Janis Zaima
Professor, San Jose State University, USA.
Kenneth Leong
San Jose State University, USA.
Stoyu Ivanov
Assistant Professor, San Jose State University, USA.A Profile of Negative EVA Firms:
Reality versus Expectations
Numerous studies have examined the relationship between market
value-added (MVA) and economic value-added (EVA), both metrics
popularized by Stern Stewart & Company. Although most studies
found a positive relationship between EVA and MVA (or stock returns),
Abate, Grant and Stewart [2004] found an anomalous observation
where some firms with negative EVAs led to higher returns. In a
separate study, Zaima [2008] investigated portfolio investing with EVA,
and found that firms with the most negative EVAs exhibited the highest
portfolio returns. Abate, Grant, and Stewart conjecture that large
returns associated with negative EVA firms may be a result of “EVA
future investment opportunities”. This study examines how negative
EVA firms fare in the future to determine whether they, indeed, realize
the “EVA future investment opportunities” or whether investor
expectations are overly optimistic. Moreover, the study sheds some
light on ways to distinguish negative EVA firms that realize their
potential growth versus the ones that do not.
Firms with negative EVAs in 2003 are identified from the Stern
Stewart 2003 database and cross sectional-time series data are obtained
from CRSP and COMPUSTAT over the 2004 to 2009 period. The firms
are categorized into quartiles using negative EVA data from 2003 and
financial profiles - net operating profits after taxes/total asssets
(NOPAT/TA), market-to-book ratios (MTB), leverage (DR), size (LnTA)
as well as the four quartiles (C1, C2, C3 and C4) - are examined. C1
represents the quartile of firms with the most negative EVA while C4
consists of firms with the last negative EVAs.
In summary, even though univariate analysis suggest that the most
negative EVA firms (C1) showed promise with its large capital
investment, in reality the results exhibited returns that were less than
expected. Multivariate analysis results show that MTB and
NOPAT/TA exhibit statistically significant relationship to portfolio
returns. We also find that leverage, size, and the four groups of firms
categorized by 2003 EVA are correlated to ex-ante portfolio returns, but
only the firms in C4 (firms with least negative EVAs) displayed
significant alphas.