See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/227417211

Managing country-of-origin choices: Competitive advantages and opportunities

Article  in  International Business Review · December 2004

DOI: 10.2139/ssrn.1554671 · Source: RePEc

CITATIONS READS

36 131

3 authors:

Glen H Brodowsky Justin Tan

California State University, San Marcos York University
18 PUBLICATIONS   238 CITATIONS    83 PUBLICATIONS   4,265 CITATIONS   

SEE PROFILE SEE PROFILE

Ofer Meilich
California State University, San Marcos
8 PUBLICATIONS   149 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Digitization of News View project

All content following this page was uploaded by Glen H Brodowsky on 18 October 2017.

The user has requested enhancement of the downloaded file.

 International Business Review 13 (2004) 729–748
www.elsevier.com/locate/ibusrev

Managing country-of-origin choices: competitive

advantages and opportunities
Glen H. Brodowskya,1, Justin Tanb, Ofer Meilicha,*
a
College of Business Administration, California State University San Marcos, San Marcos, CA 92096-0001, USA
b
Ivey School, University of Western Ontario and Department of Management, Eppley College Of Business
Administration, Creighton University, Omaha, NE 68178, USA

Received 22 April 2002; revised 6 May 2004; accepted 30 September 2004

Abstract
This study investigates the effects of country-of-design/country-of-assembly combinations on
consumers’ evaluative beliefs about and attitudes toward buying automobiles. The effects are
compared across groups of consumers differing with respect to levels of consumer ethnocentrism.
Two design countries and two assembly countries (Japan or US) were considered, yielding four
possible design-country/assembly-country combinations. The results suggest that manufacturing
products in the country in which they are sold not only provides closer access to the market, but also
allows multinational manufacturers to ‘blur the boundaries’ regarding a potentially sensitive
country-of-origin issue among highly ethnocentric consumers. At the same time, they can leverage
their country-brand images to appeal to those customers who recognize a particular country’s ability
to design high quality cars, regardless of their country of assembly.
q 2004 Elsevier Ltd. All rights reserved.

Keywords: Consumer ethnocentrism; Country effects

1. Introduction

Controversial country-of-origin-related management decisions are once again making

headlines. In an atmosphere of renewed patriotism in the wake of the events of the recent
past, and on the eve of the 2004 US presidential election, Americans are once again talking

* Corresponding author. Tel.: C1 760 750 4235; fax: C1 760 750 4250.
E-mail addresses: glenbrod@mailhost1.csusm.edu (G.H. Brodowsky), jtan@creighton.edu (J. Tan),
meilich@csusm.edu (O. Meilich).
1
Tel.: C1 760 750 4261; fax: C1 760 750 4250.

0969-5931/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ibusrev.2004.09.007
730 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

about where products are made. Over the past several decades, much attention has focused
on importing raw materials or outsourcing the manufacturing-related links of the value
chain. Many blue-collar jobs migrated from the United States to more cost effective labor
markets in Asia and Latin America. Now, Americans are faced with outsourcing other
value chain links, namely research and development and customer service. According to
an often-quoted Forrester Research report, by the year 2015, more than 3 million white
collar service sector and high technology jobs will leave the US (McCarthy, Dash, Liddell,
Ross-Ferrusi, & Tempkin, 2002).
Much research has focused on the effects of country of origin. However, consumers,
managers, and scholars have wrestled with defining just what country of origin (COO)
means. In the past, country of origin was a pre-determined product characteristic included
on a label reflecting the country from which a product had been imported. Most marketing
studies have focused on assessing how consumers use such country information cues for
product quality (Chao, 1992; Gaedeke, 1973; Heslop, Liefeld, & Wall, 1987; Johansson,
Douglas, & Nonaka, 1985; Nagashima, 1970, 1977; Reirson, 1967; Schooler, 1965, 1971;
Schooler & Wildt, 1968; Wall, Liefeld, & Heslop, 1991).
The emergence of global markets and global companies has complicated the country-
of-origin phenomenon. For global managers, country of origin has now become a
managerial decision variable. They may choose to design products in one country and
manufacture and assemble them in another country using raw materials or components
from a variety locations around the globe. Such decisions might be based on cost
considerations or on proximity to end user markets. Managers may also consider whether
or not country-of-origin choices can be used in other ways to gain competitive advantage.
For example, while some companies are looking to gain cost advantages by setting up
customer call centers in India, others may choose to emphasize their decisions to keep such
call centers in the US to attract disgruntled American consumers.
The product country image (PCI) literature, as it has come to be known, focuses on the
effects country-related information has on consumer perceptions. Nagashima (1970, p 68)
defined the image consumers associate with a given country-of-origin as “the picture, the
reputation, the stereotypes that businesses and consumers attach to products of a specific
country.” Country-related information or cues may be manifested in a variety of forms. The
simplest is the ‘made in’ label. Other forms include the explicit inclusion of country
information in brand names (American Airlines), or in the implicit use of colors used in
packaging or labeling (IKEA’s distinctive blue and yellow color scheme that evokes the
image of the Swedish flag).
Researchers have studied country-of-origin effects, or product country image, from a
number of vantage points. Identifying a product’s country of origin has become more
elusive as more and more steps of the value chain are performed in different countries.
Country of origin has become a moving but not irrelevant target. This presents challenges
and opportunities for managers. Where companies choose to design, make, or service their
products remains a concern to some consumers in some places.
Global brands and products are less easily identified with particular countries of origin.
For example, while Toyota is generally recognized as a Japanese company, many of the
vehicles it sells in the United States are manufactured in California. In fact, some of its
advertising campaigns stress that the popular Toyota Camry is the ‘best car built in
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 731

America’. In doing so, Toyota seems to be leveraging its strong positioning for Japanese
engineering while, at the same time, overcoming some consumers’ aversion to buying
imported automobiles rather than domestically manufactured ones. Cars such as the
Camry have more than one that country-of-origin. It has a country of design and
engineering (CDE) and a country of manufacturing and assembly (CMA). With a market-
leading reputation for excellence in product design and performance, however, the
question of why Toyota would not only choose to manufacture overseas, but emphasize
that choice in its marketing communications, becomes important.
According to traditional PCI research that focused on the use of country cues for
product evaluation, the explanation might be that American consumers consider
American-made products to be superior to Japanese-made ones. Objective studies of
product quality, such as that performed and published by Consumer Reports during the
1980s and 1990s, however, consistently showed Japanese cars outperforming American
cars. An alternative explanation might be that some consumers may use country-of-origin
information for reasons not related to product quality. By the late 1980s, PCI research
began to address whether some consumers, who are patriotic or ethnocentric, consider the
morality of buying products from other countries as well as, or instead of, the quality of
domestic versus imported products when making purchasing decisions.
This study explores how country information influences consumers’ perceptions of and
attitudes toward buying foreign versus domestic products. Specifically, this study
addresses whether consumers who differ in their levels of ethnocentric tendency differ in
how they use country cues. Country effects are split into two cues-country of design and
engineering (CDE) and country of manufacturing and assembly (CMA). By understanding
the dynamics of how different consumers respond to country-related information,
managers can make more informed choices about the risks and benefits of locating various
stages of the value chain in different parts of the world.

