In Brands We Trust? A Multicategory, Multicountry Investigation of Sensitivity of Consumers' Trust in Brands To Marketing-Mix Activities

In Brands We Trust?
A Multicategory,
Multicountry Investigation of Sensitivity
of Consumers’ Trust in Brands to
Marketing-Mix Activities
Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

KOUSHYAR RAJAVI
TARUN KUSHWAHA
JAN-BENEDICT E. M. STEENKAMP
The essence of a brand is that it delivers on its promises. However, consumers’

trust in brands (CTB) has declined around the world in recent decades. As a re-
sult, CTB has become a major concern for managers. The authors examine
whether CTB is influenced by marketing-mix activities (i.e., advertising, new prod-
uct introduction, distribution, price, and price promotion) implemented by brands.
The authors propose and show that the sensitivity of CTB to marketing-mix activi-
ties is moderated by consumer, category, and country characteristics, using a mul-
tisource data set consisting of a survey of 15,073 respondents and scanner panel
data on 589 brands in 46 CPG categories across 13 countries (including the four
largest emerging markets), which collectively account for half of the world’s popu-
lation. The authors find strong positive effects for advertising and new product in-
troduction intensity, weak positive effects for price and distribution intensity, and a
minor negative effect for price promotion intensity on CTB. Furthermore, the
authors find that the effect of marketing-mix activities on CTB is moderated by
consumers’ personality traits, consumers’ reliance on brands in a category, and
countries’ secular-rational and self-expression cultural values.
Keywords: consumer trust, brand trust, marketing mix, branding, information

economics, international marketing
Koushyar Rajavi (Koushyar.rajavi@scheller.gatech.edu) is an as-

sistant professor of marketing at Scheller College of Business at
Georgia Institute of Technology, 800 West Peachtree NW, Atlanta, INTRODUCTION
GA 30308. Tarun Kushwaha (tarun_kushwaha@unc.edu) is Sarah
Graham Kenan Scholar and an associate professor of marketing at the
Kenan-Flagler Business School of the University of North Carolina,
300 Kenan Center Dr, Chapel Hill, NC 27599. Jan-Benedict E. M.
Steenkamp (jbs@unc.edu) is C. Knox Massey Distinguished Professor
F or many firms, brands are among their most valuable
assets. According to brand consultancy Kantar
Millward Brown, the value of the 100 most valuable global
of Marketing at the Kenan-Flagler Business School of the University brands alone stood at $4.4 trillion in 2018 (Millward
of North Carolina, 300 Kenan Center Dr, Chapel Hill, NC 27599. Brown 2018). What makes brands so valuable? Consider
Please address correspondence to Tarun Kushwaha. The article is
based on the first essay of the lead author’s dissertation. The authors
the definition of brand proposed in the literature. One such
thank AiMark for providing data for this study. They also acknowl- definition, proposed by Kotler (2002, 593), is “a seller’s
edge the kind help of Bernadette van Ewijk for extracting data for this promise to deliver a specific set of features, benefits and
study. Supplementary materials are included in the web appendix ac- services consistent to the buyers.” If consumers trust the
companying the online version of this article.
brand to deliver on these promises, this eases their decision
Editor: J. Jeffrey Inman making, reduces costs of information gathering and proc-
essing information, reduces their purchase risk, and
Associate Editor: Russell S. Winer
increases expected utility (Erdem and Swait 1998, 2004).
Advance Access publication June 21, 2019 Thus, trust is a key factor to consider in brand success.
C The Author(s) 2019. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc.
V
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Vol. 46 2019
DOI: 10.1093/jcr/ucz026
651
652 JOURNAL OF CONSUMER RESEARCH
Kantar Millward Brown (2016) found that business-to- 13 countries, including the US; European countries such as
business (B2B) brands that rated high on brand trust grew France, Germany, and Italy; and the four leading emerging
80% in brand value in the last decade, while less trusted markets, Brazil, China, India, and Russia. Marketing-mix
brands grew only 25%. As another example, Nanda (2014) instruments are derived from surveys and household panels
reported that when it comes to convincing consumers to operated by Kantar Worldpanel, GfK, and IRI. The study
pay more, brand trust trumps other brand qualities. findings provide managers with strategic direction on how
Given the importance of consumer trust in brands, it is their marketing-mix activities affect consumer trust in one
worrying that industry evidence indicates that consumers’ of their most valuable assets and how their marketing ac-
trust in brands is slipping. Young & Rubicam (2017) ana- tivities have differential impact on CTB across different
lyzed a fairly constant set of well-known brands and consumers, categories, and countries.
reported that the proportion of brands that customers said
they trusted has fallen from 44% in 2001 to a low of 18% in

2017. The 2017 Edelman Trust Barometer found that in PREVIOUS RESEARCH
nearly half of the countries surveyed, the percentage of peo-
ple that mistrust brands’ owners exceeds the percentage of In order to delineate the position of CTB within the
people that trust them (Edelman 2017). Given these results, branding literature and highlight our contribution vis-a-vis
it is not surprising that consumer trust has moved to the top past research on CTB, we use Keller and Lehmann’s
of management’s priority list. In 2015, when the CEOs of (2003) brand value chain framework (figure 1). According
leading consumer goods firms such as P&G, Nestle, and to this framework, the brand-value-creation process begins
PepsiCo gathered for the 59th Consumer Goods Forum’s an- when the firm invests in a marketing program targeting ac-
nual summit, “Trust as a Foundation for Growth” was their tual or potential customers. The marketing activity associ-
main topic of discussion (Consumer Goods Forum 2015). ated with the program influences the customer mindset
Consumer researchers have recognized the importance of with respect to the brand—that is, what they know and feel
consumers’ trust in brands (CTB), defined as the consumer’s about the brand. This mindset, across a broad group of cus-
belief that the brand is willing and able to deliver on its tomers, then results in brand performance in the market-
promises (Chaudhuri and Holbrook 2001; Erdem and Swait place, which is the aggregate of individual customer
2004). The focus of this stream of research has largely been actions regarding quantity purchased and the price that
on the consequences of CTB (see next section). they pay. Finally, the investment community considers
We build upon and extend previous work in three mean- brand performance in valuing the company. The brand-
ingful ways. First, we shift the lens from the consequences value-chain model further specifies multipliers that
of CTB to its antecedents. More specifically, we are inter- strengthen or weaken the link between two successive
ested in the effects of marketing-mix activity on CTB. We stages, in terms of the value of these multipliers. These
examine how CTB is influenced by five key marketing moderating factors determine how value created at one
instruments: advertising intensity, new product introduc- stage transfers or “multiplies” to the next stage (Keller and
tion intensity, price, price promotion, and distribution in- Lehmann 2003, 28).
tensity. Second, we examine the sensitivity of CTB to the Past research has focused on the causal link between dif-
marketing-mix activities on a global basis. Various aca- ferent stages in the brand value chain, including the effect
demics have urged marketing scholars to investigate con- of marketing-mix instruments on consumer mindset
sumer behavior issues on an international basis (Erdem (Srinivasan, Vanhuele, and Pauwels 2010), consumer
et al. 2006). Are conclusions regarding the effects of mar- mindset on marketplace performance (Datta, Ailawadi, and
keting activities on CTB globally generalizable? Third, we Van Heerde 2017), and marketplace performance on finan-
examine boundary conditions to the findings within and cial market performance (Srinivasan et al. 2009). Other re-
across countries. We investigate whether the sensitivity of search has focused on relations between customer mindset
CTB to the marketing-mix activities systematically across metrics, which, according to Keller and Lehmann (2003,
context, where we distinguish between three context fac- 29), exhibit an “obvious hierarchy.” This stream of re-
tors: consumer, according to their personality traits search has distinguished between different customer mind-
(Inman, McAlister, and Hoyer 1990); (product) category, set metrics within a broader category. In this stream of
in terms of consumers’ reliance on brands in a category research, brand awareness (Hoyer and Brown 1990) sup-
(Fischer, Völckner, and Sattler 2010); and country, accord- ports brand associations, which drive brand attitudes and
ing to their national culture (Inglehart and Wetzel 2005). overall brand evaluations (Wilkie and Pessemier 1973) that
We put together a unique cross-sectional data set from lead to brand attachment (brand attachment, love, attitudi-
multiple sources, which contains primary (survey) data as nal loyalty, brand-self connection; Batra, Ahuvia, and
well as secondary (household panel, country) data from Bagozzi 2012; Park et al. 2010), and brand activity (behav-
15,073 respondents on 589 brands in 46 consumer pack- ioral loyalty, word of mouth [WOM]; Chaudhuri and
aged goods (CPG) categories. Our data set covers Holbrook 2001).
RAJAVI, KUSHWAHA, AND STEENKAMP 653
FIGURE 1
EXPANDED BRAND VALUE CHAIN

