Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...
Gadjah Mada International Journal of Business
September-December 2006, Vol. 8, No. 3, pp. 281–299

COMPETITIVE INTENSITY AS A QUASIMODERATOR OF THE RELATIONSHIP
BETWEEN INNOVATIVE EFFORTS AND
PERFORMANCE*
Johnny Jermias
This study empirically investigates the moderating effect of
competitive intensity on the relationship between innovative efforts
and performance. The study proposes that a firm’s competitive
intensity acts as a quasi-moderator of the relationship between
innovative efforts and performance.
The results support the predictions of the study. The results
indicate that competitive intensity moderates the relationship between innovative efforts and performance such that when competitive intensity is high, the relationship between innovative efforts and
performance is positive, and vice versa, when the competitive
intensity is low, the relationship between innovative efforts and
performance is negative. Furthermore, the results reveal that competitive intensity has a positive relationship to the level of innovative
efforts but has a negative relationship to performance.
Keywords: competitive intensity; innovative efforts; performance; quasi-moderator

* Financial support from the Social Science and Humanities Research Council of Canada is
gratefully appreciated. This paper has benefited from comments and suggestions of David Finley,
Ann-Marie Grbavec, and participants at Simon Fraser University and University of Persada
Indonesia accounting seminars.
281

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

Introduction
The relationship between innovative efforts and performance has attracted much attention in the management and business literature. Despite
all these efforts, empirical studies investigating the influence of innovative
efforts on performance often report
weak and inconclusive findings. Some
scholars (e.g., Burton et al. 2002; Foster 1986) have argued that the relationship between innovative efforts and
firms’ performance may be contingent
upon the competitive environment in
which the firms operate.
The purpose of this study is to
investigate the moderating effect of
competitive intensity on the relationship between innovative efforts and
performance. This study predicts that
competitive intensity acts as a quasimoderator of the relationship between
innovative efforts and performance.
As a quasi-moderator, competitive intensity will interact with innovative
efforts to affect performance. In addition, competitive intensity will affect
both the level of innovative efforts and
performance. Hence, this study inquires of the following research question: Does competitive intensity moderate the relationship between innovative efforts and performance?
The results support the predictions of this study. The results indicate
that competitive intensity modifies the
relationship between innovative efforts
and performance. When the competitive intensity is high, innovative efforts positively affect performance. On
282

the contrary, when competitive intensity is low, innovative efforts negatively affect performance. Furthermore, the results reveal that competitive intensity acts as a quasi-moderator of the relationship between innovative efforts and performance. As such,
competitive intensity has a positive
relationship with innovative efforts but
has a negative relationship with performance.
These findings are consistent with
the argument that when competitive
intensity is high, firms should invest
heavily in innovative efforts to increase their ability to produce unique
products/services and to gain competitive advantage (Porter 1985). By
contrast, when competitive intensity is
low, firms should focus their efforts on
achieving efficiency through the construction of efficient scale facilities,
vigorous pursuit of cost reduction from
experience, and cost minimization in
areas such as research and development (Miller 1988; Porter 1985). The
results suggest that managers need to
understand the nature of their firms’
competitive intensity before investing
in innovative efforts. This study indicates that achieving a good match between innovative efforts and types of
competitive intensity is likely to affect
performance positively.
The remainder of this paper is
organized as follows. Section two reviews the related literature as the basis
to develop the hypotheses. Section
three describes the research methods
employed to examine the hypotheses.
Section four presents the results of

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

statistical analyses. Section five discusses the main findings of this study,
limitations, and directions for future
research in this area.

Related Literature and
Hypothesis
Despite the widely-held premise
that innovation is crucial for long-term
survival of a company, empirical findings with respect to the effect of innovative efforts on performance have
been inconclusive. Based on a survey
of 400 senior executives of U.S. technological firms, Calantone et al. (2002),
for example, find a positive association between innovative efforts and
performance. Similarly, using a sample
of 531 firms from Northern Ireland
and the Republic of Ireland, Jin et al.
(2004) report that firms with high level
of investment in innovative efforts
perform better than do firms with low
level of investment in innovative efforts.
Other studies, however, have reported a negative relationship between
innovative efforts and performance.
O’Brien (2003) investigates the relationship between innovative efforts,
as measured by research and development expenditures, and the performance of firms. Using the Compustat
industrial and business segments database, he finds that innovative efforts
are negatively associated with performance. In a similar vein, based on a
survey of 31 Indonesian publicly-held
firms listed on the Jakarta Stock Exchange, Jermias and Armitage (2000)