1.1. Literature review

Consumer research has shown that individuals base their purchasing decisions on
information cues (Samiee, 1994). The importance of the country-of-origin effect lies in its
potential use by consumers as an extrinsic cue in making purchasing decisions. The most
serious consideration of this phenomenon is in situations where consumers reject a product
outright solely on the basis of its country of origin. Researchers such as Johansson (1993)
have noted an overemphasis in PCI research on studying country effects on product
evaluation or product image. He argued that such a focus lacks managerial relevance
because it fails to address how, beyond product evaluation, country of origin affects actual
consumer purchase behavior. Others have criticized the literature for its paucity of theory-
based explanations of what country of origin is, what it affects, and how consumers use
country cues (Obermiller & Spangenberg, 1989).
Some researchers have considered country effects on outcomes other than product
evaluation or product image. For example, some have investigated whether country-of-
origin (COO) affects consumers’ perceptions regarding the riskiness of purchasing foreign
versus domestic products (Wall et al., 1991). Others have measured COO effects on
purchase intention (Han, 1988; Schooler & Wildt, 1968; Shimp & Sharma, 1987).
732 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

Researchers who tested for country effects on multiple outcome variables (such as
purchase intention, product evaluation, and attitude toward buying) tested for the effects
on each dependent variable separately. They have not considered the relationships among
attitude toward buying, product evaluation, and purchase intention that may be suggested
by a belief–attitude–action framework (Fishbein & Ajzen, 1975). The current study tests
for these effects within such a theoretical framework.
Several researchers have addressed how multinational production has made the issue of
COO more complex (Brodowsky, 1998; Chao, 1992; Ettenson, 1993; Han & Terpstra,
1988; Johansson & Nebenzahl, 1986; Maronick, 1995; Samiee, 1994; Tse & Gorn, 1993).
While multinational production has made identifying a product’s country of origin less
clear to consumers, at the same time it has given managers more control over choosing a
product’s country of origin. Studying the relative effects of country of design and country
of assembly on consumer attitudes and behavior should help managers assess the
consequences of decisions concerning where to design or manufacture their products.
Another area of research investigates whether consumers’ personal characteristics
influence how they respond to country-of-origin information. Researchers have identified
consumer characteristics such as ethnocentrism (Shimp & Sharma, 1987) and consumer
patriotism (Han, 1988) that may explain why certain consumers are more likely to be
interested in knowing a product’s country of origin. Shimp and Sharma (1987) found that
ethnocentric consumers were more likely to rate foreign products negatively and less
willing to purchase imports. They developed the 17-item consumer ethnocentric tendency
scale (CETSCALE) to measure the propensity of consumers to consider the morality of
buying domestic instead of foreign products. Han (1988) found that patriotic consumers
exhibited a negative bias towards buying foreign products. However, contrary to the
findings of Shimp and Sharma (1987), he found little evidence of negative biases in
consumers’ evaluations of foreign products. Han (1988) found little evidence to support
his hypothesis that country of origin affects purchase intention through product evaluation.
Han (1988) found some evidence of stronger consumer patriotism among Caucasians,
females, older people, and blue collar workers. Herche (1992) noted that the CETSCALE
is a stronger predictor of import purchase behavior than are demographic variables.
However, while many researchers have shown that consumer ethnocentric tendency has
significant predictive effects on consumer evaluations of products and import purchase
decisions, it remains difficult to clearly profile the ethnocentric segments of the market.
Luthy and Parsa (1998) found few correlates between CETSCALE scores and
demographic variables. Nielsen and Spence (1997) found some relationship between
military service and ethnocentrism. In general, they found military personnel to exhibit
higher CETSCALE scores than non-military personnel. However, they observed rises in
non-military ethnocentrism in times of increased nationalism in the US. In a study of both
Russian and American consumers, Imbert, Jiddou, Kumar, Murillo, and Zhao (2003)
showed that younger American consumers under the age of 25, and older consumers over
the age of 45 tended to score higher on the CETSCALE than those between the ages of 25
and 45. They attributed these patterns to generational differences.
Studies of the relative effects of multiple countries of origin (countries of design and
assembly) and correlates of country effects such as patriotism and ethnocentrism are both
important areas in the literature. However, there have been few studies that have combined
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 733

these two streams of the literature. Brodowsky (1998) studied the relative impacts of
country of design and country of assembly on product beliefs and consumer attitudes
compared them across consumers differing in their levels of ethnocentric tendency. The
author found that highly ethnocentric consumers had more positive evaluative beliefs and
attitudes toward buying domestically designed or manufactured automobiles. In contrast,
he found that consumers with low levels of ethnocentric tendency expressed more positive
evaluative beliefs about and attitudes toward buying Japanese-designed or assembled
automobiles. Furthermore, the results suggested that low ethnocentric consumers are more
likely to use country cues as objective information about product quality, as has been
assumed by earlier research studies. In contrast to their low ethnocentric counterparts,
highly ethnocentric consumers appeared more likely to use country cues when forming
attitudes toward buying products. This suggests that highly ethnocentric consumers use
country cues for making patriotically based purchase decisions. Overall, the results
showed that country-of-origin information is used differently by consumers who differ in
ethnocentric tendency. The current study builds on these findings.
Since an intertwined transnational network of exchange spanning all links of the value
chain results in final products with multiple countries of origin, country cues must be
decomposed into different components (Samiee, 1994). Acharya and Elliott (2001)
examined country-of-assembly and country-of-design effects for three countries and three
product categories. These authors found that the country-of-assembly was the most
important influence on product quality evaluation and choice. Country-of-design effects
were less pronounced, they found. While this is consistent with Brodowsky (1998), these
authors did not examine these differences in light of consumer ethnocentrism.
This study incorporates the following effects: (1) country of design and engineering
(CDE), which denotes the location of design and engineering; (2) country of manufacture
and assembly (CMA), which denotes the location of final point of manufacture; and
(3) consumer ethnocentric tendencies (CETSCALE). Both country cues are incorporated
into the model to provide a more real-world definition of country of origin to compare
country effects across groups of consumers that differ in their levels of ethnocentric
tendency. The study attempts to explain halo effects in consumers’ evaluative beliefs about
domestic products and negative stereotypes in their evaluative beliefs about imports in
terms of consumer ethnocentrism. The possibility that such biases in evaluative beliefs carry
over into consumer attitudes toward buying foreign or domestic products is also addressed.