Adapted from Keller and Lehmann (2003).
NOTE.—In italics are the constructs included in the current study; constructs that were examined in previous research on CTB are indicated with continuous under-
lines; constructs that are added to the value chain are indicated with dotted underlines.
Where does trust fit into the brand value chain? such as brand consideration (Erdem and Swait 2004),
Following Chaudhuri and Holbrook (2001), we argue that purchase likelihood (Erdem, Swait, and Valenzuela
CTB is a type of brand attitude that involves “a process 2006; Herbst et al. 2011), attitudinal loyalty, and be-
that is well thought out and carefully considered” havioral loyalty (Chaudhuri and Holbrook 2001). Other
(Chaudhuri and Holbrook 2001, 82). Thus, we view CTB research has investigated the effect of CTB on brand
as a cognitive component of brand attitude (see web performance metrics, such as market share and price
appendix A for a comparison between CTB and related premium (Chaudhuri and Holbrook 2001) and con-
consumer mindset metrics in marketing). sumer sensitivity to brand price (Erdem, Swait, and
Past research on CTB has extensively examined the Louviere 2002). See table 1 for a summary of past re-
impact of CTB on other consumer mindset metrics, search on CTB.
TABLE 1
SUMMARY OF PAST RESEARCH ON CTB
Marketing-mix as
Study Data drivers of CTB Findings
Erdem and Swait (1998) Survey from students Brand investments Brand investments have a positive effect on
CTB. CTB is an integral element of brand eq-
uity. CTB leads to higher perceived quality
and lower perceived risk.
Delgado-Ballester and Survey from consumers — CTB generates consumers’ commitment, espe-
Munuera-Alema n (2001) cially in situations of high involvement.
Chaudhuri and Surveys from consumers — CTB positively influences purchase loyalty and
Holbrook (2001) and managers attitudinal loyalty. CTB indirectly leads to

increases in market share and price premium.
Erdem et al. (2002) Surveys from students — CTB lowers consumers’ price sensitivity. The
negative effect depends on the level of uncer-
tainty associated with the product category.
Erdem and Swait (2004) Surveys from students — CTB impacts brand choice and consideration
set formation, especially in contexts with high
uncertainty.
Erdem et al. (2006) Survey from students in — The positive effect of CTB on choice is greater
seven countries for consumers who rate high on either collec-
tivism or uncertainty avoidance.
Sung and Kim (2010) Surveys from students — Certain brand personality dimensions (e.g., rug-
gedness, sincerity) influence CTB more than
brand affect.
This study Survey on 15,073 consum- Advertising, innovation, Advertising, innovation, distribution, and price
ers and scanner panel distribution, price, positively affect CTB. Price promotions nega-
data in 13 countries price promotion tively affect CTB. CTB is more sensitive to
marketing mix in categories low on brand rel-
evance, and countries high (low) on secular-
rational values (self-expression values).
Using the value chain framework, our research adds to to make inferences regarding the characteristics of prod-
the literature on brand building by introducing CTB into ucts and brands (see Kirmani and Rao 2000 for an over-
the value chain, and more particularly, by examining how view of past research on the signaling role of marketing-
CTB is affected by the marketing program investments.1 mix instruments). While past research primarily focuses on
Further, we extend this framework by introducing context consumer attributions regarding overall product superiority
as a multiplier of marketing program investments on cus- (i.e., product quality), some research has found that
tomer mindset metrics. marketing-mix instruments shape consumer beliefs regard-
ing specific product and brand attributes, such as reliability
(Wiener 1985) and credibility (Erdem and Swait 1998).
CONCEPTUAL BACKGROUND
Marketing-Mix Activities as Signals for Brand
Marketing-Mix Instruments as Signals
Trust
Inspired by Kirmani and Rao (2000), we use signaling
The essence of CTB is consumers’ belief that a brand
theory as the theoretical lens to understand why marketing-
delivers on its promises, time and time again. The nature of
mix investments affect CTB. A considerable body of
these promises can vary from physical attributes (e.g., or-
marketing studies has documented the signaling role of
ganic ingredients) to functional benefits (taste of coffee) to
brand-specific investments. More specifically, past re-
self-expressive benefits (e.g., “smart shopper,” brand
search shows that advertising expenditures (Kirmani
CSR). But what should hold a brand back from reneging
1990), high price (Rao and Monroe 1988, 1989), product
on its promises (e.g., selling products with nonorganic
warranties (Boulding and Kirmani 1993; Wiener 1985),
ingredients, using cheaper coffee beans, not being involved
brand name (Erdem 1998), price promotions (Yoo,
in CSR)? And how can the consumer trust the brand to de-
Donthu, and Lee 2000), and distribution outlets (Chu and
liver on its promises?
Chu 1994) effectively serve as signals that consumers use
Signaling theory, which is grounded in information eco-
1 Erdem and Swait (1998) examine the effect of general brand invest-
nomics (Klein and Leffler 1981, Milgrom and Roberts
ments on CTB. Their research, however, does not distinguish between 1986), provides an explanation (Kirmani and Rao 2000).
different marketing-mix instruments. Signaling theory recognizes the asymmetrical information
structure of the market and proposes that brands can use rationality of firms and consumers). The value of this the-
market signals to convey information to imperfectly in- ory is that it is able to explain many real-world outcomes.
formed consumers. Klein and Leffler (1981) demonstrate As Boulding and Kirmani (1993, 121–122) argue, “the
analytically that market prices above the competitive price power of the signaling framework lies in its ability to make
(i.e., price premium) are a means of enforcing brand prom- predictions based on a single behavioral assumption—
ises. Klein and Leffler (1981) further discuss that brand- rational firms and consumers. [. . .] However, it says nothing
specific expenditures on advertising that are observable to about the underlying processes by which consumers make
consumers are lost if the brand cheats. Large advertising such inferences or even whether they actually do so.”2
expenditures inform consumers of the magnitude of sunk What, then, is the empirical evidence on consumer use of
costs and hence the opportunity cost to the brand if it marketing instruments for CTB? Erdem and Swait (1998)
cheats. This provides another incentive for firms to keep find that brand investments (measured with two items:
their promises. “This brand spends lots of money on ads, commercials,

Although Klein and Leffler (1981) focus on advertising promotions, event sponsorships, celebrity endorsements,
as brand-specific marketing program investment, their ana- etc.,” and “This brand has spent a lot on the community
lytical conclusions apply to any kind of observable brand- over the years”) have a significant effect on CTB. Other
name expenditures (Milgrom and Roberts 1986, 799–800), empirical support for the use of marketing-mix activity on
including new product introductions under a given brand CTB is largely indirect; the customer mindset metric stud-
name (Milgrom and Roberts 1986) and distribution (Rao ied was perceived quality, not brand trust. Nevertheless, as
and Mahi 2003). New product introductions help a brand quality is essential to the ability of a brand to fulfill its
differentiate itself with its competitors. The innovative promises—there are few brands that do not make quality
brand relies on consumers’ repeat purchases to recoup promises—this work is informative for our purposes. A
R&D, new packaging, and other innovation-related costs. large body of marketing studies shows that consumers use
Thus, innovative brands signal to consumers that they are price (Dawar and Parker 1994; Rao and Monroe 1989) and
motivated to deliver on their promises; otherwise, they advertising (Kirmani 1990; Kirmani and Wright 1989) as
would incur great losses (fixed cost of innovation) signals of product quality. Past research in marketing has
(Milgrom and Roberts 1986). Similarly, a brand with an further found that price promotions damage brand attitude
extensive distribution network is viewed as a strong and re- (Yi and Yoo 2011), brand loyalty (Papatla and
sourceful brand that has been able to attract interest from Krishnamurthy 1996), and brand equity (Yoo et al. 2000).
multiple retailers. Consumers interpret a brand’s ubiqui- Based on the above discussion, we expect that CTB is posi-
tous presence as a sign of its consistent performance across tively affected by the brand’s advertising intensity, new
different markets. Extensive distribution costs—associated product introduction intensity, distribution intensity, and
with high expenditures on slotting allowances, in-store pro- price, and negatively affected by the brand’s price promo-
motion material, and other expensive retail investments— tion intensity.
would be lost if the brand does not deliver on its promises Consistent with the rational perspective underlying sig-
(Rao and Mahi 2003). naling theory, the above discussion abstracts from the con-
In sum, signaling theory proposes that consumers use text in which the formation of CTB judgments take place.
the extent of investments in various marketing program However, from a behavioral point of view, judgment for-
elements (high price, advertising activity, new product ac- mation is affected by the psychological makeup of the con-
tivity, distribution coverage) as signals that the brand will sumer in question, characteristics of the category involved,
deliver what it promises (i.e., that it can be trusted). But and characteristics of the country in which the consumer
what is the role of price promotions? Price promotions lead lives. We propose that these three “context” multipliers
to lower price premiums. According to the information strengthen or weaken the effect of marketing-mix activities
economics perspective, the lower price premiums caused on CTB (see figure 2).
by frequent price promotions indicate that the brand is
more likely to cheat on its promises. Marketing offers a
complementary explanation on the negative impact of price Moderating Role of Personality Traits
promotions on consumer attitude. Blattberg and Neslin We start by examining variation in the sensitivity of
(1989) argue that heavy price promotions raise suspicions CTB to marketing-mix activities in terms of the personality
in consumers’ minds regarding the capability of brands. profile of the consumer (Baumgartner 2002). Personality
When exposed to frequent price promotions by a brand, the traits are “dimensions of individual differences in tenden-
consumer “questions why it is necessary to keep promoting cies to show consistent patterns of thoughts, feelings, and
the product and concludes there is something wrong with actions” (McCrae and Costa 1990, 29). These traits are
the product” (Blattberg and Neslin 1989, 90). exhibited by individuals across a wide range of situations,
We acknowledge that information economics makes
strong assumptions, which are not always realistic (e.g., 2 We thank an anonymous reviewer for raising this point.
FIGURE 2
RESEARCH FRAMEWORK