report a negative relationship between
innovative efforts and performance.
Miller (1988) suggests that researchers need to investigate the moderating effects of the environment on
the relationship between innovative
efforts and performance since the relationship between these two variables
might be different across different environmental conditions. In a similar
vein, Foster (1986) emphasizes the
need to consider the environment in
which a firm operates when investigating the effect of innovative efforts
on performance. He argues that a generic prescription of the relationship
between innovative efforts and performance can be as harmful as they are
helpful because innovative efforts can
only be effective under high competitive intensity.
This study adopts a contingency
approach and argues that competitive
intensity acts as a quasi-moderator of
the relationship between innovative
efforts and performance. Sharma et al.
(1981) propose that as a quasi-moderator, the moderator variable should
interact with the predictor variable to
modify the form of the relationship
between the predictor and the criterion
variables. In addition, they stipulate
that the moderator variable should also
have a significant relationship with
both the predictor and the criterion
variable. Related to this study, the
quasi-moderating relationship between
innovative efforts and performance
suggests that the impact of innovative
efforts on performance varies depending upon the level of competitive in283

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

tensity. In addition, competitive intensity affects both the level of innovative
efforts and the performance of the
firms.
Milliken (1987) suggests that
when competitive intensity is low,
firms are able to assess both the present
and future state of the environment
reasonably well, enabling them to determine the potential impact of decision making on current and future business activities. In such an environment, firms will gain competitive advantage by planning their activities
reasonably well and realize efficiency
(see also Burton et al. 2002). Porter
(1985) suggests that to increase efficiency, firms should produce standard
products and employ standardized operating procedures. Miller (1987) contends that although these practices reduce flexibility in responding to
changes in technology, customer demand, and competitors, they increase
efficiency by narrowing the range of
tasks and by making activities more
routine. Increased efficiency will, in
turn, lower the cost of products or
services and will lead to superior performance (Porter 1985).
In contrast, when competitive intensity is high, firms should invest
heavily in research and development
activities to help them create firmspecific assets (D’Aveni 1994). Likewise, Miller (1987) asserts that investment in research and development activities increases firms’ ability to produce new and better products and help
them keep up with the innovations of

284

competitors and the changing environment.
Chandler (1962) proposes that innovation is most likely to be found in
industries characterized by high competitive intensity in which the rates of
new product introduction and technological change are high. If competitive
environment is a critical factor in determining the level of innovative efforts and firms need to properly adjust
the level of innovative efforts to adapt
to the environment in which the firms
operate, then a proper match between
innovative efforts and competitive
environment should enhance the firms’
performance.
Previous discussion suggests that
the impact of innovative efforts on
performance varies, depending upon
the level of competitive intensity.
When the competitive intensity is high,
innovative efforts will have a positive
impact on performance due to an increased ability to produce new and
better products to keep up with competitors’ innovations and the changing
environment. When competitive intensity is low, however, innovative
efforts will have a negative impact on
performance due to the need to increase efficiency by minimizing costs,
including research and development
activities. Therefore, the following
hypothesis is examined:
H1: The relationship between innovative efforts and performance is
moderated by competitive intensity such that when competitive
intensity is high, the relationship

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

between innovative efforts and
performance will be positive, but
when competitive environment is
low, the relationship between innovative efforts and performance
will be negative.
Furthermore, this study proposes
that competitive intensity acts as a
quasi-moderator for the relationship
between innovative efforts and performance. As a quasi-moderator variable,
competitive intensity will affect the
level of innovative efforts (Miller 1987;
Tushman and Anderson 1986; Bourgeois 1985) and performance
(Calantone et al. 2002; Baker and
Sinkula 1999; Prescott 1986; Dess and
Beard 1984). Miller (1987) argues that
when the competitive intensity is high,
innovation is crucial to determining
firms’ competitive advantage and market success. Innovative products enable the firms to adapt to their rapidly
changing environment (Miles and
Snow 1978) and to take advantage of
emerging trends (Miller 1987). Innovation-based competition thus necessitates the creation and development
of products that are superior to others
in the market (Mia and Clarke 1999).
Consequently, firms operating in high
competitive intensity are often characterized by their heavy investments in
research and development activities
(Porter 1985). By contrast, when the
competitive intensity is low, firms tend
to be focused on efficiency by producing standardized products, using routine operating procedures, and limiting their spending on areas such as
research and development (Miller