1.2. Research hypotheses

Combining multiple country cues with consumer ethnocentric tendency, this study
proposes several research hypotheses related to evaluative beliefs about and attitude
toward buying automobiles. The two country cues included are country of design and
engineering (CDE), and country of manufacturing and assembly (CMA). Brodowsky
(1998) demonstrated that the main effects of CDE and CMA on evaluative beliefs and
attitude toward purchase were significant for both high and low ethnocentric groups.
However, the country biases toward or against domestic or imported products were
opposite for the two ethnocentric groups. Highly ethnocentric consumers’ evaluative
beliefs and attitudes toward buying were consistently biased in favor of American design
734 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

and American manufacturing. In contrast, low ethnocentric consumers’ evaluative beliefs

and attitudes toward buying were significantly biased in favor of both Japanese design and
Japanese manufacturing.
In the current study, attention is directed toward comparisons in evaluative beliefs and
attitudes toward buying specific design/assembly–country combinations across low,
moderate, and high ethnocentric consumers. There are two categories of CDE/CMA
combinations considered. Uni-national cars are those for which the country of design and
engineering is the same as the country of manufacturing and assembly. These cars would
be described either as either purely American or purely Japanese. Bi-national cars are
those whose country of design and engineering and country of manufacturing and
assembly differ. These binational cars are described hereafter as either American-
designed/Japanese-made cars, or Japanese-designed/American-made cars.
The first set of hypotheses considers the impact of both CDE and CME biases for
highly ethnocentric consumers. Shimp and Sharma (1987) defined highly ethnocentric
consumers as those who are more concerned about the righteousness and morality of
buying domestic products than imports. Brodowsky (1998) found that highly
ethnocentric consumers have more positive evaluative beliefs about and more positive
attitudes toward buying cars designed and engineered in the US than those designed
and engineered in Japan. These consumers also exhibited more positive evaluative
beliefs about and more positive attitudes toward buying cars manufactured in the US
than those manufactured and assembled in Japan. Thus, the following hypotheses are
proposed for examining biases in favor purely American uni-national cars among
highly ethnocentric consumers:
H1a. Highly ethnocentric American consumers will express more positive evaluative
beliefs about purely American cars than purely Japanese ones.
H1b. Highly ethnocentric American consumers will express more positive attitudes toward
buying purely American cars than purely Japanese ones.
The way Shimp and Sharma (1987) defined consumer ethnocentric tendency offers
some clues about how highly ethnocentric consumers might react to bi-national cars—
those in which the CME and CDA differ. These authors suggested that individuals whose
livelihoods are most likely to be threatened by competition from imports would exhibit
higher levels of ethnocentricity. The threat to highly ethnocentric consumers would most
likely come through job loss due to overseas manufacturing of products. Thus, highly
ethnocentric consumers would more likely be concerned about buying products
manufactured and assembled in the US than buying products designed and engineered
in the US. Indeed, the CETSCALE specifically measures consumer resentment toward
buying products that ‘put American workers out of work’ and ‘allow foreign countries to
get rich off us’. Thus, we propose that given the choice between American-designed/
American-made cars and American-designed/Japanese-made cars, highly ethnocentric
consumers would be more positively disposed toward the former than the latter. Formally:
H2a. Highly ethnocentric American consumers will express more positive evaluative
beliefs about American-designed/American-made cars than American-designed/
Japanese-made cars.
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 735

H2b. Highly ethnocentric American consumers will express more positive attitudes toward
buying American designed/American-made cars than American-designed/Japanese-
made cars.
If, indeed, highly ethnocentric consumers are more concerned about the threats of
foreign products to American labor, then they might be more positively disposed toward
Japanese designed cars if they are manufactured and assembled in the US rather than
Japan. Such a finding might lend support to the strategies of many Japanese automakers
that have chosen to manufacture and assemble their cars in American factories. Thus:
H3a. Highly ethnocentric American consumers will express more positive evaluative
beliefs about Japanese-designed/American-made cars than Japanese-designed/Japa-
nese-made cars.
H3b. Highly ethnocentric American consumers will express more positive attitudes toward
buying Japanese-designed/American-made cars than Japanese-designed/Japanese-
made cars.
In contrast, low ethnocentric consumers are not likely to be particularly concerned with
the morality or righteousness of buying domestic rather than imported cars. Brodowsky
(1998) found that low ethnocentric consumers exhibited no positive biases in their beliefs
about or attitudes toward buying domestic versus imported cars. However, the author
found systematic biases in favor of Japanese cars over American ones among low
ethnocentric consumers. A possible explanation is that low ethnocentric consumers use
country cues as indicators of product quality. Objective information about automobile
performance provided by non-advocate media sources such as Consumer Reports Auto
Buying Guide sheds some light on this issue. For instance, the 1995 report included ratings
of mid-sized cars in the same price range as the Performa, the hypothetical car considered
in this study. Six of the top-10 rated mid-sized cars carried Japanese brand names, while
only one American car was included in this group. Among the 10 lowest-rated cars, eight
carried American brand names. Thus, comparing purely Japanese cars to purely American
cars, it is expected that low ethnocentric consumers would favor the Japanese. Therefore,
the following hypotheses are proposed for examining biases in favor of purely Japanese
uni-national cars among low ethnocentric consumers:
H4a. Low ethnocentric consumers will express more positive evaluative beliefs about
purely Japanese cars than purely American ones.
H4b. Low ethnocentric consumers will express more positive attitudes toward buying
purely Japanese cars than purely American ones.

