operating activities – net cash flow,’’ is used. And,
(2) For years before 1988, according to Bowen et al. (1986, 1987),
this study uses the definition: funds from operations (D110) + D in
current liabilities (D5)  D in debt in current liabilities (D34)  D
in current assets (D4) + D in cash and cash equivalents (D1), where
442 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445

D denotes changes. ROA (ROE) is: Operating Income Before
Depreciation/Total Assets (Total Equity). In addition to the three
accounting performance measurements, the study includes a
control variable, SIZE, to control for any systematic effects
generated by firm size. SIZE is estimated by the natural log of
sales and a negative relationship is expected with MAR according
to Berk (1995) that argued for a negative relationship between SIZE
and the expected return. In the hospitality literature, Lee and
Upneja (2008) also found a negative relationship between SIZE and
the cost of equity capital.
USD), the hotel sector shows the greatest number ($2,984.99),
followed by the restaurant sector ($2,644.12) and the casino sector
($1,085.34). The hospitality industry as a whole presents a mean
value of $2,035.78 for its sales. Last, in the summary of market
adjusted returns (MARs), the casino sector appears to be the best
performer in terms of MARs, followed by the restaurant sector
(0.10) and the hotel sector (0.01). The hospitality industry’s mean
MAR is 0.11 for the sample period.
5.2. Summary of Pearson’s correlations

5. Findings The study next analyzes correlations among the variables as a

preliminary examination and also to see any potential relation-
5.1. Summary of descriptive statistics ships that may raise a question of a multicollinearity problem
(Table 2). With the hospitality sample, EVA shows a significant
The study first summarizes descriptive statistics of the variables relationship with all other variables except with MAR (r = 0.05). All
included in the model; Table 1 presents results. One interesting significant relationships appear to be positive, except for REVA
finding appearing in Table 1 is that during the sample period, REVA (r = 0.18; p-value < 0.01). The relationships for the hospitality
for all three hospitality sectors and the entire hospitality industry industry in terms of EVA seem to derive mainly from the restaurant
had negative values (all in millions USD) on average: for the sector where similar relationships of EVA with other variables
hospitality sample, $135.32, the hotel sample, $195.97, the appear. The only difference in findings between the restaurant
restaurant sample, $131.38, and the casino sample, $68.25. For sector and the hospitality industry is that EVA positively correlates
EVA (in millions USD), the hospitality sample shows a positive with REVA in the restaurant sector (r = 0.42; p-value < 0.01). This
value ($7.77) that appears to be mainly attributed from a positive positive relationship between EVA and REVA persists in the hotel
EVA of the restaurant sample ($48.39). EVAs of the hotel and casino and casino sectors.
samples, on average, are negative with values of $86.69 and In addition to a relationship with EVA, REVA shows a negative
$2.84, respectively. On the other hand, MVA (in millions USD) relationship with MVA in the hospitality (r = 0.70), restaurant
appears to be positive for all four sample groups: the hospitality (r = 0.73) and casino (r = 0.23) samples. REVA positively
sample, $2,864.13, the hotel sample, $2,314.47, the restaurant correlates with MAR in the hospitality (r = 0.10) and casino
sample, $3,989.27, and the casino sample, $1,205.69. REVA with (r = 0.16) samples, but does not correlate in the hotel and
more negative values than EVA can be explained by REVA having
Table 2
greater cost of capital due to a use of market value, which is
Pearson’s correlation matrix.
normally larger than book value which is used in EVA calculation to
compute cost of capital. Also, according to the three EVA-related REVA MVA CFO ROA ROE SIZE MAR
measures, the restaurant sector shows the best firm performance Total
during the sample period compared to the hotel and casino sectors. EVA 0.18** 0.57** 0.44** 0.44** 0.16** 0.20** 0.05
For the three traditional accounting performance measures, REVA 0.70** 0.70** 0.09 0.08 0.20** 0.10*
MVA 0.84** 0.07 0.04 0.46** 0.04
again, the restaurant sector shows better performance in general.