such as child rearing, work interactions, and personal con- (1996) distinguished between two thinking styles:
sumption. In the last few decades, the Big Five has analytical-rational and intuitive-experiential. An analytical
emerged as the most influential personality trait model thinking style is characterized by deliberate, slow, analytic
(McCrae and Costa 2008). The Big Five framework distin- information processing. An intuitive thinking style is char-
guishes between five fundamental traits: openness to expe- acterized by automatic and quick information processing
rience, conscientiousness, extraversion, agreeableness, and and is associated with the use of heuristics. This suggests
neuroticism. that intuitive thinking is associated with greater reliance on
These personality traits can be linked to information market signals as heuristics in CTB, while analytical think-
processing styles. People have habitual tendencies, com- ing is associated with less reliance on market signals. The
monly called thinking styles (Epstein et al. 1996), to ap- two thinking styles supplement each other because attitude
proach various problems in consistently similar ways. formation and behavior are a joint function of both modes
These thinking styles vary between people, are stable over of processing (Epstein et al. 1996, 392).
time (Scott and Bruce 1995), and are related to people’s Subsequent empirical research examined the relationship
personality (Epstein et al. 1996). Epstein and colleagues between these thinking styles and the Big Five personality
traits (Pacini and Epstein 1999; Riaz, Riaz, and Batool Theory, the likelihood that an input will be used for judg-
2012), finding that the tendency to engage in rational, ana- ment is determined by accessibility of the input in memory
lytical thinking is related to openness to experience and (i.e., ease of retrieval), perceived diagnosticity of the input
conscientiousness, while the intuitive thinking style is as- (i.e., attribute relevance), and availability of other inputs in
sociated with agreeableness and extraversion. Neuroticism memory. Factors that increase the accessibility of an input
did not exhibit consistent relations with either thinking will increase the probability that the input is used in judg-
style. Thus, we expect marketing-mix activities to have ment formation (Herr, Kardes, and Kim 1991).3 Drawing
larger effects on CTB for consumers who are high on ex- on Accessibility-Diagnosticity Theory, several studies
traversion and agreeableness and smaller effects on CTB have shown that when the accessibility of brand-related in-
for consumers who are high on conscientiousness and formation increases, the likelihood that consumers use
openness to experience. such information as an input for brand evaluations
increases (Li and He 2013; Menon and Raghubir 2003).

Moderating Role of Brand Relevance in We argue that when BRiC is high, brands are important to
consumers, and consumers are paying close attention to the
Category marketing activities of these brands (Fischer et al. 2010).
The thinking-style perspective describes modes of infor- As a result, brand-related information is more accessible to
mation processing that are generalized across situations. them and hence CTB is more likely to be influenced by
However, ever since the work of Belk (1974), consumer marketing activities when BRiC is high. In sum, while the
researchers have been aware that attitude formation can be Elaboration Likelihood Model leads to the prediction that
(more) strongly affected by the actual judgment situation. marketing activities have larger (smaller) effects on CTB
Past consumer research has studied the role of various cate- when BRiC is low (high), Accessibility-Diagnosticity
gory characteristics in attitude formation, such as category Theory leads to the prediction that marketing activities
risk, involvement, utilitarian versus hedonic, and purchase have larger (smaller) effects on CTB when BRiC is high
frequency. We believe that a relatively recently introduced (low).
construct, “brand relevance in a category” (Fischer et al.
2010), is particularly useful for the purposes of the present
Moderating Role of National Culture
study. Implicit in the information economics perspective is
that consumers can—and do—rely on brands in their deci- For studying the multiplier effect of the country context
sion making. After all, if consumers do not care about on the effect of marketing-mix activities on CTB, we take
brands, they will not closely follow the brands, and, a cultural approach. According to Tse et al. (1988, 82), na-
therefore, will not notice brands’ marketing activities. tional culture influences consumers’ “rules for selective at-
Fischer et al. (2010) have shown that the role of the brand tention, interpretation of environmental cues, and
in consumer decision-making processes systematically responses.” This suggests that consumers’ utilization of
varies by product categories. They further showed that marketing signals varies predictably across countries
brand relevance in a category (BRiC) varies between depending on prevailing cultural values (Aaker 2000).
consumers. The best-known national-cultural systems include the
If the BRiC is high, brands are of greater relevance to frameworks proposed by Hofstede, Inglehart, Schwartz,
the consumer. According to the Elaboration Likelihood and Triandis (see Vinken, Soeters, and Ester 2004 for an
Model of persuasion (Petty and Cacioppo 1986), this overview). Given that brands are a key element, and car-
would suggest that the consumer is more likely to thought- rier, of the materialistic culture (Holt 2002; McCracken
fully follow brands’ product and ethics-related actions and 1986), Inglehart’s framework (Inglehart and Baker 2000;
communications (as well as assessing those actions against Inglehart and Welzel 2005), which is the only framework
brands’ promises) and weigh peripheral cues less—that is, that is explicitly grounded in (post)materialism, is the most
follow the central route of persuasion. On the other hand, if useful for our purposes. For previous marketing applica-
the BRiC is low, the Elaboration Likelihood Model sug- tions of Inglehart’s theory, see Steenkamp and de Jong
gests that the consumer relies more on peripheral cues. (2010) and Steenkamp and Geyskens (2012, 2014).
Note that the Elaboration Likelihood Model parallels Inglehart identifies two bipolar cultural dimensions: tra-
Epstein et al.’s (1996) two thinking styles with the impor- ditional versus secular-rational values, and survival versus
tant difference that the Elaboration Likelihood Model self-expression values. Countries that score high on the
makes situation-specific predictions. secular-rational dimension are characterized by
Yet there is a compelling rival theory on situational in-
formation processing: Accessibility-Diagnosticity Theory 3 Moreover, Menon and Raghubir posit that accessibility of an input
can “be used as a reasonable proxy for the diagnosticity of the input”
(Feldman and Lynch 1988), where the focus is less on cen- (Menon and Raghubir 2003, 231). Thus, factors that increase accessi-
tral versus peripheral information than on the accessibility bility of an input also influence attitude by increasing diagnosticity
of information. According to Accessibility-Diagnosticity value of that input.
materialistic secularism (Inglehart and Welzel 2005, 26, TABLE 2

31). Brands—as one of the most visible exponents of a ma- DIRECTIONAL EFFECTS
terialistic world (McCracken 1986)—are expected to be of
greater relevance in these societies (Rindfleisch, Variable Expected direction
Burroughs, and Wong 2009; Steenkamp and de Jong
Marketing-mix activities
2010). Applying the tenets of Accessibility-Diagnosticity Advertising þ
Theory in this context suggests that brand-related New product introductions þ
marketing-mix activity is more accessible to consumers in Distribution þ
countries high on secular-rational values and hence more Price þ
Price promotion –
likely to be used in attitude formation. The Elaboration
Likelihood Model predicts the opposite effect. Because of Multiplier effects
Consumer
the greater relevance of brands in secular-rational coun-