2. Methodology

2.1. Research design

The automobile product category was chosen for this study of design-country and
assembly-country effects. Competition between American and Japanese automakers has
736 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

been the center of much controversy in recent years. Many Americans have strong
opinions about differences in the quality of automobiles from each country. Some resent
the strong position of Japanese automakers in the US market. Furthermore, it has become
increasingly common for Japanese automakers to design and engineer cars that they
manufacture and assemble in American factories. Similarly, while most cars designed by
American automakers are manufactured and assembled domestically, none have been
manufactured and assembled in Japan.
Four separate survey instruments, one for each of the treatment groups, were
developed. The survey instruments corresponded to the four design-country and
assembly-country combinations. In order to test the effects of multiple country cues,
both CDE and CMA were each allowed to take one of two values (Japan or the US).
Including only two values for each country cue yields four possible CDE/CMA
combinations.
Subjects were randomly assigned to one of four treatment groups. A between-
subjects design was chosen to minimize the possibility of subjects discovering the
CDE and CMA experimental manipulations. Members of each treatment group were
presented with the description of a hypothetical four-cylinder, air-conditioned compact
automobile called the Performa. They were told that the car cost $16,000 and to
assume that they had enough savings to buy the car in cash. The only differences
between treatment groups were the CDE and CMA information associated with each
car. Members of the first two groups were asked to express evaluative beliefs about
and attitudes toward buying uni-national cars. The first group considered a Performa
whose CDE and CMA were both Japan. The second group considered a Performa
whose CDE and CMA were both the United States. The third and fourth groups were
asked to express their evaluative beliefs and attitudes toward buying bi-national cars.
The third group considered a Performa whose CDE was Japan and CMA was the US.
The fourth group considered a Performa whose CDE was the United States and whose
CMA was Japan.
At the time these data were collected, Japanese brands such as Toyota, Nissan, and
Honda had already achieved significant shares of the US automobile market. Each had
strong brand equity in the minds of consumers. Publications such as Consumer Reports
had rated these brands highly. The focus of this research was on the power of country of
design and engineering and country of manufacturing and assembly, rather than on brand
equity. Therefore, it was decided to downplay the role of actual brands and emphasize
the ‘brand equities’ of countries as places of design, engineering, manufacturing or
assembly. This was the primary motivation behind choosing a fictitious car, rather than
an actual car for this study. Such a choice was also useful because, while some Japanese
brands such as Toyota were deliberately positioning their cars as ‘made in America’, no
American-branded cars were positioning themselves as ‘made in Japan’. Also, at about
the time of the study, IBM had introduced a Personal Computer model called the
Performa. Thus, this familiar-sounding name borrowed from a different product category
was used. It is interesting to note that not one respondent mentioned that Performa was a
computer brand name. However, several respondents made comments such as ‘I know I
have seen the Performa on the road. Who makes it?’
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 737

Table 1
Research design

Japanese manufactured US manufactured Total

Japanese designed Japan–Japan, NZ105 (27%) Japan–US, NZ102 (26%) 207 (53%)
US designed US–Japan, NZ92 (23%) US–US, NZ95 (24%) 187 (47%)
Total 197 (50%) 197 (50%) 394 (100%)

2.2. Data collection

A mailing list of 2000 households in New York State was purchased from Polk
Research. Polk was given a breakdown of zipcodes and the number of households desired
at various income levels within each zipcode to obtain a diverse sample. The survey
instrument was mailed in envelopes addressed to male heads of households in 1995.
Of the 2000 surveys mailed, 422 completed surveys were returned representing a 21%
response rate. Of these respondents, 394 provided data on all of the variables used in this
study. Approximately equal numbers of each of the four versions of the survey instrument
were returned. Table 1 lists the number of respondents in each of the four treatment
groups.
The sample was predominantly male (81%). Females accounted for approximately
11.8% of the survey while the remaining 7.1% did not provide a response to this question.
This heavy male-representation of the sample may be attributed to the fact that the surveys
were addressed to male heads of households. Fifty percent of the respondents were 44
years of age or younger, while 46.1% of the respondents were over 45 years of age. The
remaining 8.5% did not report their ages. The majority of the sample may be classified as
middle income, with 62% of the respondents reporting annual household incomes ranging
from $20,000 to $60,000. Less than 4% reported incomes of less than $20,000 while
15.9% reported household incomes greater than $75,000. Fourteen percent of the
respondents did not report their household incomes. Ninety-eight percent of respondents
indicated that they had completed a high school education and 45.2% of the sample had
completed four or more years of college.

2.3. Measurement scales

2.3.1. Evaluative beliefs

A major objective of this study was to incorporate and combine major streams from the
PCI literature. Therefore, it was important to include traditional measures of country-of-
origin-based product evaluations. To maintain consistency with earlier research, the same
13 attributes rated by respondents in an earlier study of country effects on automobile
evaluations (Johansson & Nebenzahl, 1986) were used. These 13 items were measured on
7-point semantic-differential scales. The anchors for these 13 semantic-differential items
are summarized in Table 2.
Each participant was asked to rate one automobile, the Performa, on all 13 items. The
car was described in terms of both country of engineering and design country of
manufacturing and assembly. Each respondent’s overall evaluation of the car was
738 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

Table 2
13-Item scale for evaluative beliefs

1 Unreliable/reliable
2 Unreasonably priced/reasonably priced
3 Common/exclusive
4 Not-so-careful workmanship/careful workmanship
5 Imitative/innovative
6 No pride of ownership/pride of ownership
7 Conventional/stylish
8 For older people/for younger people
9 Costly to run/economical to run
10 Low quality/high quality
11 Low performance/high performance
12 Not durable/Durable
13 High service cost/low service cost

computed by summing responses to the 13 items in the scale. In subsequent discussion,

this overall score is referred to as EVALUATION. The mean value of EVALUATION for
the entire sample (NZ394) was 53.15. The 13-item scale proved to be highly reliable, with
a Cronbach’s alpha of 0.89.

2.3.2. Attitude toward buying

Fishbein and Ajzen (1975), defined attitude toward an action as an individual’s
assessment of the favorableness of the consequences of engaging in that action. They
suggested that the assessment of the consequences of engaging in an action can be
measured using four 7-point semantic-differential items anchored as follows: good/bad;
foolish/wise; pleasant/unpleasant; beneficial/harmful. When the action is the purchase of a
product, attitude toward the action is usually measured by presenting respondents with a
statement such as ‘.buying the Performa would be.’ followed by the four 7-point
semantic-differential items. Based on data provided by 394 respondents, the four-item
scale proved highly reliable, with Cronbach’s alpha of 0.94. Attitude toward the action is
the summation of the scores across the four items. This was the same way Shimp and
Sharma (1987) measured attitude toward buying foreign cars. The mean score of
ATTITUDE, representing the sum of the four-item scale for each respondent, for the 394
respondents included in this analysis, was 16.58.