CFO 0.04 0.02 0.62** 0.02
The four sample groups show a mean value of CFO (in millions ROA 0.32** 0.16** 0.12*
USD) of $338.66 for the hospitality group, $326.62 for the hotel ROE 0.17** 0.14**
group, $426.03 for the restaurant group, and $193.04 for the casino SIZE 0.09
group. ROAs (ROEs), on average, for the hospitality, hotel, Hotel
restaurant and casino groups are 0.16 (0.42), 0.08 (0.36), 0.20 EVA 0.66** 0.14 0.32* 0.10 0.05 0.44** 0.06
(0.49), and 0.15 (0.49), respectively. Regarding SALEs (in millions REVA 0.44 0.56** 0.22 0.11 0.51** 0.05
MVA 0.76** 0.38** 0.05 0.63** 0.22
CFO 0.45** 0.15 0.81** 0.16
Table 1 ROA 0.41** 0.51** 0.30*
Descriptive statistics. ROE 0.07 0.15
SIZE 0.05
Variables EVA ($) REVA ($) MVA ($) CFO ($) ROA ROE SALE ($) MAR
Restaurant
Total (n = 353) EVA 0.42** 0.77** 0.72** 0.23** 0.22** 0.55** 0.00
Mean 7.77 135.32 2864.13 338.66 0.16 0.42 2035.78 0.11 REVA 0.73** 0.71** 0.08 0.05 0.43** 0.15
Max 551.30 629.45 50379.94 3903.60 0.41 3.44 19064.70 2.06 MVA 0.87** 0.03 0.04 0.61** 0.01
Min 425.44 2776.28 1808.50 114.77 0.16 0.75 1.00 1.39 CFO 0.04 0.06 0.70** 0.08
Hotel (n = 46) ROA 0.11 0.06 0.13
Mean 86.69 195.97 2314.47 326.62 0.08 0.36 2984.99 0.01 ROE 0.18* 0.10
Max 58.54 544.05 10261.61 891.00 0.15 3.22 12034.00 1.87 SIZE 0.13
Min 425.44 755.84 1808.50 6.12 0.16 0.44 32.77 1.40 Casino
Restaurant (n = 134) EVA 0.41** 0.50** 0.03 0.52** 0.34** 0.17* 0.07
Mean 48.39 131.38 3989.27 426.03 0.20 0.49 2644.12 0.10 REVA 0.23** 0.34** 0.19* 0.17* 0.18* 0.16*
Max 551.30 629.45 50379.94 3,903.60 0.35 11.35 19064.70 1.74 MVA 0.31** 0.13 0.10 0.15 0.18*
CFO 0.08 0.15 0.70** 0.03
Min 124.20 2776.28 278.43 1.53 0.00 0.00 28.88 1.05
ROA 0.48** 0.23** 0.09
Casino (n = 135) ROE 0.32** 0.21*
Mean 2.84 68.25 1205.69 193.04 0.15 0.49 1085.34 0.17 SIZE 0.08
Max 257.94 193.06 13594.43 1031.14 0.41 3.44 4896.00 2.06
Min 273.17 501.22 1625.55 114.77 0.07 0.75 1.00 1.39 All data is at firm level; Total includes hotels, restaurants, casinos, cruise lines and a
theme park; REVA represents refined EVA; MAR represents market adjusted return.
*
All data is at firm level; Total includes hotels, restaurants, casinos, cruise lines and a Represents significance levels of 0.05.
**
theme park; REVA represents refined EVA; MAR represents market adjusted return. Represents significance levels of 0.01.
 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445 443

restaurant samples. MVA consistently demonstrates a high and (t-value = 2.35; p-value < 0.01), and casino (t-value = 3.37; p-
positive correlation with CFO and SIZE in all four samples, except value < 0.001) sectors, and the same positive incremental
with SIZE in the restaurant sample (r = 15, p-value > 0.05). MVA explanatory power of MVA on MAR appears in the hospitality
shows a significant correlation with MAR only in the casino sample (t-value = 3.14; p-value < 0.01), hotel (t-value = 2.46; p-
(r = 0.18). value < 0.05), restaurant (t-value = 2.10; p-value < 0.05), and
CFO shows a positive correlation with SIZE and no correlation casino (t-value = 3.08; p-value < 0.05) samples.