1987). These practices enhance the
firms’ ability to gain competitive advantage by becoming the lowest producers in the industry.
If competitive intensity influences
the level of innovative efforts, there
will be a good reason to predict that
firms operating in high competitive
intensity will spend more on innovative efforts than will those operating in
low competitive intensity. Specifically,
the following hypothesis is examined:
H2: There is a positive relationship
between competitive intensity and
innovative efforts.
Competitive intensity has been
hypothesized and empirically demonstrated to have significant effects on
performance (Keats and Hitt 1988;
Dess and Beard 1984; Porter 1985).
High competitive intensity may influence a firm’s performance negatively
due to an increased risk and uncertainty surrounding the firm’s products
and services (Keats and Hitt 1988).
Empirical studies have consistently
reported that competitive intensity
negatively affects performance. For
example, based on data from 110 large
manufacturing firms listed on Fortune
500 database, Keats and Hitt (1988)
report that competitive intensity has a
negative relationship with operating
performance. Similar results are also
reported by Prescot (1986) and Dess
and Beard (1984). Therefore, the following hypothesis is tested:
H3: There is a negative relationship
between competitive intensity and
performance.
285

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

This study uses the following
moderated regression model to test the
hypotheses:
PERFORMi = 0 + 1 INOVi +
2 ENVi +
3 INOVi*ENVi +
4 SIZEi +
5 LEVi + 
where:
PERFORMi : Financial performance
of firm i in terms of ROA
or ROS
The level of innovative
INOVi :
efforts as measured by
research and development expenses divided
by net sales revenues.
An indicator equal to 1
ENVi :
for firms competing in
low competitive intensity and 0 for firms competing in high competitive intensity.
The size of firm i as meaSIZEi :
sured by the logarithmic
function of total assets.
LEVi :
Leverage of firm i as
measured by the ratio of
debt to equity.
Hypothesis H1 predicts that the
coefficient of INOV*ENV will be
negative and significant. In addition,
hypothesis H1 also implies that the
coefficient of INOV will be positive
but the sum of the coefficients of INOV
and INOV*ENV will be negative. To
test hypotheses H2 and H3, this study
employs the three-step moderated re286

gression analysis procedures suggested
by Sharma et al. (1981). Assuming
that the coefficient of INOV*ENV will
be significant (Hypothesis H1), Hypothesis H2 will be examined by regressing the performance variables on
ENV without the interaction term. It is
expected that the coefficient of ENV
will be significant. To test hypothesis
H3, the moderator variable (ENV) will
be regressed on the predictor variable
(INOV) and it is expected that the
relationship will be significant. The
additional variables in the model control for firm size (SIZE) and monitoring mechanisms by creditor (LEV).

Research Methods
Sample selection
Data were collected from the
North American Compustat S&P 500
database. The sample used in this study
consists of firms in the manufacturing
industry (SIC 2000 and SIC 3000).
The manufacturing industry was selected as the research sample because
firms in this industry tend to employ
different levels of innovative efforts to
compete effectively.
Information on industry concentration ratios was obtained from the
1997 U.S. economic census from the
U.S. Department of Commerce. To be
included in the sample, a firm must
report research and development expenses, sales revenues, total assets,
total liabilities, and total equities for
five consecutive years (1997-2001).
In addition, the industry concentration

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

Table 1.Sample Selection for Compustat S&P 500 Firms in Manufacturing
Industry
Total number of firms listed on Compustat S&P 500

500

Less:
Non-manufacturing firms:
- Mineral industry (SIC 1000-1499)

20

- Construction industry (SIC 1500-1799)

4

- Transportation, communications and utilities
(SIC 4000-4999)

64

- Wholesale trade (SIC 5000-5199)

11

- Retail trade (SIC 5200-5999)

39

- Financial, insurance, and real estate industry
(SIC 6000-6799)

81

- Service industry (SIC 7000-8999)

49

- Other industry (SIC 9000-9999)

3

Sample before data restrictions

229

Incomplete data for five consecutive years
(1997-2001)

39

Unable to obtain data from proxy statements

74

Total

ratio for the industry in which the firm
belongs to (based on three-digit SIC
codes) must be available in the report
from the 1997 U.S. economic census.
The number of firms that meet these
data requirements is 116. Table 1 summarizes the sample selection.

Variables Measurement
Performance. This study utilizes
two accounting-based performance
measures as the dependent variables
since it is interested in investigating
the outcome of managerial discretionary efforts engaging in innovative ef-

(271)

(113)
116

forts. Some studies have shown that
managers prefer accounting-based performance measures that tend to be
more controllable than market-based
measures (Verrecchia 1986; Elitzur
and Yaari 1995). Therefore, this study
uses return on assets (ROA) and return
on sales (ROS) to measure firms’ performance. Some researchers argue that
ROA and ROS are the two most commonly used indicators of profitability
(Vining and Broadman 1992; Parker
and Hartley 1991). Hutchinson and
Gul (2004) contend that accountingbased performance measures such as
287