2.3.3. CDE and CMA

Two country cues were hypothesized to affect evaluative beliefs about and attitudes
toward buying automobiles. The first was CDE, which represented the country in
which automotive engineers specify all physical aspects of the car. These include the
aesthetic, mechanical, and safety features of the car. CMA is the second country cue.
This cue represented the factory location and nationality of the assembly-line workers.
There were two possible values for CDE in the research design. These included Japan
and United States. Similarly, there were two possible values for CMA, Japan or the
United States.
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 739

Table 3
The consumer ethnocentrism scale, Shimp and Sharma (1987)

1 Americans should only buy American-made products

2 Only those products that are unavailable in the US should be imported
3 Buy American products; keep Americans working
4 American products first, last, and foremost
5 Purchasing foreign-made products is un-American
6 It is not right to buy foreign products
7 A real American always buys American-made products
8 We should purchase products manufactured in the US instead of letting others
get rich off of us
9 It is always best to buy American-made products
10 There should be very little purchasing of goods from other countries, unless out of
necessity
11 Americans should not buy foreign products because this hurts American business
and causes unemployment
12 Curbs should be put on all imports
13 It may cost me more, but I prefer to buy American products
14 Foreigners should not be allowed to put their products on our markets
15 Foreign products should be taxed heavily to reduce their entry into the US
16 We should buy from foreign countries only those products that we cannot obtain
in our own country
17 American consumes who purchase products mad in other countries are responsible
for putting their fellow Americans out of work

2.3.4. Ethnocentrism
Shimp and Sharma (1987) defined highly ethnocentric consumers as those whose sense
of belonging to the national ‘in group’ is strengthened by buying domestic products and
refusing to buy imports. Highly ethnocentric consumers consider the morality of buying
domestic rather than imported products. They score highly on the 17-item CETSCALE.
Conversely, low scorers on the CETSCALE—low-ethnocentric consumers—are not
driven by concerns about the morality of buying domestic or imported products. Their
purchase decisions are not guided by a need to feel a sense of belonging to the national in-
group. An individual’s ethnocentrism is measured as the summed scores of the 17 items on
the CETSCALE. Based on the data collected, scores on the CETSCALE (hereafter
referred to as CETSCALE) ranged from 17 to 119, with a mean score of 61.68. Using
CETSCALE scores, the sample was split into three groups. The group with the lowest
scores was labeled low (NZ134), those with mid level scores were labeled medium
(NZ125), and those with the highest scores were labeled high (136). The 17-item
CETSCALE (Shimp & Sharma, 1987) is reproduced in Table 3.

3. Results

Consistent with much previous research, few relationships between consumer

ethnocentric tendency and demographics were identified. Cross-tabulations of ethno-
centric tendency and age group showed some consistency with earlier studies.
Respondents over the age of 55 and under the age of 35 appeared slightly more likely
740 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

to be highly ethnocentric than those in the middle age ranges of 35–55, who were less
likely to be ethnocentric (c2Z7.9, p!0.10). This is consistent with the findings of Imbert
et al. (2003). Additionally, there seemed to be some association between union
membership and ethnocentric tendency. Members of unions appeared more likely to
express higher levels of ethnocentric tendency compared to non-union members (c2Z6.7,
p!0.05). No education level or gender effects were identified. Patterns related to military
service and ethnocentric tendency were non-significant.
The proposed research hypotheses were tested via a 3!4 completely between-subjects
multivariate analysis of variance design (MANOVA). Analysis was performed using the
GLM multivariate function in SPSS 11.0. The two dependent variables in the analysis
were evaluative beliefs (EVALUATION) and attitude toward purchase (ATTITUDE).
The independent variables included ethnocentrism (CETSCALE) with three levels (low,
moderate, or high) and country treatment (FORM) with four levels (Japan–Japan, Japan–
US, US–Japan, and US–US). Both independent variables were entered simultaneously.
Bartlett’s Test of Sphericity was significant, indicating that the two dependent variables
were correlated, making a multivariate analysis appropriate. Checks of equality of
covariance matrices across groups, including Box’s M were also acceptable. The mean
scores of the two dependent variables for each of the 12 treatment groups are presented in
Table 4.
The results of the MANOVA for the whole sample are presented in Table 5. The
multivariate statistics showed significant (p!0.001) main effects. The multivariate-F
statistics for the FORM and CETSCALE main effects were 9.08 and 7.97, respectively.
The two-way interaction between FORM and CETSCALE was also significant (FZ7.14,
p!0.001). Adjusted R2 values for the two dependent variables indicated that the combined
Table 4
Mean scores on dependent variables for whole sample and treatment groups

Whole sample Low ethnocentric Medium ethnocentric High ethnocentric

(NZ394) (NZ134) (NZ125) (NZ136)
Evalu- Attitude Evalu- Attitude Evalu- Attitude Evalu- Attitude
ation ation ation ation
Japanese- 54.88 14.96 59.50 19.30 52.11 15.04 47.35 10.48
designed/ (nZ105) (nZ105) (nZ37) (nZ37) (nZ28) (nZ28) (nZ40) (nZ40)
Japanese-
made
Japanese- 55.73 17.21 59.14 20.03 54.60 17.65 50.26 14.13
designed/ (nZ102) (nZ102) (nZ35) (nZ35) (nZ36) (nZ36) (nZ31) (nZ31)
American-
made
American- 53.68 16.30 56.80 19.42 53.15 16.58 47.64 12.91
designed/ (nZ92) (nZ92) (nZ26) (nZ26) (nZ33) (nZ33) (nZ33) (nZ33)
Japanese-
made
American- 53.61 18.34 47.83 15.55 51.96 17.85 58.59 21.63
designed/ (nZ95) (nZ95) (nZ36) (nZ36) (nZ27) (nZ27) (nZ32) (nZ32)
American-
made
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 741

Table 5
MANOVA results for the whole sample (NZ394)

Effect Dependent DF Multivariate F Univariate F Eta squared

variable
Intercept 2 5422.79*** 0.97
Evaluation 1 10664.69*** 0.92
Attitude 1 4114.81*** 0.98
Form 6 7.97*** 0.06
Evaluation 3 0.90 0.01
Attitude 3 7.84*** 0.06
Cetscale 4 9.08*** 0.05
Evaluation 2 7.64*** 0.04
Attitude 2 18.15*** 0.09
Form!Cetscale 12 7.14*** 0.10
Evaluation 6 9.28*** 0.13
Attitude 6 14.71*** 0.19

*p!0.10, **p!0.05, ***p!0.001.

main effects and interactions explained 14.1% of the variation in EVALUATION and
26.5% of the variation in ATTITUDE for the entire sample.
Univariate F statistics provided more insight into between-group differences with
respect to each dependent variable. The main effect of FORM was significant only for
ATTITUDE (FZ7.84, p!0.001). However, there was no significant main effect of the
country treatment (FORM) on product evaluation. The CETSCALE main effect on
ATTITUDE was statistically significant (FZ18.14, p!0.001). The main effect of
CETSCALE on EVALUATION was also significant (FZ7.64, p!0.001). The two-way
interaction between CETSCALE and FORM was significant for both EVALUATION and
ATITUDE (FZ9.27, p!0.001 and FZ14.71, p!0.001, respectively).
The main effects of CETSCALE and its interactions with FORM were significant for
both dependent variables. The sample was split into low, moderate, and high ethnocentric
groups to examine differences in attitudes and evaluations of the four types of cars (defined
in terms of both country cues) for the three ethnocentrism groups separately. Using
EVALUATION and ATTITUDE as dependent variables, multivariate analyses of
variance were performed for the low, moderate, and high ethnocentric groups separately.
The independent variable used in these analyses was FORM, representing the four design-
country/assembly-country experimental treatments.