with MAR for the all four samples. ROA and ROE show a positive The three traditional accounting firm performance measures do
correlation in the hospitality, hotel and casino samples, but not in not, in general, present an incremental explanatory power on MAR,
the restaurant sample (r = 0.11; p-value > 0.05). ROA demon- except a few cases. ROE shows a positive impact in the hospitality
strates a positive correlation with MAR in the hospitality (r = 0.12) (t-value = 2.59; p-value < 0.05) and casino (t-value = 2.42; p-
and hotel (r = 0.30) samples; whereas, ROE shows a positive value < 0.05) samples; whereas, ROA shows a positive relationship
correlation with MAR in the hospitality (r = 0.14) and casino with MAR only in the hotel sector (t-value = 3.22; p-value < 0.01).
(r = 0.21) samples. SIZE generally presents a positive correlation SIZE variable consistently shows a negative and significant impact
with variables throughout the four samples, but not a single on MAR, except in the restaurant sector (t-value = 1.74; p-
correlation with SIZE appears to be significant with MAR in all four value > 0.05), implying that larger size results in worse market
samples. With regard to the multicollinearity problem, only a few adjusted return.
relationships appear to signal caution: CFO-MVA and CFO-SIZE By examining adjusted R2 and F-values, the model analyzed in
correlations higher than 0.80. Accordingly, the main analysis this study seems to explain MAR better for the hotel (adjusted
estimates a variance inflation factor (VIF) to examine the issue. R2 = 0.19; F-value = 2.50) and casino (adjusted R2 = 0.15; F-
value = 4.36) sectors. Therefore, although, in the restaurant sector,
5.3. Main findings REVA and MVA explain MAR at a statistically significant level, their
practical explanatory power should be cautiously interpreted. The
The study performs a pooled regression as the main analysis study also estimates VIF to determine whether a severe multi-
and Table 3 presents the results. First, EVA shows a negative collinearity problem exists in any of the four analyses. According to
explanatory power for MAR in the hospitality (t-value = 2.34; p- Ott and Longnecker (2001), who contended that VIF less than 10 is
value < 0.05), hotel (t-value = 2.31; p-value < 0.05), and casino acceptable, a severe multicollinearity problem does not seem to
(t-value = 2.35; p-value < 0.05) samples, but does not explain exist in this study; the highest VIF appears in the restaurant sample
MAR in the casino sample which is, in general, inconsistent with with value of 6.98.
the expected positive impact. On the other hand, REVA and MVA
show a positive and significant explanatory power on MAR in all 6. Discussion and suggested future research
four samples, except that REVA demonstrates an insignificant
relationship with MAR in the hotel sample (t-value = 1.66). In The purpose of the study is to compare incremental explanatory
detail, a positive incremental impact of REVA on MAR exists in power of six firm performance measures: three EVA-related
the hospitality (t-value = 3.37; p-value < 0.001), restaurant performance measures (i.e., EVA, REVA, and MVA) and three
traditional accounting performance measures (i.e., CFO, ROA, and
Table 3 ROE) on the market adjusted return, controlled for firm size. The
Pooled regression analysis with EVA, REVA, MVA, CFO, ROA and ROE MAR = -
a0 + a1EVA + a2REVA + a3MVA + a4CFO + a5ROA + a6ROE + a7SIZE + e.
study makes contributions, not only by introducing a modified EVA
version (i.e., REVA), but also by investigating, separately, the three
EVA REVA MVA CFO ROA ROE SIZE hospitality sub-sectors (i.e., hotel, restaurant, and casino). In
addition, examining a more closely considered dependent variable,
Total market adjusted return, makes the analysis more valid and
Coeff 0.19 0.28 0.37 0.09 0.11 0.14 0.23 reliable; the dependent variable adjusts for market return and
t-value 2.34* 3.37*** 3.14** 0.78 1.77 2.59* 3.25***
uses a calculation period beginning 3 months after a firm’s 10K
VIF 2.37 2.59 5.12 5.28 1.47 1.13 1.85
Adj. R2 0.07 reporting month. The latter calculation practice complies better
F-value 4.75*** with a generally accepted market efficiency theory: semi-strong
Hotel
efficient market. This is because in a semi-strong efficient market,
Coeff 0.58 0.36 0.54 0.14 0.76 0.17 0.98 stock prices do not tend to reflect insider information, but only
t-Value 2.31* 1.66 2.46* 0.50 3.22** 1.05 3.21** public information such as annual reports (10Ks). Therefore,
VIF 3.46 2.67 2.72 4.35 3.12 1.53 5.16 considering the fact that such public information becomes
Adj. R2 0.19
available within a nominal 3 months after the accounting period
F-value 2.50*
ends, analysis is more real and reliable.