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

ROA and ROS tend to be more controllable by managers and, therefore,
are considered by many as appropriate
proxies for management performance.
ROA is measured by the ratio of income before extraordinary items to
total assets. Subsequently, ROS is
measured by the ratio of income before extraordinary items to total sales
revenue. To mitigate the influence of
some temporary unusual circumstances
occurring in a particular year, this study
uses five-year averaged values for ROA
and ROS (1997-2001).
Innovative Efforts. This study
uses research and development (R&D)
intensity as a proxy for a firm’s innovative efforts. Although there is no
guarantee that large expenditures on
R&D intensity will lead to successful
production of innovative products/services, a firm investing in R&D at much
higher rate than do its competitors is
most likely trying to compete on the
basis of innovativeness to respond to
its competitive environment. R&D intensity is measured by the ratio of
research and development expenses to
total sales revenues and is averaged
over a five-year period (1997-2001).
Competitive Intensity. This study
utilizes the industry concentration ratios, measured by Herfindahl Index, to
measure the competitive intensity. The
Herfindahl Index is calculated based
on the sum of squared of market shares
of all firms in an industry using the
following formula:
n

Hi =

288

 (marketshare )

2

i=1

i

where i refers to an individual firm in
the industry. The higher the index, the
more concentrated is the industry.
Therefore, the Herfindahl Index is
negatively associated with competitive intensity.
Control Variables. Previous studies suggest that size and leverage may
influence the performance of firms
(Frank and Goyal 2003; Ramaswamy
2001). This study, therefore, includes
size and leverage as control variables.
A logarithmic function of total assets
is employed to control for the effects
of firm size on the performance of the
firm. Leverage is measured as the ratio
of total liabilities to total equity.

Data Analysis and Results
Descriptive Statistics
Table 2 reports descriptive statistics for all sample and are partitioned
by competitive intensity. As suggested
by Sharma et al. (1981), this study uses
median split approach to classify the
sample firms into high and low competitive intensity based on the industry
concentration ratios. Although the dichotomization of continuous variables
is likely to result in loss of information, it improves the understandability
and clarity of interpreting the regression results. To examine whether the
results are sensitive to different specification, however, the regression using competitive intensity as a continuous variable is also performed. The
results are, in general, consistent with
those using the dichotomous variable.

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

Table 2. Descriptive statistics
Variables

ROA

Panel A: All sample (n=116)
Mean
8.762
S.D.
6.027
Min
-116.35
Max
42.52

ROS

INOV

SIZE

LEV

-0.002
0.557
-4.790
0.390

0.099
0.101
0.000
0.740

3.661
0.519
2.560
5.100

1.548
1.306
0.110
7.640

Panel B: Partitioned by competitive intensity
a. High Intensity (n=58)

Mean

11.604

0.079

0.124

3.735

1.735

S.D.
Min
Max

8.125
-3.580
40.150

0.059
-0.050
0.250

0.124
0.010
0.740

0.525
2.570
5.100

1.201
0.230
7.640

b. Low Intensity (n=58)
Mean
5.691
S.D.
21.192
Min
-116.35
Max
42.52

-0.090
0.797
-4.790
0.390

0.074
0.065
0.000
0.330

3.580
0.505
2.560
4.600

1.344
1.393
0.110
7.500

ROA is return on assets as measured by the ratio of net income to total assets
ROS is return on sales as measured by the ratio of net income to total net sales revenue
INOV is innovative efforts as measured by the ratio of research and development expenses to total
net sales revenue
SIZE is measured by the logarithmic function of total assets
LEV is the leverage as measured by the ratio of debt to equity

Panel A of Table 2 shows that the
average ROA and ROS for the entire
sample is 8.762 and -0.002. Panel B of
Table 2 shows that firms in high competitive intensity, on average, perform
better (have higher ROA and ROS)
than do firms in low competitive intensity (11.604 and 0.079 as compared to
5.691, and -0.090 for ROA and ROS,
respectively).
The average spending associated
with innovative efforts for the overall
firms is 10 percent of the total sales
revenues. Consistent with our predic-

tion, firms in high competitive intensity show higher innovative efforts as
compared to those in the low competitive intensity (12.4 and 7.4% respectively).

Hypotheses Testing
To investigate the moderating effects of competitive intensity on the
relationship between innovative efforts
and performance, this study employs
the three-step moderated regression
analysis procedure proposed by
Sharma et al. (1981) and uses Chow’s
289

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

(1960) approach to testing the equality
of regression coefficients between the
subgroups. Table 3 reports the regression results for ROA and ROS using
both the dichotomous [columns (2)
and (3)] and continuous [columns (4),
and (5)] specifications of competitive
environment (ENV). The F-statistics
for all regressions are statistically significant (F=4.118, p=0.002, R2=0.159

for column (2); F=3.538, p=0.005,
R2=0.139 for column (3); F=4.788,
p=0.001, R2=0.180 for column (4);
and F=4.512, p=0.001, R2=0.170 for
column (5). Since the results are consistent across both specifications, this
study discusses only the results using
the dichotomous competitive environment (ENV) measure.