Marketing-mix activities Strengthens marketing-mix
tries, consumers are less likely to use peripheral cues such extraversion effectsa
as marketing-mix variables in attitude formation. Marketing-mix activities Strengthens marketing-mix
According to Inglehart (Inglehart and Welzel 2005, 115), agreeableness effects
Marketing-mix activities Weakens marketing-mix effectsb
“Materialist/postmaterialist values are a key component of conscientiousness
the survival/self-expression dimension.” Countries that score Marketing-mix activities Weakens marketing-mix effects
high on the self-expression dimension are characterized by a openness to experience
relative deemphasis of materialistic ideologies and emerging Category
post-materialist ideologies. In these societies, “the ‘quality Marketing-mix activities Weakens (ELM)/strengthens
of experience’ replaces the quantity of commodities as the brand relevance in a (ADT) marketing-mix effects
category
prime criterion for making a good living” (Inglehart and
Welzel 2005, 25). Maximizing well-being rather than maxi- Culture
Marketing-mix activities Weakens (ELM)/strengthens
mizing material possessions becomes a guiding motivation secular-rational culture (ADT) marketing-mix effects
to people, and their interest in the marketplace for achieving Marketing-mix activities Strengthens (ELM)/weakens
life goals declines. Postmaterialist priorities are associated self-expression culture (ADT) marketing-mix effects
with reduced importance of brands (Holt 2002). According NOTE.—ELM ¼ Elaboration Likelihood Model; ADT ¼ Accessibility-
to Accessibility-Diagnosticity Theory, in countries low on Diagnosticity Theory.
a
self-expression values (i.e., high on survival values), That is, – for price promotion, þ for the other marketing-mix activities
b
That is, þ for price promotion, – for the other marketing-mix activities
marketing-mix activities are expected to be more accessible
to consumers and as such have greater impact on consum-
ers’ attitude and their trust in brands. The Elaboration nine developed countries (Denmark, France, Germany,
Likelihood Model again offers the opposite perspective. Great Britain, Italy, Netherlands, Spain, Sweden, and the
Table 2 gives an overview of the directional effects. United States) and four emerging markets (Brazil, China,
India, and Russia). In each country, respondents—the per-
Sociodemographics son in the household that was responsible for grocery pur-
chases—answered questions regarding a maximum of
Four sociodemographics are included in our framework three brands of consumer product goods in a product cate-
(figure 2) and analyses. Although they are not the focus of gory. The selected brands were the top three national
this study, CTB could systematically vary across age, gen- brands in their category in 2013 (based on annual volume
der, social class, or education of the respondent. We cannot market share). The total number of different product cate-
make predictions as to their likely effect since most previ- gories included in the survey was 46. The specific catego-
ous research used student samples (table 1). However, fac- ries included varied across countries to reflect usage
tual findings for these variables might be of interest to patterns and the needs of GfK and Kantar Worldpanel.
applied researchers. Moreover, controlling for these varia- The questionnaire was developed in English and trans-
bles provides for a stronger test of our focal effects. lated into local languages using the back-translation
method. Respondents answered questions regarding the
METHOD marketing activities of a brand and their trust in it. In the
survey, advertising and new product introduction intensity
Data were operationalized with two items each, using items de-
We combine consumer survey data, scanner data, and veloped by Steenkamp and Geyskens (2014). CTB was
country data to examine the proposed main and moderating operationalized using two items drawn from Chaudhuri
effects. The individual-level survey data was collected via and Holbrook (2001). Respondents answered questions re-
the internet by the global market research agencies GfK garding BRiC with the four-item scale developed by
and Kantar Worldpanel in 2015 in 13 countries, including Fischer et al. (2010). The Big Five personality traits were
measured using items developed by Donnellan et al. Model and Estimation

(2006). Additionally, respondents reported their sociode-
Our model consists of variables at three levels: brand,
mographic information (i.e., gender, age, education, social
consumer, and country. We model consumers’ trust in
status). Social desirability responding tendencies were
brands as a function of marketing-mix instruments and
measured using items developed by Hays, Hayashi, and
their interactions with category and country-level modera-
Stewart (1989).
tors. Web appendix E gives details on model development.
We obtained household scanner data for all 13 countries
Our estimation equation is:
from GfK, Kantar Worldpanel, and IRI. Specifically, we
acquired average shelf price (price per volume for a brand),
distribution intensity (percentage of retailers that sold a CTBijk ¼ d000 þ d100 ADVijk þ d200 NPIijk þ d300 DISTik
brand, weighted by retailers’ annual market share), and Xp¼5
price promotion intensity (the brand’s annual value sold on þ d400 PRICEik þ d500 PROMik þ d0p0 PRSNpjk

p¼1
promotion divided by the brand’s total annual sales) during
X
q¼5 X
p¼4
2014. To render the measure for price comparable across þ dqp0 MKTMIXqijk PRSNpjk
categories, we compute z-scores for brand price based on q¼1 p¼1
the price of the top 10 national brands in each category. To X
q¼5
ensure temporal separation the scanner data are from 2014 þ d060 BRICjk þ dq60 MKTMIXqijk BRiCjk
so that they lag the brand trust measure collected in 2015. q¼1
Country data on Inglehart’s cultural values were Xq¼5
obtained from World Values Survey. We also obtained a þ d001 SECRATk þ dq01 MKTMIXqijk
measure of generalized societal trust from World Values q¼1
Survey to control for cross-country variations in disposi- SECRATk þ d002 SELFEXPRk
tion to trust. Variables and operationalizations are summa- X
q¼5
rized in table 3. þ dq02 MKTMIXqijk SELFEXPRk

We merged the scanner data with consumer survey data q¼1
to construct our final data set. Our final sample consisted X

p¼11 X
p¼15
of 35,028 observations from 15,073 respondents and 589 þ d0p0 SOCIOpjk þ d0p0 CATTYPEpjk
p¼7 p¼12
brands across 46 distinct CPG categories in 13 countries
(average of 26 categories in each country). Web appendix þ d003 STRk þ d004 EURk þ Wijk
B presents category-country combinations in our data set (1)
(as well as grouping them into low, medium, and high cate-
gories with respect to average BRiC ratings). We provide where i denotes the brands, j denotes the consumers, and k
examples of low, medium, and high average brand trust denotes the countries in our data. CTBijk denotes the trust
(compared to country mean) in web appendix C. that consumer j in country k has in brand i. ADVijk, NPIijk,
DIST , PRICEik, and PROMik (collectively
Pq¼5ik
q¼1 MKTMIX qijk ) refer to advertising intensity (q ¼ 1),
new product introduction intensity (q ¼ 2), distribution in-
Cross-National Measurement Validation tensity (q ¼ 3), price (q ¼ 4), and promotion intensity
Pp¼5
(q ¼ 5).5 p¼1 PRSNpjk denotes the Big Five personality
Following Steenkamp and Baumgartner (1998), first we
traits extraversion (p ¼ 1), agreeableness (p ¼ 2), conscien-
establish the cross-national invariance of measurement
tiousness (p ¼ 3), openness to experience (p ¼ 4), and neu-
instruments. Results of the measurement invariance analy-
roticism (p ¼ 5). BRiCjk represents consumer j’s reliance
ses (reported in web appendix D) support metric and scalar
on brands in category k. SECRATk and SELFEXPRk refer
invariance for CTB and metric invariance for the survey-
to the secular-rational and self-expression dimensions, re-
based marketing-mix instruments, BRiC, personality traits,
spectively. We include several sociodemographic variables
and social desirability responding.4 Figure 3 shows country Pp¼11
means for CTB, with their 95% confidence intervals. The ( p¼7 SOCIOpjk ) to control for heterogeneity across con-
three countries where mean CTB is highest—Brazil, India, sumers. The SOCIO variable captures gender (p ¼ 7), age
and China—are all emerging markets. Noteworthy is that (p ¼ 8), education (p ¼ 9), social class (p ¼ 10), and social
the US is significantly higher on CTB than any other devel- desirability responding tendency of the consumer (p ¼ 11).
P
oped market. We include four category dummies ( p¼15 p¼12 CATTYPEpjk )
to account for five different types of product categories:
4 Since we do within-country mean-centering for our predictor varia-
bles in our main analysis, establishing scalar invariance is not required
for advertising, new product introduction, brand relevance in a 5 DISTik, PRICEik, and PROMik do not vary across survey respond-
category (BRiC), personality traits, and social desirability responding. ents (hence, they have no j subscript).
TABLE 3
VARIABLES AND DESCRIPTIONS
Variable Operationalization Reference Source

Consumers’ trust in brands 1) Brand m is a brand I trust. 2) Brand m delivers what Chaudhuri and Holbrook (2001) Consumer surveys
(a ¼.79) [CTB] it promises.
Advertising intensity 1) Brand m is heavily advertised in newspapers, maga- Steenkamp et al. (2010) Consumer surveys
(a ¼.87) [ADV] zines, TV, or internet. 2) Brand m advertises a lot.
New product introduction 1) Brand m frequently introduces new products. 2) Steenkamp et al. (2010) Consumer surveys
intensity Brand m has many new product introductions.
(a ¼.84) [NPI]
Distribution intensity Percentage of retailers that sold brand m during a Sotgiu and Gielens (2015) Scanner data
[DIST] year, weighted by retailers’ market shares in the pre-