Restaurant General findings of the study support several previous studies
Coeff 0.22 0.35 0.48 0.08 0.12 0.12 0.22
(Biddle et al., 1997; Chen and Dodd, 2001; de Villiers, 1997; Kim,
t-Value 1.27 2.35* 2.10* 0.36 1.28 1.33 1.74
VIF 4.16 3.02 6.98 6.39 1.19 1.14 2.14 2006; Weissenrieder, 1998; Zimmerman, 1997) that EVA does not
Adj. R2 0.05 appear to be as a good firm performance measure as otherwise
F-value 1.89 thought. Moreover, EVA shows a negative explanatory power in
Casino general which is different from the expected positive explanatory
Coeff 0.35 0.36 0.40 0.18 0.07 0.23 0.26 power. Acceptable VIFs found in the study suggest that the
t-Value 2.35* 3.37** 3.08* 1.33 0.64 2.42* 2.04* negative coefficients of EVA are not due to a multicollinearity
VIF 3.50 1.85 2.65 2.83 1.70 1.37 2.64
problem that often leads to producing different (from expecta-
Adj. R2 0.15
F-value 4.36***
tions) and unstable coefficients. At this point, why the negative
coefficient is prevalent between EVA and the market adjusted
All data is at firm level; Total includes hotels, restaurants, casinos, cruise lines and a return is unclear, beyond the possibility that EVA is simply an
theme park; REVA represents refined EVA; MAR represents market adjusted return.
*
Represents significance levels of 0.05.
inaccurate performance measure. Some may argue that a possible
**
Represents significance levels of 0.01. reason for a negative relationship between EVA and the market
***
Represents significance levels of less than 0.001. adjusted return appeared during the sample period because the
444 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445
financial market mainly used other performance measures rather
than EVA in evaluating hospitality companies. However, the fact
that REVA and MVA which are other EVA-related performance
measures show good explanatory power with positive coefficients
may weaken such an argument for EVA. Future studies are clearly
required to shed more lights on this issue.
The study provides interesting and meaningful findings that
REVA and MVA can be considered good performance measures
throughout the three hospitality sectors (i.e., hotel, restaurant and
casino). According to the findings, REVA and MVA significantly
explain the market adjusted return by presenting positive
coefficients. A Z-test to identify any statistical differences in terms
of the explanatory power between REVA and MVA reveals no
differences. Therefore, either or both REVA and MVA can be used as
a good performance measures for hospitality firms.
In addition, the study found that traditional accounting
performance measures (i.e., CFO, ROA, and ROE) do not explain
much of market adjusted return after considering REVA and MVA.
One of a few exceptions is that ROA shows a positive explanatory
power only in the hotel sector. This finding is, in fact, consistent
with the findings of Lee (2008). Therefore, for the hotel sector, a
cautious step in determining performance measures may be
required, and ROA should not be dismissed but considered an
important performance measure along with MVA. REVA’s positive
coefficient is found to be not statistically significant at the 5% level,
but significant at the 10% level. This low level of significance could
be due to small sample size for the hotel sector with 46
observations, but this is an empirical issue and a further
examination is necessary.
One possible explanation of the study findings that REVA and
MVA explain more of MAR is because of the common nature of the
three variables (i.e., REVA, MVA and MAR): future-oriented
whereas other accounting performance measures focus on the
past year’s firm performance. However, considering that even ROA,
an accounting performance measure focusing only on the past
year, shows a significant explanatory power in a hotel setting, it is
premature to generalize this explanation and we leave this issue
for future studies.
The findings of this study provide not only theoretical
contributions to the hospitality EVA literature, but also practical
implications to hospitality related academicians and practitioners.