Table 3.Regression Results of ROA and ROS on Independent and Control
Variables using Dichotomous and Continuous Measures of Innovative Effortsa
Variables

Prediction

(1)

Dichotomous INOV
 (t-statistics)b

Continuous INOV
 (t-statistics)

ROA
(2)

ROS
(3)

ROA
(4)

ROS
(5)

Intercept

?

0.000
1.992 **

0.000
1.689 *

0.000
1.642 *

0.000
1.369

INOV

+

0.096
0.467

0.077
0.383

0.494
1.693

0.498
1.757

ENV

-

0.086
0.664

0.122
0.936

-0.114
-1.018

-0.139
1.234

INOV*ENV

-

-0.560
-2.352 **

-0.516
-2.221 **

-0.918
-3.041 ***

-0.897
-3.066 ***

SIZE

?

-0.078
-0.839

-0.137
-1.455

-0.065
-0.706

-0.127
-1.375

LEV

?

-0.077
-0.759

-0.027
-0.263

-0.063
-0.631

-0.017
-0.173

0.159
4.118 ***
116

0.139
3.538 ***
116

0.180
4.788 ***
116

0.170
4.512 ***
116

R2
F
Sample size
a

The coefficients are based on the standardized form. INOV is innovative efforts as measured by
research and development expenses divided by net sales revenues. ENV is a dummy variable as
measured by the industry concentration ratio (1 if high intensity; 0 if low intensity). Size is the
logarithmic function of total assets. Lev is leverage as measured by the ratio of debt to equity.
b
***, **, and *, denote the significant level at 0.01, 0.05. and 0.1, respectively.
290

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

Consistent with Hypothesis H1,
the coefficients of INOV*ENV are
negative and significant (=-0.560,
p<.05 for ROA, and =-0.516, p<.05
for ROS). Furthermore, the results also
show that the coefficients of INOV are
positive but not significant, and the
sum of the coefficients of INOV and
INOV*ENV are negative. These results indicate that competitive intensity moderates the relationship between
innovative efforts and performance. In
addition, the results reveal that when
competitive intensity is high, innovative efforts have a positive relationship with performance (conditioned
on the ENV being equal to 0 for high
competitive intensity, coefficient of
INOV indicates the relationship between INOV and PERFORM in high

competitive intensity). When competitive intensity is low, however, innovative efforts have a negative relationship with performance (conditioned
on the ENV being equal to 1 for low
competitive intensity, the sum of the
coefficients of INOV, ENV, and
INOV*ENV signifies the relationship
between INOV and PEFORM in low
competitive intensity).
The moderating effects of competitive intensity on the relationship
between innovative efforts and performance can be better interpreted in
graphical representation. Figure 1
shows the fitted value of PERFORM
based on the regression estimates reported in Table 3 using the dichotomous INOV variable found in columns (2) and (3). The relationship

Figure 1. The Moderating Effect of Competitive Intensity on the Relationship
between Innovative Efforts and Firm Performance
Panel A: ROA
0.2
0.1

ROA

0
-0.1

0.2

0.4

0.6

0.8

1

-0.2
-0.3
-0.4
-0.5
Innovative Efforts
High Intensity

Low Intensity
291

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

Panel A: ROS
0.2
0.1

ROS

0
-0.1

0.2

0.4

0.6

0.8

1

-0.2
-0.3
-0.4
-0.5
Innovative Efforts
High Intensity

between INOV and PERFORM is positive when competitive intensity is high
(coefficients on INOV are 0.096 and
0.077 for ROA and ROS, respectively).
However, the relationship between
INOV and PERFORM is negative
when competitive intensity is low (the
sum of the coefficients of INOV, ENV
and INOV*ENV are -0.378 (0.096 +
0.086-0.560) for ROA and -0.317
(0.077 + 0.122-0.516) for ROS.
The significant interaction between INOV and performance reported
in Table 3 suggests that ENV is either
a quasi-moderator or a pure-moderator of the relationship between innovative efforts and performance. To ex-

Low Intensity

amine whether competitive intensity
acts as a quasi-moderator of the relationship between innovative efforts and
performance (test of hypothesis H2 and
hypothesis H3), this study employs the
three-step moderated regression analysis procedures proposed by Sharma et
al. (1981). The first step is to investigate whether the moderator variable
(ENV) is related to the criterion variable (PERFORM). The two criterion
variables (ROA and ROS) are regressed
on the competitive intensity variables
(two dummy variables representing
high and low competitive intensity).1
Table 4 reports the results of this analysis. The results show that competitive

1
As suggested by Prescott (1986), this study validates the results by regressing ROA and ROS
on the continuous measures of competitive intensity (ENV). The results are consistent with those
of the dichotomous measures of competitive intensity (Adjusted R2= 0.018, F= 0.747, p= 0.526 for
ROA and Adjusted R2= 0.027, F=1.168, p=0.325).