3.1. High ethnocentric group

Using data from the high ethnocentric group, a multivariate analysis was performed.
The results are presented in Table 6. The multivariate test proved significant (FZ10.91,
p!0.001). This indicated the presence of at least one significant difference in highly
ethnocentric consumers’ evaluative beliefs or attitudes toward cars carrying differing
country cues. Univariate F statistics showed significant country treatment effects on both
EVALUATION (FZ8.25, p!0.001) and ATTITUDE (FZ23.77, p!0.001). Adjusted R2
statistics showed that the country main effect explained 14 and 34% of the variation in
742 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

Table 6
MANOVA results for the high ethnocentric group (NZ136)

Effect Dependent DF Multivariate F Univariate F Eta squared

variable
Intercept 2 1591.65*** 0.96
Evaluation 1 3139.36*** 0.96
Attitude 1 889.20*** 0.87
Form 6 11.544*** 0.20
Evaluation 3 6.51*** 0.16
Attitude 3 23.77*** 0.35

*p!0.10, **p!0.05, ***p!0.001.

EVALUATION and ATTITUDE, respectively. Examination of the marginal means

(Table 4) showed that highly ethnocentric consumers expressed the most positive
evaluative beliefs (58.59) for uni-national American-designed/American-made cars. They
expressed the least positive evaluative beliefs (47.64) about bi-national American-
designed/Japanese-made cars. High ethnocentric consumers expressed the most positive
attitude toward buying the uni-national American-designed/American-made car (21.63),
and the least positive attitude toward buying the Japanese-designed/Japanese-assembled
car (10.48).

3.2. Moderate ethnocentric group

Next, data provided by moderately ethnocentric respondents were analyzed. In

spite of a marginally significant multivariate F statistic (FZ2.30, p!0.05) there were
no significant design-country/assembly-country combination main effects on moder-
ately ethnocentric respondents’ evaluative beliefs about or attitudes toward buying
cars.

3.3. Low ethnocentric group

Finally, analysis was performed using data from the low ethnocentric group only.
Results are presented in Table 7. The analysis showed a significant country treatment main
effect (multivariateZF 4.83, p!0.001). Univariate statistics showed significant country-
based differences in both EVALUATION (FZ9.30, p!0.001), and ATTITUDE
(FZ6.22, p!0.05). Adjusted R2 values show the country treatment main effect explained
16% of the variation in EVALUATION and 11% of the variation in ATTITUDE.
Examination of the marginal means of the two dependent variables showed that low
ethnocentric consumers expressed the most positive evaluative beliefs (59.50) about the
Japanese-designed/Japanese-made car Japan. They expressed the least positive evaluative
beliefs (47.83) about the American-designed/American-made car. Low ethnocentric
consumers expressed the most positive attitudes toward buying the bi-national Japan-
designed/American assembled car (20.03), and the least positive attitudes toward buying
the American-designed/American-made car (15.55).
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 743

Table 7
MANOVA results for the low ethnocentric group (NZ134)

Effect Dependent DF Multivariate F Univariate F Eta squared

variable
Intercept 2 2720.32***
Evaluation 1 5292.49*** 0.96
1 2494.99*** 0.93
Form 6 4.56***
Evaluation 3 8.06*** 0.11
Attitude 3 3.52** 0.05

*p!0.10, **p!0.05, ***p!0.001.

3.4. Hypotheses testing

A series of post-hoc comparisons was performed using data from both the low and high
ethnocentrism groups. Using data from within each ethnocentric group separately, mean
scores on evaluative beliefs and attitudes toward buying each design-country/assembly-
country combination were compared. Multiple comparisons were conducted using post-
hoc Tukey’s honestly significant difference test. This test uses the studentized range
statistic to make all pair-wise comparisons between groups, and sets the experiment-wise
error rate for the collection of all pair-wise comparisons. These comparisons are presented
in Table 8.
According to the table, a clear bias in favor of the purely American vehicle appears
among the highly ethnocentric respondents. Both evaluative beliefs about and attitudes
toward buying the uni-national American car are significantly higher (p!0.05) than
beliefs or attitudes toward any of the other design/assembly-country combinations for this
group. This is illustrated by the negatively signed mean differences that are highlighted in
the table. These finding support the first two hypotheses, H1a and H1b. These results also
support H2a and H2b, which posited that highly ethnocentric consumers would have more
positive evaluative beliefs about and attitudes toward buying uni-nationally American
designed and assembled cars than American-designed cars made in Japan.

Table 8
Post hoc tests of mean differences for low and high ethnocentric groups

County of Country of Low ethnocentric consumers High ethnocentric consumers

design and manufacturing
Mean differ- Mean differ- Mean differ- Mean differ-
engineering and assembly
ence attitude ence evaluative ence attitude ence evaluative
(CDE) (CMA)
beliefs beliefs
Japan, Japan Japan, US K0.73 0.32 K3.65** K2.9
Japan, Japan US, Japan K0.13 2.65 K2.43 K0.29
Japan, Japan US, US 3.74** 11.62** K11.15** K11.24**
Japan, US US, Japan 0.61 2.34 1.22 2.62
Japan, US US, US 4.47** 11.31** K7.50** K8.34**
US, Japan US, US 3.87** 8.97** K8.71** K10.96**

*p!0.10, **p!0.05, ***p!0.01.