vious year.
Price Value sales of brand m divided by its volume sales, av- Sotgiu and Gielens (2015) Scanner data
[PRICE] eraged over all purchase occasions, in the previous
year. For comparability across categories and coun-
tries, based on price of the top 10 brands in category
n, we computed z-scores for brand prices.
Price promotion intensity Total absolute value sales sold on promotion by brand Sotgiu and Gielens (2015) Scanner data
[PROM] m, divided by total absolute value sold by brand m,
per year (in the previous year).
Brand relevance in a 1) In category n the brand plays—compared to other Fischer et al. (2010) Consumer surveys
category things—an important role. 2) In category n I focus
(a ¼.89) [BRiC] mainly on the brand. 3) In category n it is important
to purchase a brand name product. 4) In category n
the brand plays a significant role as to how satisfied I
am with the product.
Traditional versus secular- Country scores derived from responses to multiple Inglehart and Welzel (2005) WVS – Wave 5
rational values items in large representative surveys. Scores range
[SECRAT] from –2.0 to 2.0. Higher scores indicate a stronger
secular-rational culture.
Survival versus self-ex- Country scores derived from responses to multiple Inglehart and Welzel (2005) WVS – Wave 5
pression values items in large representative surveys. Scores range
[SELFEXPR] from –2.5 to 2.5. Higher scores indicate a stronger
self-expression culture.
Extraversion [PRSN1] I see myself as someone who 1) is the life of the party Donnellan et al. (2006) Consumer surveys
(a ¼.78) 2) talks a lot 3) talks to a lot of different people at
parties.
Agreeableness [PRSN2] I see myself as someone who 1) sympathizes with Donnellan et al. (2006) Consumer surveys
(a ¼.75) others’ feelings 2) feels others’ emotions 3) is really
interested in others.
Conscientiousness I see myself as someone who 1) gets chores done right Donnellan et al. (2006) Consumer surveys
[PRSN3] away 2) likes order 3) makes a mess of things (re-
(a ¼.65) verse-coded).
Openness to experience I see myself as someone who 1) has a vivid imagina- Donnellan et al. (2006) Consumer surveys
[PRSN4] tion 2) is interested in abstract ideas 3) has a good
(a ¼.71) imagination.
Neuroticism [PRSN5] I see myself as someone who 1) has frequent mood Donnellan et al. (2006) Consumer surveys
(a ¼.68) swings 2) is relaxed most of the time (reverse-
coded) 3) gets upset easily.
Gender [SOCIO1] What is your gender? Consumer surveys
Age [SOCIO2] What is your age? Consumer surveys
Education [SOCIO3] Which of these best describes your highest level of ed- Consumer surveys
ucation? (No formal education; Education up to age:
12, 14, 16, 18; Higher education; University)
Social class [SOCIO4] If people in our society are divided into upper, upper Consumer surveys
middle, middle, lower middle, working, and lower
classes, which class do you think you belong to?
SDR [SOCIO5] 1) I sometimes feel resentful when I don’t get my way Hays et al. (1989) Consumer surveys
(a ¼.64) 2) I sometimes try to get even rather than forgive 3)
There have been occasions when I took advantage
of someone.
TABLE 3 (CONTINUED)
Variable Operationalization Reference Source
Societal trust Self-reported trust in others, constructed as the per- WVS – Wave 5
[STR] centage of respondents answering yes to the ques-
tion “generally speaking, would you say that most
people can be trusted?”
Product category type General product category specification (0 ¼ food; 1 ¼ Consumer surveys
[CATTYPE] beverage; 2 ¼ household care; 3 ¼ personal care; 4
¼ animal food)
NOTE.—CTB, ADV, NPI, PRSN1–PRSN5, BRiC, and SDR were scored on a seven-point scale where 1 ¼ “very strongly disagree,” 2 ¼ “disagree,” 3 ¼
“somewhat disagree,” 4 ¼ “neither agree nor disagree,” 5 ¼ “somewhat agree,” 6 ¼ “agree,” and 7 ¼ “very strongly agree.” Our initial analysis showed that for per-
sonality traits and social desirability responding, some of the original items (which were reverse-coded) did not correlate well with other items or exhibited lack of
measurement variance across countries. We dropped those items.

FIGURE 3
MEAN CTB ACROSS THE COUNTRIES
beverage (p ¼ 12), personal care (p ¼ 13), household care d260, d360, d460, and d560 represent the moderating effect of
(p ¼ 14), and pet food (p ¼ 15), with food as the baseline BRiC on the sensitivity of CTB to the marketing activities.
category. We also include two country-level control varia- d101, d201, d301, d401, and d501 represent the moderating ef-
bles in our model: STRk and EURk. STRk captures general- fect of secular-rational values on the sensitivity of CTB to
ized trust in others in a country. In our data, we have nine the marketing-mix instruments. Finally, d102, d202, d302,
European countries; therefore, we include a dummy vari- d402, and d502 test the moderating impact of self-expression
able (i.e., EURk) to capture unobserved region-specific values on the sensitivity of CTB to the marketing-mix
effects. instruments.
d100, d200, d300, d400, and d500 representP
the main effect Wijk is the composite error term that includes a cross-
q¼5 Pp¼4
of marketing-mix instruments on CTB. q¼1 p¼1 dqp0 classified brand random effect mikbrand, which is normally
capture the moderating impact of the four personality traits distributed with zero mean and variance r22. mikbrand cap-
on the effectiveness of marketing activities on CTB. tures brand-specific unobserved heterogeneity that might
Retaining all 20 interactions between marketing-mix impact CTB. We use grand-mean centering for country-
instruments and personality traits might lead to multicolli- level variables and within-group centering for other varia-
nearity and unstable results. Therefore, we retain only bles that vary within consumers and across consumers.
interactions that are significant at the .10 level (see Since we examine relationships at multiple levels simulta-
Steenkamp and Geyskens 2014 for a similar practice). d160, neously, we use iterative maximum likelihood.
Common Method Bias and Endogeneity p < .001), distribution intensity (c300 ¼ .182, p ¼ .025),
and price (c400 ¼ .028, p ¼ .008) have a significant posi-
While we are interested in examining the effect of
tive effect on CTB, while price promotion intensity nega-
marketing-mix instruments on CTB, one could argue that
tively impacts CTB (c500 ¼ –.241, p ¼ 058). These results
the observed relationships between the marketing-mix
are consistent with our predictions regarding the effects of
instruments and CTB could be because the level of CTB
marketing activities on CTB.
influences managerial strategy in setting the level of
Figure 4 illustrates the magnitude of these effects by pre-
marketing-mix activities. For example, if price promotions
senting observed mean CTB scores for observations that
are used frequently by a brand in a particular country, is it
are at least one standard deviation above (below) the mean
because the brand has problems in that country (e.g., low
of the marketing-mix instrument in question.7 The largest
CTB), or did the price promotions reduce CTB? Moreover,
difference is found for new product introduction intensity.
there could be unobserved variables that influence both
CTB for brands low on this marketing-mix instrument is

marketing-mix activities and CTB (e.g., access to capital,
on average 4.41 versus an average CTB score of 5.93 for
managerial talent, suppliers, and social media capabilities).
brands high on this instrument. Cohen’s d for this effect is
Hence, the effect of marketing-mix instruments on CTB
1.44, which meets the cutoff for a large effect size.8
might be overstated if, for example, managerial talent
Advertising also has a large effect (Cohen’s d ¼ .80): CTB
drives both. Additionally, the same individuals who rate
for brands low (high) on advertising is on average 4.56
CTB also rate advertising intensity, new product introduc-
(5.47). Price has a small effect (d ¼ .23), while the effects
tion intensity, and BRiC. This could lead to common
of distribution (d ¼ .15) and price promotion (d ¼ .09) are
method bias.
below the cutoff for a small effect size.
In order to address the common method bias and endo-
geneity concerns, we generate instrumental variables. We
exploit the multimarket nature of our data set to construct The Moderating Role of Personality Traits
valid Hausman-style instruments. We obtain meaningful Consistent with our earlier discussion, we find that the
instrumentation for marketing-mix instruments by using a effect of advertising (c110 ¼ .010, p ¼ .003) and new prod-
brand’s average marketing-mix values in the same cate- uct activity (c210 ¼ .006, p ¼ .088) is larger for consumers
gory across other countries (see web appendix F for a de- high on extroversion. Moreover, the effect of advertising is
tailed description of our IV approach). Similarly, for BRiC, smaller for consumers that are higher on openness to expe-
we use average values across consumers in other countries. rience (c140 ¼ –.017, p < .001), which is in line with our
The instruments are sufficiently strong, as evidenced by predictions. Mixed support is found for conscientiousness.
the first-stage R-squared and F-statistics. Across the six Consistent with our expectations, the effect of advertising
scenarios, we obtain an average R-squared of 30%, and all is lower for consumers high on this trait (c130 ¼ –.014, p ¼
incremental F-values exceed the common threshold of 10 .001). However, new product activity has a more positive
(on average, the incremental F-values are 3,021). We esti- effect on CTB for more conscientious consumers (c230 ¼
mate six predicted residuals from the first-stage regressions .020, p < .001). This might be explained by the cognitive
and then add the estimated residuals as control functions to nature of new product activity. By their very nature, new
the main model (Petrin and Train 2010). The instruments products introduce some new information in the market,
that we use are obtained from responses by other respond- which takes some effort to process. Conscientious people
ents in different countries. Therefore, our instruments not may be more prone to do that and, based on their process-
only address the endogeneity concerns, but also account ing of this information, feel that the brand is trying to im-
for the common method bias. prove and better meet consumer needs, which increases
brand trust. Finally, we find that advertising has a larger ef-
fect on CTB for more agreeable consumers (c120 ¼ –.010,
RESULTS p ¼ .009), which contradicts our prediction. Steenkamp
Main Effects of Marketing-Mix Activities and Maydeu-Olivares (2015) argued that individuals high
on agreeableness have more positive attitudes toward ad-
Parameter estimates for the model specified in equation 1 vertising. This might explain the positive effect we find in
are reported in table 4.6 Note that we report unstandardized this study.
coefficients. In multilevel modeling, standardized coeffi- To get a sense of the magnitude of these interaction
cients are problematic because variance is partitioned effects, figure 5 (panel A) provides the observed mean
across different levels. Advertising intensity (c100 ¼ .063,
p ¼ .037), new product introduction intensity (c200 ¼ .381, 7 Everywhere in the results section, when we talk about high versus
low values on a variable, it refers to one standard deviation above or
6 Out of the 20 interactions between marketing-mix instruments and below the mean.
personality traits, we retain and report only those interactions that are 8 The cutoff value for a small, medium, and large effect size is .2, .5,
significant at p < .10. and .8, respectively (Cohen 1988).
TABLE 4
RESULTS
Covariate Parameter Expected sign Estimate p-value
Intercept c000 5.347 < .001