292

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

Table 4. Regression Results of ROA and ROS on Competitive Intensity
Variable (Dichotomous and Continuous) and Control Variables
SS

DF

MS

F

P

R2

ROA
Regression
Residual
Total

3,377.04
31,752.10
35,129.14

3
113
116

1,125.68
280.99

4.502

0.036

0.034

ROS
Regression
Residual
Total

2.79
39.03
41.82

3
113
116

0.93
0.34

3.050

0.083

0.023

ROA
Regression
Residual
Total

1588.29
32,348.35
33,936.64

3
113
116

529.43
286.27

3.079

0.072

0.016

ROS
Regression
Residual

1.77
39.37

3
113

0.59
0.35

3.168

0.058

0.015

Total

41.14

116

Criterion Variable
Dichotomous Variable

Continuous Variable

ROA is return on assets as measured by the ratio of net income to total assets
ROS is return on sales as measured by the ratio of net income to total net sales revenues

intensity has a significant effect on
performance (R2 =0.034, F= 4.502,
p=0.036 for ROA and R2 =0.023,
F=3.050, p=0.083 for ROS).2 The significant effect of ENV on PERFORM

indicates that competitive intensity acta
as a quasi-moderator of the relationship between innovative efforts and
performance.

2
This being the case, the competitive intensity acts as a quasi-moderator of the relationship
between innovative effort and performance (see Sharma et al. (1981) for excellent discussion on this
topic).

293

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

Table 5. The Chow Test for equality of the regression pairs based on the
competitive intensity variable (high vs. low) and control variables
Criterion Variable

SS

DF

MS

F

p

ROA
Regression
Residual
Total

489.31
3,384.56
3,873.87

3
55
58

163.10
61.54

2.602

0.061

ROS
Regression
Residual
Total

0.07
0.14
0.208

3
55
58

0.02
0.003

8.417

0.000

ROA
Regression
Residual
Total

5,279.64
20,281.96
25,561.59

3
55
58

1,759.88
368.76

4.599

0.006

ROS
Regression
Residual
Total

2.58
13.73
16.31

3
55
58

0.86
0.25

3.315

0.027

ROA
Regression
Residual
Total

3,064.06
27,459.75
30,523.81

3
113
116

1,021.35
243.01

4.129

0.008

ROS
Regression
Residual
Total

1.57
15.26
16.83

3
113
116

0.52
0.135

3.836

0.012

High intensity

Low intensity

Overall Sample

ROA is return on assets as measured by the ratio of net income to total assets
ROS is return on sales as measured by the ratio of net income to total net sales revenues

294

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...

The second step is to test the equality of the regression pairs (high intensity vs. low intensity) using the procedure suggested by Chow (1960).3 This
study uses median split to divide the
sample into subgroups on the basis of
the competitive intensity variable. The
performance variables are then regressed on the innovative efforts variable for each subgroup as well as for
the overall sample. Table 5 reports the
results of this procedure. The results
indicate that the coefficient of regression for high intensity subgroup differs significantly from the coefficient
of regression for low intensity sub
group [F (3,110)= 5.216, p<.05 for
ROA, and F(3,110)=3.65, p<.05 for
ROS].4 These results further substantiate the hypothesized quasi-moderating effect of environment on the relationship between innovative efforts and
performance.
The third step is to correlate INOV
with ENV. As predicted, competitive
intensity is positively related to innovative efforts (r=.247, p=.008), indicating that firms are more willing to
spend on innovative efforts as competitive intensity increases.