Whoever attempts to evaluate hospitality firm performances or
managers’ performances can adopt REVA or MVA as an evaluation
tool in addition to, or by replacing, traditional accounting
performance measures. For example, in forming a hospitality
investment portfolio, analysts can utilize REVA or MVA as a
portfolio selection criterion; whereas boards of directors of
hospitality companies can adopt such performance measures to
determine their executives’ or managers’ compensations. Acade-
micians, especially hospitality researchers in financial manage-
ment areas, may want to examine REVA or MVA as a key
performance measure in their studies.
7. Limitations
Although the current study makes several meaningful
contributions, several limitations exist. First, the sample period
represents 1985–2004, according to the availability of EVA and
EVA-related data directly from Stern Stewart & Co. The most
recent years are excluded from the analysis and the findings of
this study may not be generalizable to such periods. However,
considering such excluded periods are only 3 years, impact may
be minimal rather than substantial. Second, according to the
findings, practical use of REVA or MVA may require potential
users to purchase the data set from Stern Stewart & Co. rather
than generating the measures independently. This is not a
critical issue because the cost of purchasing the data set is
comparatively low for firms included in this study. Moreover,
practical calculation steps of EVA can be found in Kim (2006) and
by using those steps, one can also estimate REVA and MVA.
However, caution is necessary when adopting this approach
because such estimates may be inconsistent with the data set
provided by Stern Stewart & Co.
References
Aliouche, E.H., Schlentrich, U., 2005. Does franchising create value? An analysis of
the financial performance of US public restaurant firms. In: Proceedings of the
Annual International Society of Franchising Conference, London.
Bacidore, J.M., Boquist, J.A., Milbourn, T.T., Thakor, A.V., 1997. The search for the best
financial performance measure. Financial Analysts Journal May/June 11–20.
Berk, J., 1995. A critique of size-related anomalies. Review of Financial Studies 8 (2),
275–286.
Biddle, G., Seow, G., Siegel, A., 1995. Relative versus incremental information
content. Contemporary Accounting Research 12 (1), 1–23.
Biddle, G.C., Bowen, R.M., Wallace, J.S., 1997. Does EVA beat earnings? Evidence on
associations with stock returns and firm values. Journal of Accounting and
Economics 24 (3), 301–336.
Bowen, R., Burgstahler, D., Daley, L., 1986. Evidence on the relationships between
earnings and various measures of cash flow. The Accounting Review 61 (4),
713–747.
Bowen, R., Burgstahler, D., Daley, L., 1987. The incremental information content of
accrual versus cash flows. The Accounting Review 62 (4), 723–747.
Bowen, R., Johnson, M., Shevlin, T., Shores, D., 1989. Informational efficiency and the
information content of earnings during the stock market crash of 1987. Journal
of Accounting and Economics 11 (2/3), 225–254.
Brossy, R., Balkcom, J.E., 1994. Getting executives to create value. Journal of Business
Strategy January/February 18–21.
Chen, S., Dodd, J.L., 1997. Economic value added (EVA): an empirical examination of
a new corporate performance measure. Journal of Managerial Issues 9 (3), 318–
333.
Chen, S., Dodd, J.L., 2001. Operating income, residual income and EVA: Which metric
is more value relevant. Journal of Managerial Issues 13 (1), 65–86.
Chow, C.W., Haddad, K.M., Leung, S., Sterk, W., 2003. Using value-based perfor-
mance measures to improve hotel profitability. International Journal of Hospi-
tality & Tourism Administration 4 (1), 23–50.
Clinton, B.D., Chen, S., 1998. Do new performance measures measure up? Manage-
ment Accounting 80 (4), 38–43.
de Villiers, J., 1997. The distortions in economic value added (EVA) caused by
inflation. Journal of Economics and Business 49 (3), 285–300.
Ezzamel, M., Burns, J., 2005. Professional competition, economic value added and
management control strategies. Organizations Studies 26 (5), 756–776.
Fisher, A., 1995. Creating stockholder wealth. Fortune 132, 105–117.