Main effects marketing-mix activities
Advertising intensity (ADV) c100 þ .063 .037
New product introduction intensity (NPI) c200 þ .381 < .001
Distribution intensity (DIST) c300 þ .182 .025
Price (PRICE) c400 þ .028 .008
Price promotion intensity (PROM) c500 – –.241 .058
Interactions with personality traits
Extraversion ADV c110 þ .010 .003

Extraversion NPI c210 þ .006 .088
Agreeableness ADV c120 þ –.010 .009
Conscientiousness ADV c130 – –.014 .001
Conscientiousness NPI c230 – .020 < .001
Openness to experience ADV c140 – –.017 < .001
Interactions with brand relevance in category
BRiC ADV c160 ELM(–), ADT(þ) .021 < .001
BRiC NPI c260 ELM(–), ADT(þ) .017 < .001
BRiC DIST c360 ELM(–), ADT(þ) .063 .014
BRiC PRICE c460 ELM(–), ADT(þ) .015 < .001
BRiC PROM c560 ELM(þ), ADT(–) .007 .839
Interactions with national culture
SECRAT ADV c101 ELM(–), ADT(þ) –.011 .364
SECRAT NPI c201 ELM(–), ADT(þ) .076 < .001
SECRAT DIST c301 ELM(–), ADT(þ) .017 .919
SECRAT PRICE c401 ELM(–), ADT(þ) .057 .001
SECRAT PROM c501 ELM(þ), ADT(–) –.733 .006
SELFEXPR ADV c102 ELM(þ), ADT(–) –.061 < .001
SELFEXPR NPI c202 ELM(þ), ADT(–) –.043 < .001
SELFEXPR DIST c302 ELM(þ), ADT(–) .141 .180
SELFEXPR PRICE c402 ELM(þ), ADT(–) –.026 .071
SELFEXPR PROM c502 ELM(–), ADT(þ) .121 .573
Sociodemographics and controls
Gender (male: 0, female: 1) c070 –.026 .056
Age c080 –.002 .008
Education (in years) c090 –.006 .184
Social class c0100 .002 .642
Extraversion c010 .004 .509
Agreeableness c020 .084 < .001
Conscientiousness c030 .078 < .001
Openness to experience c040 .038 < .001
Neuroticism c050 –.031 < .001
Brand relevance in category (BRiC) c060 .191 < .001
Secular-rational culture (SECRAT) c001 –.123 .331
Self-expression culture (SELFEXPR) c002 –.116 .041
Socially desirable responding tendency c0110 .008 .130
Generalized societal trust (STR) c003 –.001 .730
European countries dummy (EUR) c004 –.441 < .001
Category type (CATTYPE) c0120 –c0150 Included
Six control functions and brand random effect Included
NOTE.—N ¼ 35,028; p-values are one-sided for hypothesized effects and two-sided for others; ELM ¼ Elaboration Likelihood Model; ADT ¼ Accessibility-
Diagnosticity Theory.
CTB score for low versus high level of the marketing-mix low on extraversion (d ¼ .55) and high on extraversion (d
instrument in question—and the associated Cohen’s d—for ¼ .92) for a difference of .37 indicates that the interaction
low versus high value of the moderator. To assess the ef- is of small-medium effect size. The interaction between ex-
fect size of the interaction, the difference between the two traversion and new product activity is associated with Dd
Cohen’s ds (Dd) is informative. The figure shows that the of .33. The only other interaction that meets the cutoff for
strongest effect sizes are associated with extraversion. a small effect size is between advertising and openness to
Comparing the effect of advertising between consumers experience (Dd ¼ .21).
The Moderating Role of Brand Relevance in a exceed the cutoff for a small effect size. The strongest
Category effects by far are found for the interactions between self-
expression values and advertising intensity (Dd ¼ .72) and
We proposed rival explanations regarding the moderat- new product activity (Dd ¼ .75).
ing role of BRiC as Accessibility-Diagnosticity Theory and
the Elaboration Likelihood Model of persuasion posit dia-
metrically opposing effects. We find that advertising inten- Sociodemographics and Control Variables
sity (c160 ¼ .021, p < .001), new product introduction We find that women have lower trust in brands (c070 ¼
intensity (c260 ¼ .017, p < .001), distribution intensity –.026, p ¼ .056) and that CTB declines with age (c080 ¼
(c360 ¼ .062, p ¼ .014), and price (c460 ¼ .015, p < .001) –.002, p ¼ .007). We also find that CTB is lower in
have a stronger impact on CTB when BRiC is high European countries (c004 ¼ –.441, p < .001). This finding
compared to when it is low. The effect for price promotion is in line with the fact that private labels command a larger

is not significant. Thus, we find strong support for market share in Europe than in any other continent.
Accessibility-Diagnosticity Theory’s predictions. Figure 5 We included the main effects of our moderators for
(panel B) again provides observed mean CTBs. It high- proper interpretation of the interactions. Although they are
lights that the interaction effects are mostly substantial. For not the focus of our study, they yield some interesting
advertising, Dd ¼ .69, a medium-large effect, for new results. We find that people high on the trait of agreeable-
products and price, the effect is close to medium (Dd ¼ ness exhibit higher CTB (c020 ¼ .084, p < .001). This is
.45). The effect for distribution is more modest: Dd ¼ .28. consistent with past research that identified the trait of trust
as a subscale of agreeableness (McCrae and Costa 2008).
The Moderating Role of National Culture We also find that consumers scoring high on conscientious-
ness (c030 ¼ .078, p < .001) and openness to experience
We proposed rival predictions regarding how the secular- (c040 ¼ .038, p < .001) are higher on CTB, while neuroti-
rational and self-expression cultural dimensions moderate cism has a negative effect on CTB (c050 ¼ –.031, p <
the sensitivity of CTB to the marketing-mix activities. .001). This finding is consistent with the notion that an in-
Consistent with the arguments of Accessibility- dividual holding a neurotic personality has a pessimistic at-
Diagnosticity Theory, we find that higher new product titude toward individuals and objects, which makes it
introduction intensity (c201 ¼ .076, p < .001), higher price difficult for him/her to trust individuals/entities (Tan and
(c401 ¼ .057, p ¼ .001), and lower price promotion intensity Sutherland 2004).
(c501 ¼ –.733, p ¼ .006) have a stronger impact on CTB in The main effect of BRiC is positive and significant (c060
countries high on secular-rational values compared to coun- ¼ .191, p < .001), suggesting that CTB is higher when in a
tries low on secular-rational values. The interactions for ad- particular product category, and brands are more important
vertising and distribution intensity are not significant. to a consumer. We find that consumers have lower trust in
We find significant interactions for self-expression cul- brands in countries high on self-expression values (c002 ¼
ture in the direction specified by Accessibility- –.116, p ¼ .041). This finding is consistent with previous
Diagnosticity Theory, albeit not for all marketing-mix research, which argued that brands are expected to do
instruments. Consistent with Accessibility-Diagnosticity worse in postmaterialistic countries (Holt 2002; Steenkamp
Theory, advertising intensity (c102 ¼ –.061, p < .001), new and Geyskens 2014).
product introduction intensity (c202 ¼ –.043, p < .001),
and price (c402 ¼ –.026, p ¼ .071) have a weaker impact
on CTB in countries high on self-expression values. The
Robustness Checks
other two interactions are not significant. We conducted a series of checks to assess the robustness
Figure 5 (panel C) depicts mean CTBs associated with of our findings, including analyses with median split mod-
the significant interaction effects. Four interaction effects erators, and models with manufacturer fixed effects and
FIGURE 4
MEAN CTB AT VARYING LEVELS OF MARKETING-MIX ACTIVITIES