Discussion, Limitations, and
Implications for Future
Research
Overall, the results of this study
are of consistence with the hypotheses. First, this study finds that the
relationship between innovative efforts
and performance is moderated by competitive intensity. The results indicate
that innovative efforts have a positive
effect on performance when competitive intensity is high. However, when
competitive intensity is low, innovative efforts have a negative impact on
performance. These findings are consistent with Porter’s (1985) argument
that when competitive intensity is high,
firms should invest heavily in research
and development activities to promote
innovativeness. These findings are also
of consistence with the argument that
firms operate in low competitive intensity should adopt a cost leadership
strategy through aggressive construction of efficient scale facilities, vigorous pursuit of cost reduction, tight cost
control, and cost minimization in areas such as research and development,
services, sales force and advertising

3

The Chow Test uses the following formula:
(SSER - SSEU) / K
to calculate the F-value with K and (T1 + T2-K) for the numerator and deSSER / (T1 + T2 - 2K)
and denominator degree of freedom, respectively. SSE R is the sum of squared residual for the
overall sample, SSE U is the sum of the two sums of squared residual from the two subgroups, K
is the number of independent variables, T1 and T2 are the number of sample in each subgroup.
4
Since the observed F-value of 5.216 and 3.650 are bigger than the critical value of 2.70 (degree
of freedoms of 3 and 110 for the numerator and the denominator, respectively), the coefficients of
regressions in the two subgroups differ significantly.

295

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3

(Porter 1985; Hambrick 1983). These
findings suggest that innovative efforts should be evaluated in relation to
the types of environment in which the
firm operates since good fit between
these two variables is likely to affect
performance positively.
Second, the results indicate that
competitive intensity acts as a quasimoderator of the relationship between
innovative efforts and performance.
As such, not only does competitive
intensity affect the relationship between innovative efforts and performance, but it also acts as a predictor
for the performance variable (Sharma
et al. 1981). The negative relationship
between competitive intensity and performance is consistent with previous
findings investigating the independent,
direct effects of environment on performance (Keats and Hitt 1988; Dess
and Beard 1984)
Third, firms are inclined to align
their innovative efforts with the level
of competitive intensity in comprehensive, intuitively meaningful, and
often predictable pattern (Zahra and
Covin 1993). Firms operating in high

competitive intensity spend more on
innovative efforts than do those operating in low competitive intensity.
These findings suggest that as the competitive intensity increases, firms tend
to pursue innovative strategy to gain
competitive advantage by intensifying
their investments in research and development activities.
The results of this study, however, should be interpreted in light of
two limitations. First, this study utilizes single industry data to test our
hypotheses. Although the most interesting issue regarding the relationship
between innovative efforts and performance is to investigate the variability
within an industry, the external validity of the results will be enhanced if
studies including other industries find
similar results. Second, this study uses
Herfindal index as a proxy for competitive environment. Other measures
such as earning volatility, variation in
sales or executive perceptions of the
environment in which the firms operate may be harnessed to validate the
results of this study.

References
Argyris, C. 1999. On Organizational Learning (2nd ed.). Oxford: Blackwell Business.
Baker, W. E., and J. M. Sinkula. 1999. The synergistic effect of market orientation and
learning orientation on organizational performance. Journal of Academic Marketing
Science 27: 411-427.
Burgeois, III, L. J. 1985. Strategic goals, perceived uncertainty, and economic performance in volatile environments. Academy of Management Journal 28: 548-573.

296

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...
Burton, R. M., J. Lauridsen, and B. Obel. 2002. Return on assets loss from situational and
contingency misfits. Management Science 48 (11): 1461-1485.
Calantone, R. J., S. T. Cavusgil, and Y. Zhao. 2002. Learning orientation, firm innovation
capability, and firm performance. Industrial Marketing Management 31: 515-524.
Chandler, A. D. Jr. 1962. Strategy and Structure: Chapters in the History of the Industrial
Enterprise. Cambridge, Mass: MIT Press.
Chow, G. C. 1960. Test of equality between sets of coefficients in two linear regressions.
Econometrica 28: 591-605.
Cooper, R. G. 2000. New product performance: What distinguishes the star products.
Australian Journal of Management 25: 17-45.
D’Aveni, R. A. 1994. Hyper-competition: Managing the Dynamics of Strategic Maneuvering. New York: Free Press.
Dess G. G., and D. W. Beard. 1984. Dimensions of organizational task environments.
Administrative Science Quarterly 29: 52-73.
Douglas, S. P., and D. K. Rhee. 1989. Examining generic competitive strategy types in U.S.
and European markets. Journal of International Business Studies (Fall): 437-463.
Elitzur, R. R., and V. Yaari. 1995. Executive incentive compensation and earnings
manipulation in a multi-period setting. Journal of Economic Behavior and Organization (26): 201-219.
Foster, R. N. 1986. Innovation: The Attacker’s Advantage. New York: Summit Books.
Frank, M. Z., and V. K. Goyal. 2003. Testing the pecking order theory of capital structure.
Journal of Financial Economics 67: 217-248.
Garrat, B. 2000. The Learning Organization: Developing Democracy at Work. London:
HarperCollins Business.
Gerdin, J., and J. Greve. 2004. Forms of contingency fit in management accounting
research—A critical review. Accounting, Organizations, and Society 29: 303-326.
Golder, P. N., and G. J. Tellis. 1993. Pioneering advantage: Marketing logic or marketing
legend? Journal of Marketing Research 30: 158-170.
Hambrick, D. C. 1983. High profit strategies in mature capital goods industries: A
contingency approach. Academy of Management Journal 26: 687-707.
Hutchinson, M., and F.A. Gul. 2004. Investment opportunity set, corporate governance
practices and firm performance. Journal of Corporate Finance 10: 595-614.
Jermias, J., and H. M. Armitage. 2000. Management accounting in Indonesia: Analysis of
current system, potential for change and forces behind Innovation. The International
Journal of Accounting and Business Society 8: 36-57.
Jermias, J., and L. Gani. 2004. Integrating business strategy, organizational configurations
and management accounting systems with business unit effectiveness: A fitness
landscape approach. Management Accounting Research 15: 179-200.
Jin, Z., N. Hewitt-Dundas, and N. J. Thompson. 2004. Innovativeness and performance:
Evidence from manufacturing sectors. Journal of Strategic Marketing 12: 255-266.