Fletcher, H.D., Smith, D.B., 2004. Managing for value: developing a performance
measurement system integrating economic value added and the balanced
scorecard in strategic planning. Journal of Business Strategies 21 (1), 1–17.
Garvey, G.T., Milbourn, T.T., 2000. The optimal and actual use of EVA versus earnings
in executive compensation. Working Paper. Claremont Colleges.
Grant, J.L., 1996. Foundations of EVA for investment managers. Journal of Portfolio
Management 23 (1), 41–48.
Kefgen, K., Mahoney, R., 1996. Economic Value Added: A New Performance Measure
for Incentive Pay. Retrieved: October 2, 2008, from http://www.hotel-online.-
com/Trends/HVS/Kefgen/PerformanceMeasure_IncentivePay.html.
Kim, W.G., 2006. EVA and traditional accounting measures: which metric is a better
predictor of market value of hospitality companies? Journal of Hospitality &
Tourism Research 30 (1), 34–49.
Kramer, J.K., Pushner, G., 1997. An empirical analysis of economic value added as a
proxy for market value added. Financial Practice and Education 7 (1), 41–49.
Lee, S., 2008. Examination of various financial risk measures for lodging firms.
Journal of Hospitality & Tourism Research 32 (2), 255–271.
Lee, S., Upneja, A., 2008. Is Capital Asset Pricing Model (CAPM) the best way to
estimate cost-of-equity for the lodging industry? International Journal of
Contemporary Hospitality Management 20 (2), 172–185.
Lehn, K., Makhija, A.K., 1996. EVA & MVA as performance measures and signals for
strategic change. Strategy & Leadership 24 (3), 34–38.
Lieber, R., 1996. Who are the real wealth creators? Fortune 134, 107–116.
Madanoglu, M., Chang, D.Y., Chu, Y.-H., 2004. Creating economic value in the US
airline industry: are we missing the flight? International Journal of Contem-
porary Hospitality Management 16 (5), 294–298.
Milunovich, S., Tsuie, A., 1996. EVA in the computer industry. Journal of Applied
Corporate Finance 9 (1), 104–115.
Mir, A.E., Seboui, S., 2006. Corporate governance and earnings management and the
relationship between economic value added and created shareholder value.
Journal of Asset Management 7 (3/4), 242–254.
O’Byrne, S.F., 1996. EVA and market value. Journal of Applied Corporate Finance 9
(1), 116–125.
Ott, R.L., Longnecker, M., 2001. Statistical Methods and Data Analysis, 5th ed.
Duxbury, Pacific Grove.
Rutledge, J., 1993. De-jargoning EVA. Forbes 152, 148.
 S. Lee, W.G. Kim / International Journal of Hospitality Management 28 (2009) 439–445
Stewart III, G.B., 1991. The quest for value: The EVA management guide. Harper
Business, New York.
Stewart III, G.B., 1994. EVATM—Fact or fantasy. Journal of Applied Corporate Finance
7 (2), 71–84.
Tabachnick, B.G., Fidell, L.S., 2001. Using Multivariate Statistics, 4th ed. Allyn &
Bacon, Boston.
Teitelbaum, R., 1997. America’s greatest wealth creators. Fortune 136, 265–276.
Tully, S., 1993. The real key to creating wealth. Fortune September 20, 38–50.
Uyemura, D.G., Kantor, C.C., Pettit, J.M., 1996. EVA1 for banks: value creation, risk
management, and profitability measurement. Journal of Applied Corporate
Finance 9 (2), 94–113.
445
Walbert, L., 1994. The Stern Stewart performance 1000: using EVA to build market
value. Journal of Applied Corporate Finance 6 (4), 109–120.
Weissenrieder, F., 1998. Value based management: economic value added or cash
value added? Working Paper. Gothenburg University.
Zimmerman, J.L., 1997. EVA and divisional performance measurement: capturing
synergies and other issues. Journal of Applied Corporate Finance 10 (2),
99–109.
Zinkhan, G.M., Verbrugge, J.A., 2000. Special issue on the marketing/finance inter-
face. Journal of Business Research 50 (2), 139–142.
Zinkin, J., 2006. Strategic marketing: balancing customer value with shareholder
value. The Marketing Review 6 (2), 163–181.