brand fixed effects. We also assessed our model’s out-of- sophisticated, and expensive-looking supermarkets (Child,
sample predictive power across three sets of analyses. The Kilroy, and Naylor 2015). Such outlets have more charac-
results are reported in web appendixes G–I. The overall teristics of expensive brand-specific capital expenditures.
conclusion is that our model findings exhibit a high degree We use the richness of our data set to explore possible
of robustness. differences in the effects of marketing-mix activities on
CTB across different country-level categorizations (e.g.,
Hofstede’s cultural dimensions, country’s religion, coun-
Other Country-Level Categorizations try’s economic well-being) in an exploratory fashion. We
While we focus on country differences using Inglehart’s use median values to categorize countries into low and
cultural framework, another, managerially more interest- high on different factors and run separate analyses on
ing, classification is between developed and emerging mar- countries that are above versus below the median on a par-
kets (Burgess and Steenkamp 2006). Our set of countries ticular characteristic, with only the marketing-mix activi-

includes the four most important emerging markets: Brazil, ties as regressors. We report the results of those analyses in
China, India, and Russia. We find that the signaling value web appendix J. The results confirm the importance of new
of advertising and price is stronger in emerging markets product introduction intensity as the most important driver
than in developed markets (see Table 5). Compared to de- of CTB; its effect on CTB is significant across all country-
veloped markets, brands have not been around as long in level categorizations. Advertising is also generally an im-
emerging markets; hence, knowledge about products and portant driver of CTB; however, in certain contexts it does
brands is generally less deep (Burgess and Steenkamp not significantly influence CTB (e.g., countries low on un-
2006). In these contexts, advertising fulfills a more impor- certainty avoidance or low on power distance). Overall, the
tant role in creating brand awareness and communicating results highlight interesting patterns that could be of inter-
the brand message (Pauwels, Erguncu, and Yildirim 2013). est to marketing managers and researchers. Discussing the
The larger role of price in emerging markets is consistent possible underlying mechanisms behind these results is
with recent research that showed that while consumers in outside the scope of this article, but our findings may offer
emerging markets are more price conscious, they also rely inspiration for future research.
more on price as an indicator to infer product performance
(Zielke and Komor 2015). Distribution has a large effect DISCUSSION
on CTB in developed markets but not in emerging markets.
This may be due to the fact that in emerging markets, infor- The brand value chain (Keller and Lehmann 2003) has
mal distribution and small, relatively unsophisticated emerged as a useful framework to understand the chain of
mom-and-pop shops play a large role, while in developed events through which brands create value. The key first
markets, brands are almost exclusively offered in large, link in this framework is that marketing activity of the firm
FIGURE 5
MODERATING ROLE OF PERSONALITY TRAITS, BRAND RELEVANCE IN CATEGORY, AND NATIONAL CULTURE
TABLE 5 In our theorizing, we discussed rival predictions for

THE EFFECT OF MARKETING-MIX ON CTB ACROSS BRiC and national culture, based on Accessibility-
Diagnosticity Theory and the Elaboration Likelihood
Emerging Developed p-value Model of persuasion. Our findings across categories and
Markets Countries (difference) countries consistently supported Accessibility-
Advertising .133 (< .001) .034 (.039) .018
Diagnosticity Theory. This is interesting; after all, both are
New products .419 (< .001) .417 (< .001) .966 established theories of attitude formation. Yet they could
Distribution –.037 (.651) .270 (< .001) .001 be reconciled if, for consumers, firms’ marketing activities
Price .045 (.133) .007 (.523) .234 contain real information that requires some thoughtful con-
Price promotion –.271 (.179) –.308 (.009) .872
sideration. If consumers are economic experts, they would
NOTE.—Numbers in parentheses are two-sided p-value. be aware of the information content of the market signals.
They would intuitively understand that if firms renege on

their promises, they would lose repeat business, and hence,
influences the consumer mindset with respect to the brand. the signals are valid. There are indirect indications that this
If the firm is ineffective in affecting the customer mindset might be the case. The two marketing-mix activities for
with its marketing, the subsequent links in the framework which the strongest interactions were consistently found
become moot. Our study elaborates and extends the brand are advertising and new product activity. For advertising,
value chain. First, it expands the framework by introducing Kirmani (1990) and Kirmani and Wright (1989) provide
brand trust as an important customer mindset element. evidence that consumers may indeed link adverting to
CTB has become a top managerial concern and has also quality and sales potential. And for many consumers, new
spawned a considerable amount of academic research, doc- product activity is important, as it can fill relevant niches
umenting its effect on various brand performance metrics. in the marketplace (Schmalensee 1978). According to
Second, we extend the brand value chain by identifying the Nielsen (2015), almost two-thirds of respondents in a
context in which the firm operates as a key multiplier, in- global survey said they like it when manufacturers intro-
tervening between marketing activity and customer mind- duce new products, and more than half said they purchased
set. We specify three context domains: consumer, a new product during their last shopping trip.
category, and country. We develop arguments regarding We find a strong multiplier role for Inglehart’s theory of
the main effects of five marketing-mix activities on CTB national culture. So far, his theory has received little atten-
as well as the multiplier effects of personality traits (con- tion in the literature. Yet his work is rooted in (post)materi-
sumer), brand relevance in a category (category), and alism and (post)modernity, which are concepts that are of
secular-rational and self-expression values (country). The great importance to consumer researchers. We believe his
effects were examined using a dedicated data set that com- theory deserves more attention in marketing and consumer
bined consumer surveys, household scanner data, and behavior.
country data across 589 brands in 46 CPG categories,
across 13 countries, which collectively account for half of Managerial Implications
the world’s population. The cross-national data used in this
study provide a strong test of the generalizability of the The brand value chain is a structured means for manag-
findings. Consistent with the expanded brand-value-chain ers to understand where and how value is created and sug-
framework, marketing-mix activity affected CTB, and the gests where to look to improve that process. According to
judgment context acted as a systematic multiplier, weaken- Keller and Lehmann (2003), brand and category marketing
ing or strengthening the effects of marketing program managers are likely to be particularly interested in the cus-
investments on CTB. tomer mindset and the impact of the marketing program on
We find that, with the exception of extraversion, the customers. They create value through smart investments in
multiplying role of consumers’ personality traits was small their marketing program and by maximizing the multiplier
to negligible, and also not always consistent with their effects to the extent possible. So, what are smart invest-
overall thinking styles. On the other hand, brand relevance ments, from the perspective of maximizing CTB (i.e., allo-
in a category (BRiC) had a much larger multiplier effect. cating more resources to advertising and new product
This suggests that overall thinking styles are less useful to activity)? These two marketing activities have a large ef-
understand the use of market signals in attitude formation fect (in Cohen’s sense) on CTB, while the effect of the
than category-specific measures. This does not detract other three instruments is small at best. Regarding advertis-
from the theory of thinking styles but suggests that to un- ing, this gives brand managers additional leverage to make
derstand formation of attitudes toward specific objects (in their case that it should not be evaluated only on sales lift.
our case, trust in specific brands), we need more specific Advertising builds brand trust, and this message should
measures. resonate with the C-suite, as brand trust is top of mind for
many CEOs (Consumer Goods Forum 2015). Our findings
also support continued investments in innovation by firms, Marke garantiert den Unterschied” [the brand guarantees
which is contrary to industry practice, at least for large the difference] and “Das Original: Achten Sie auf die
firms (Steenkamp and Sloot 2019). Marke” [The original: Pay attention to the brand]. The mo-
We further find that the effectiveness of these two major tivation for this campaign was that “many consumers think
trust-building marketing activities is substantially (in terms that PLs and NBs are actually the same product, only in
of effect size) moderated by a person’s degree of extraver- different packaging.” The campaign has since been
sion, category BRiC, and country espousal of self- adopted by other European associations of brand
expression values. Segmenting the market on a personality manufacturers.
trait like extraversion is possible, but the likelihood of suc- Using advertising to increase BRiC kills multiple birds
cess is probably not high (Wedel and Kamakura 1998, 16). with one stone. First, it is yet another way in which adver-
National culture is a given, too. In this sense, whether these tising contributes to the brand value chain. Second, BRiC
multipliers inhibit or facilitate value creation may be itself has a direct effect on CTB. Third, higher BRiC (ei-