744 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

Hypotheses H3a and H3b considered whether highly ethnocentric consumers would
prefer Japanese cars manufactured in the US to uni-national Japanese cars. Underlying
these hypotheses was the question of whether highly ethnocentric consumers were thus
motivated by their desire to protect the US car industry the jobs of US auto workers. The
results in Table 8 support H3b, that posited a more positive attitude toward buying
Japanese cars that are American made than Japanese cars made in Japan at p!0.05.
However, H3a was not supported. Evaluative beliefs about Japanese-designed cars, among
highly ethnocentric consumers, do not significantly differ based upon country of
manufacturing and assembly.
In contrast to their highly ethnocentric counterparts, low-ethnocentric respondents’
biases appear to be clearly opposite in favor of the purely Japanese-designed cars. The low
ethnocentric consumers appear to have significantly more favorable attitudes toward
buying and evaluative beliefs about the purely Japanese cars than toward the American-
designed and assembled cars (p!0.05). These findings support H4a and H4b, which posited
that in the absence of high ethnocentrism, consumers would have more positive evaluative
beliefs about and attitudes toward buying Japanese versus American automobiles. The
results in Table 8 illustrate significant bias against the purely American-designed and
engineered cars among low-ethnocentric consumers. Evaluative beliefs about and
attitudes toward buying the purely American-designed and engineered cars are
significantly less positive (p!0.05) than those for any of the other design/assembly-
country combinations. According to the results of the post hoc tests, there were no
differences in these low ethnocentric respondents’ evaluative beliefs about attitudes
toward buying any car that was at least partially Japanese.

4. Discussion and conclusions

Country-of-origin choices are not inconsequential. Choosing to design, source raw

materials, or assemble in one country versus another may lead to cost savings or quality
improvements. However, poorly conceived design-country (CDE) or assembly-country
(CMA) choices may lead to negative consequences. In recent years, some American
clothing manufacturers attempted to manipulate CMA information by setting up
sweatshops in places like Saipan, an American territory. Technically, the clothing could
carry ‘Made-in-the-US’ labels. However, the workers in these sweatshops were not
American citizens. Rather, they were young women from China and other Southeast Asian
countries working for low wages in deplorable conditions. New light was shed on old
issues as the American public looked not only at immediate questions concerning what
country-of-origin is and how it could or should be defined, but to issues of corporate
citizenship and morality. Managers must consider consumers’ perceptions about the
morality of country of design and country of assembly choices, in addition to how these
choices affect cost or perceived product quality, because such attitudes also affect
consumer purchase decisions.
Much of the earlier product–country–image literature focused only on how consumers
use country information as cues for product quality. This study examined the effects of
design- and assembly-country information on perceived product quality. Beyond that, it
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 745

also considered whether and how such cues affect consumers’ feelings about the goodness
or rightness of buying products based upon these two country cues. Incorporating
measures of consumer ethnocentric tendency allowed for studying how these effects vary
across different market segments.
It is not surprising that highly ethnocentric American consumers have more
positive evaluative beliefs about American-designed-and-manufactured automobiles
than purely Japanese ones. Nor is it terribly surprising that this group of consumers
feels better about buying purely American cars than buying purely Japanese ones. In
fact, this segment of the sample had significantly more positive evaluative beliefs and
attitudes toward buying uni-national American-designed-and-manufactured cars than
cars with any of the other design-country/assembly-country combinations. It is
possible that for them, any Japanese involvement raises red flags. On the other hand,
for the low ethnocentric consumer, any Japanese involvement might actually make the
products more attractive.
The interesting results lies in the post-hoc comparisons of the evaluative ratings of and
attitudes toward buying Japanese-designed cars manufactured in Japan, or in the United
States. There were no significant differences in evaluative beliefs about Japanese-
designed/Japanese-manufactured cars and the Japanese-designed/American-manufactured
cars. That is, regardless of the degree of ethnocentric tendency, the results show that
consumers’ evaluative beliefs about the Japanese-designed cars do not appreciably change
when production is moved from Japan to the United States.
Even though highly ethnocentric respondents appear to evaluate the two types of
Japanese-designed vehicles equally, these respondents expressed more positive
attitudes toward buying Japanese cars that are manufactured in the US. There were
no such country-of-manufacturing-and-assembly-based differences in low ethnocentric
respondents’ attitudes toward buying Japanese-designed cars. In real-world terms,
these findings suggest that Japanese auto makers like Toyota may have increased the
likelihood of selling Japanese cars to all American consumers by choosing to
manufacture in US. In doing so, they could overcome highly ethnocentric consumers’
reluctance to buying imports without jeopardizing their high-quality brand image
among consumers who are less ethnocentric. Thus, the company continues to
strategically leverage its positive country-of-design image while overcoming consumer
objections through its choice of country of assembly.
By including both country of design and engineering as well as country of
manufacturing and assembly, the results show that Japanese design transcends country
of manufacture choice, at least when the two potential manufacturing countries are the
US and Japan. While highly ethnocentric consumers may continue to prefer buying
purely American products over ones that are in any part Japanese, they do not appear
to believe that manufacturing in the US improves the quality of Japanese products.
From this, it seems that highly ethnocentric consumers tend to consider the morality
of country-based purchasing decisions over the effects of country on product quality.
In contrast, low ethnocentric consumers perceive the purely American cars to be
inferior in quality and are averse to buying them. However, any input from Japanese
companies—either in terms of manufacturing or design processes, may improve low
ethnocentric consumers’ likelihood of purchase. In the absence of strong patriotic
746 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

emotion, the data in this study suggest that these low ethnocentric consumers are
more likely to use country information when evaluating and forming beliefs about
products. Understanding the dynamics of how consumers of varying levels of
ethnocentric tendency respond to Japanese or American design or manufacturing
should be helpful to managers of Japanese as well as American firms. The results of
this study seem to indicate that the Japanese companies perhaps understand this better
than their American competitors.
The data used in this study were collected in one region of the United States in 1995 and
focused on only one product category. Automobiles, by their very nature as high
involvement products and their importance to the US and world economy, may not be
typical of all product categories. PCI research, popular in the 1980s and 1990s seemed to
have concluded that country of origin matters to some consumers for some products, but
its very definition as a construct had become hazy.
The analysis presented in this paper shows highly ethnocentric American consumers
have distinct biases in favor of American-designed American-made cars. Low
ethnocentric consumers have distinct biases against American-designed American-made
cars and in favor of Japanese vehicles. Such distinctly opposite reactions to design and
manufacturing country choices might cancel each other out if consumer characteristics
such as ethnocentric tendency are not considered. Failing to consider such differences may
lead managers to erroneously conclude that country-of-origin choices do not affect
consumer behavior. Overlooking country-based effects may not only be costly in terms of
lost customers, but also in terms of lost opportunities.
Although the data used in this study were collected in the mid 1990s, the phenomena
that they captured then remain relevant today. By the end of the 1990s, multinational
production had become so commonplace that many researchers began to think that PCI
studies had outlived their relevance. Just as many might have concluded that PCI studies
were passé, the issue, in a slightly mutated form, burst onto the front pages of the press.
Many American consumers had already come to accept the inevitable loss of
manufacturing sector jobs to cheaper markets. A good proportion of those willing to
accept the inevitable progress in global efficiency may have taken comfort in the perceived
security of their high tech or service sector jobs.
Things seem to have progressed a bit faster than many may have liked. Now, as non-
manufacturing jobs are sent offshore, there seems to be a renewed sense of ethnocentrism
among American consumers. Certainly, the media have identified this as a key issue. The
foundation established through a three-decade rich body of PCI literature, and notions of
consumer ethnocentric tendency, may provide a solid framework for studying these issues.
Whether in manufacturing or services, the emotional and moral considerations of country-
based buying decisions remain an important concern for some people. That consumer
ethnocentrism may begin to manifest itself in more segments of American society (those
working in high tech jobs, for example) remains a rich area for further exploration.
Although this study revisits a manufacturing-based model, it focuses on attitudes toward
buying based upon choosing to perform some links of the value chain offshore. Thus, it
provides a useful starting point for studying the issue with respect to other links in the
value chain.
 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748 747