297

Gadjah Mada International Journal of Business, September - December 2006, Vol. 8, No. 3
Keats, B. W., and M. A. Hitt. 1988. A causal model of linkages among environmental
dimensions, macro organizational characteristics, and performance. Academy of
Management Journal 31 (3): 570-598.
Li, T., and R. J. Calantone. 1998. The impact of market knowledge competence on new
product advantage: Conceptualization and empirical examination. Journal of Marketing 62: 13-29.
Lynn, T. A. 1994. Learning from competition. Journal of Accountancy (February): 43-46.
Mansfield, E. 1961. Technical change and the rate of imitation. Econometrica 29: 741766.
Manu, F. A. 1992. Innovation orientation, environment and performance: A comparison
of U.S. and European markets. Journal of International Business Studies 23 (2): 333352.
March, J. G. 1991. Exploration and exploitation in organizational learning. Organization
Science 2 (1): 71-87.
Mia, L., and B. Clark. 1999. Market competition, management accounting systems and
business unit performance. Management Accounting Research 10: 137-158.
Miles, R. E., and C. C. Snow. 1978. Organizational Strategy: Structure and Process. New
York: McGraw-Hill.
Miller, A. 1988. A Taxonomy of technological settings with related strategies and
performance levels. Strategic Management Journal 9: 239-254.
Miller, D. 1987. The structural and environmental correlates of business strategy.
Strategic Management Journal 8: 55-76.
Milliken, F. J. 1987. Three types of perceived uncertainty about the environment: state,
effects, and response uncertainty. Academy of Management Review 12: 133-143.
Mone, M. A., W. McKinley, and V. L. Barker. 1998. Organizational decline and
innovation: A contingency framework. Academy of Management Review 23 (1): 115132.
O’Brien, J. P. 2003. The financial leverage implications of pursuing a strategy of
innovation. Strategic Management Journal 24: 415-431.
Parker, D., and K. Hartley. 1991. Do changes in organizational status affect financial
performance? Strategic Management Journal 12: 631-641.
Porter, M. E. 1985. Competitive Advantage. New York: Free Press.
Prescott, J. E. 1986. Environments as moderators of the relationship between strategy and
performance. Academy of Management Journal 29: 329-346.
Ramaswamy, K. 2001. Organizational ownership, competitive intensity, and firm performance: An empirical study of the Indian manufacturing sector. Strategic Management Journal 22: 989-998.
Rolfe, A. J. 1992. Profiability reporting techniques bridge information gap. The Journal
of Business Strategy: 32-37.

298

Jermias—Competitive Intensity as a Quasi-moderator of the Relationship between ...
Schneeweis, T. 1983. Determinants of profitability: An international perspective. Management International Review 23 (2): 15-21.
Sharma, S., R. M. Durand, and O. Gur-Arie. 1981. Identification and analysis of moderator
variables. Journal of Marketing Research XVIII: 291-300.
Tushman, M. L., and P. Anderson. 1986. Technological discontinuities and organizational
environments. Administrative Science Quarterly 31: 439-465.
Verrecchia, R. E. 1986. Managerial discretion in the choice among financial reporting
alternatives. Journal of Accounting and Economics 9: 175-196.
Vinning, A., and A, Broadman. 1992. Ownership versus competition: Efficiency in public
enterprise. Public Choice 73: 205-239.
Zahra, S., and J. Covin. 1993. Domestic and international competitive focus, technology
strategy and firm Performance: An empirical analysis. Technology Analysis and
Strategic Management 6 (1): 39-54.

299