References

Acharya, C., & Elliott, G. (2001). An examination of the effects of ‘country-of-design’ and ‘country-of assembly’
on quality perceptions and purchase intentions. Australasian Marketing Journal, 9(1), 61–75.
Brodowsky, G. H. (1998). The effects of country of design and country of assembly on evaluative beliefs and
attitudes toward buying them: A comparison between low and high ethnocentric consumers. Journal of
International Consumer Marketing, 10(3), 85–113.
Chao, P. (1992). Partitioning country-of-origin effects: Consumer evaluations of a bi-national product. Journal of
International Business Studies, 24(2), 291–306.
Ettenson, R. (1993). Brand name and country of origin effects in the emerging market economies of Russia,
Poland, and Hungary. International Marketing Review, 10(5), 14–36.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research.
Reading, MA: Addison Wesley.
Gaedeke, R. (1973). Consumer attitudes toward products ‘made-in’ developing countries. Journal of Retailing,
49(2), 13–24.
Han, M. C. (1988). The role of consumer patriotism in the choice of domestic versus foreign products. Journal of
Advertising Research, 28(3), 25–32.
Han, M. C., & Terpstra, V. (1988). Country-of-origin effects for uni-national and bi-national products. Journal of
International Business Studies, 19(2), 235–254.
Herche, J. (1992). A note on the predictive validity of the CETSCALE. Journal of the Academy of Marketing
Science, 20(3), 261–264.
Heslop, L. A., Liefeld, J. P., & Wall, M. (1987). An experimental study of the impact of country-of-origin
information. In R. E. Turner, Marketing (Vol. 8) (pp. 179–185). Toronto, Ont.: Proceedings, Administrative
Sciences Association of Canada—Marketing Division, 179–185.
Imbert, G., Jiddou, K., Kumar, S., Murillo, A., & Zhao, P. (2003). Analysis of Russian values. A Directed study at
Wayne State Univerity, http://www.gerardimbert.com/files/mkt7995_research.pdf.
Johansson, J. K. (1993). Missing strategic opportunity: Mangers’ denial of country-of-origin effects. In N. A.
Papadopoluous, & L. A. Heslop (Eds.), Product country images, impact and role in international marketing
(pp. 77–86). Binghamton, NY: International Business Press an imprint of the Haworth Press Inc, 77–86.
Johansson, J. K., Douglas, S. P., & Nonaka, I. (1985). Assessing the impact of country-of-origin on product
evaluations: A new methodological perspective. Journal of Marketing Research, 22, 388–396.
Johansson, J. K., & Nebenzahl, I. D. (1986). Multinational production effects on brand value. Journal of
International Business Studies, 17(3), 101–126.
Luthy, M. R., & Parsa, R. A. (1998). Student ethnocentrism and attitudes toward international trade agreements.
Presented at the Marketing Management Association, Chicago, IL.
Maronick, T. J. (1995). An empirical investigation of consumer perceptions of ‘made-in USA’ claims.
International Marketing Review, 12(3), 38–52.
McCarthy, J. C., Dash, A., Liddell, H., Ross-Ferrusi, C., & Tempkin, B. D. (2002, November 11). 3.3 million US
service jobs to go offshore. Retrieved April 27, 2004 from http://ww.forrester.com/ER/Research/Brief/
Excerpt/0.1317.15900.00.html.
Nagashima, A. (1970). A comparison of Japanese and U.S. attitudes towards foreign products. Journal of
Marketing, 34, 68–74.
Nagashima, A. (1977). A comparative ‘made-in’ product image survey among Japanese businessmen. Journal of
Marketing, 41, 95–100.
Nielsen, J. A., & Spence, M. T. (1997). A test of the stability of the Cetscale, a measure of consumers’
ethnocentric tendencies. Journal of Marketing Theory and Practice, 5(4), 68–76.
Obermiller, C., & Spangenberg, E. (1989). Exploring the effects of country-of-origin labels: An information
processing framework. In T. K. Srull, Advances in consumer research (Vol. 16) (pp. 454–459). Provo, UT:
Association of Consumer Research, 454–459.
Reirson, C. C. (1967). Attitude change toward foreign products. Journal of Marketing Research, 4, 385–387.
Samiee, S. (1994). Customer evaluations of products in a global market. Journal of International Business
Studies, 25(3), 579–604.
 748 G.H. Brodowsky et al. / International Business Review 13 (2004) 729–748

Schooler, R. D. (1965). Product bias in central American product market. Journal of Marketing Research, 5,
296–299.
Schooler, R. D. (1971). Bias phenomena attendant to the marketing of foreign goods in the U.S. Journal of
International Business Studies, 2, 71–80.
Schooler, R. D., & Wildt, A. R. (1968). Elasticity of product bias. Journal of Marketing Research, 5, 78–81.
Shimp, T. A., & Sharma, S. (1987). Consumer ethnocentrism: Construction and validation of the CETSCALE.
Journal of Marketing Research, 24, 280–289.
Tse, D. K., & Gorn, G. (1993). An experiment on the salience of country of origin in an era of global brands.
Journal of International Marketing, 1(1), 57–76.
Wall, M., Liefeld, J. P., & Heslop, L. A. (1991). Impact of country-of-origin cues on consumer judgments in
multi-cue situations: A covariance analyze. Journal of the Academy of Marketing Science, 19(2), 105–113.

View publication stats