largely out of the hands of the marketer. Yet, as argued by ther via increased CTB or directly; this has not been
Keller and Lehmann (2003, 31), recognizing their uncon- researched yet) is associated with improved brand perfor-
trollable nature is important to help put in perspective the mance, especially price premiums and brand equity
relative success or failure of trust-building programs. (Fischer et al. 2010).
Marketers cannot be logically held accountable for context
multipliers that they cannot influence. So, if marketers in a
brand strategy session are comparing trust in the firm’s
brand across countries, and they find that CTB is lower in
Limitations and Further Research
the US than in China, before concluding that the US man- Our empirical setting casts a wide net, leading to empiri-
ager is not doing their job, they need to correct scores to cal generalizations. Yet our data are not without limita-
account for the fact that the US rates high on self-expres- tions, which offers three lines of further inquiry. First, our
sion, while China rates low. Our parameter estimates and data are cross-sectional. An interesting question is whether
the publicly available country scores on the Inglehart and how CTB changes over time. Addressing this question
dimensions can be used to make the necessary correction. requires longitudinal data with repeated measurements of
However, creative managers can do something about CTB on the same people. This allows one to estimate latent
brand relevance in a category, either individually or via change trajectories (Steenkamp and Maydeu-Olivares
trade organizations. The two constituent components of 2015) and relate the parameters of change to marketing
BRiC are risk reduction and social demonstrance (Fischer program investments. Second, field experiments or lab
et al. 2010). Advertising can play a major role to highlight experiments can be used for a more detailed causal expla-
the adverse consequences of not selecting your brand—re- nation of the observed regularities in our study. Lab experi-
jection by peers (social demonstrance) or disappointing ments would specifically be helpful in determining the
product performance.9 This will be most effective if the underlying mechanisms at play. Researchers can test the
market leader or multiple brands in the same category do accessibility-diagnosticity mechanism against alternative
this. In many product categories, and certainly in the CPG explanations to detect the underlying mechanism causing
industry studied in this article, private labels are the main the observed relationships between marketing-mix activi-
homogenizing factor reducing BRiC. Individual or collec- ties and CTB. We speculated that perhaps the Elaboration
tive actions to differentiate the brand (brands) from private Likelihood Model and Accessibility-Diagnosticity Theory
labels, then, can help increase BRiC. An example of indi- might be reconcilable if, for consumers, firms’ marketing
vidual brand action is to communicate to consumers that activities contain real information that requires some
the brand does not manufacture private labels. Tylenol thoughtful consideration. Future research should investi-
runs TV ads in which employees make the following prom- gate this possibility. Third, in our research, advertising in-
ise: “We don’t make store brand pain relievers. We make tensity and new product introduction intensity were
Tylenol.” Pledge announces on its packaging in red, bold, operationalized using survey items developed and vali-
capital letters that “THIS FORMULA IS NOT SOLD TO dated in past research (Steenkamp, van Heerde, and
ANY RETAILER AS A STORE BRAND.” An example of Geyskens 2010; Steenkamp and Geyskens 2014). Future
collective action is the long-running advertising campaign research should examine whether these findings are repli-
run by the Austrian Association of Brand Manufacturers cated with secondary data.
(which counts companies like Mars, P&G, and Colgate- The brand-value-chain model includes program quality
Palmolive as members), which uses slogans like “Die multipliers, which were not considered in this study. It
remains to be seen whether program quality multipliers in-
9 We focus on advertising, as Fischer et al. (2010) documented that
teract with the context multipliers we have introduced in
heavy advertising is associated with higher BRiC. New product activ- strengthening or weakening the effect of marketing-mix
ity may be another way to increase BRiC. activities.
In our study, we find that with the exception of extraver- Blattberg, Robert C. and Scott A. Neslin (1989), “Sales
sion, the moderating role of personality traits is very small. Promotion: The Long and the Short of It,” Marketing Letters,
This raises several questions. First, what is special about 1 (1), 81–97.
Boulding, William and Amna Kirmani (1993), “A Consumer-Side
extraversion that underlies this fairly strong effect? Experimental Examination of Signaling Theory: Do
Second, might direct measures of thinking style (Pacini Consumers Perceive Warranties as Signals of Quality?”
and Epstein 1999) yield stronger results? Third, might con- Journal of Consumer Research, 20 (1), 111–23.
sumer traits that are conceptually related to the Big Five Burgess, Steven M. and Jan-Benedict E. M. Steenkamp (2006),
(Steenkamp and Maydeu-Olivares 2015) be more pertinent “Marketing Renaissance: How Research in Emerging Markets
consumer multipliers than personality traits? Future re- Advances Marketing Science and Practice,” International
search should look into these questions. Journal of Research in Marketing, 23 (4), 337–56.
Campbell, Margaret C. and Kevin Lane Keller (2003), “Brand
In information economics theory, it is crucial that con- Familiarity and Advertising Repetition Effects,” Journal of
sumers observe brand-specific investments. The brands in Consumer Research, 30 (2), 292–304.

our study were the largest brands in their category. It is Chaudhuri, Arjun and Morris B. Holbrook (2001), “The Chain of
likely that their marketing activity is more easily observ- Effects from Brand Trust and Brand Affect to Brand
able than that of minor brands. For example, past research Performance: The Role of Brand Loyalty,” Journal of
has showed that advertising intensity has differential Marketing, 65 (2), 81–93.
Child, Peter, Thomas Kilroy and James Naylor (2015), “Modern
impacts across well-known brands and lesser-known Grocery and the Emerging-Market Consumer: A
brands (Campbell and Keller 2003). Future research could Complicated Courtship,” McKinsey, http://bit.ly/1Uf9aBY.
extend our work by examining CTB and the role of Chu, Wujin and Woosik Chu (1994), “Signaling Quality by
marketing-mix activities therein for lesser-known brands. Selling through a Reputable Retailer: An Example of Renting
Finally, future research could extend empirical testing to the Reputation of Another Agent,” Marketing Science, 13
consumer durables. Do marketing-mix activities still play (2), 177–89.
the same trust-building role in these categories? We specu- Consumer Goods Forum (2015), “Trust Tops the CEO Agenda at
the Consumer Goods Forum Annual Summit,” April 28,
late that this will indeed be the case since the emergence of https://bit.ly/2Io2RjD.
e-WOM means that any attempt to renege on promises will Datta, Hannes, Kusum L. Ailawadi, and Harald J. van Heerde
quickly be known to multitudes of consumers who just en- (2017), “How Well Does Consumer-based Brand Equity
tered the market. We believe this topic requires further Align with Sales-based Brand Equity and Marketing-Mix
investigation. Response?” Journal of Marketing, 81 (3), 1–20.
Dawar, Niraj and Philip Parker (1994), “Marketing Universals:
Consumers’ Use of Brand Name, Price, Physical
Appearance, and Retailer Reputation as Signals of Product
Quality,” Journal of Marketing, 58 (2), 81–95.
Delgado-Ballester, Elena and Jose Luis Munuera-Aleman (2001),
DATA COLLECTION INFORMATION “Brand Trust in the Context of Consumer Loyalty,”
European Journal of Marketing, 35 (11/12), 1238–58.
The individual-level survey data that we used in this
Donnellan, M. Brent, Frederick L. Oswald, Brendan M. Baird, and
study was collected by the global market research agencies Richard E. Lucas (2006), “The Mini-IPIP Scales: Tiny-Yet-
GfK and Kantar Worldpanel in 2015. The household scan- Effective Measures of the Big Five Factors of Personality,”
ner panel data, which was used for measures on price, dis- Psychological Assessment, 18 (2), 192–203.
tribution, and price promotion, was collected by GfK, Edelman (2017), “2017 Edelman Trust Barometer,” January 21,
Kantar Worldpanel, and IRI. The authors acquired survey https://tinyurl.com/ybu9bh2j.
data and scanner panel data from AiMark. The first author Epstein, Seymour, Rosemary Pacini, Veronika Denes-Raj, and
Harriet Heier (1996), “Individual Differences in Intuitive–
conducted all empirical analyses in this study. Experiential and Analytical–Rational Thinking Styles,”
Journal of Personality and Social Psychology, 71 (2), 390–405.
REFERENCES Erdem, Tülin (1998), “An Empirical Analysis of Umbrella
Branding,” Journal of Marketing Research, 35 (3), 339–51.
Aaker, Jennifer L. (2000), “Accessibility or Diagnosticity? Erdem, Tülin and Joffre Swait (1998), “Brand Equity as a
Disentangling the Influence of Culture on Persuasion Signaling Phenomenon,” Journal of Consumer Psychology, 7
Processes and Attitudes,” Journal of Consumer Research, 26 (2), 131–57.
(4), 340–57. ———. (2004), “Brand Credibility, Brand Consideration, and
Batra, Rajeev, Aaron Ahuvia, and Richard P. Bagozzi (2012), Choice,” Journal of Consumer Research, 31 (1), 191–8.
“Brand Love,” Journal of Marketing, 76 (2), 1–16. Erdem, Tülin, Joffre Swait, and Jordan Louviere (2002), “The
Baumgartner, Hans (2002), “Toward a Personology of the Impact of Brand Credibility on Consumer Price Sensitivity,”
Consumer,” Journal of Consumer Research, 29 (2), 286–92. International Journal of Research in Marketing, 19 (1), 1–19.
Belk, Russell W. (1974), “An Exploratory Assessment of Erdem, Tülin, Joffre Swait, and Ana Valenzuela (2006), “Brands
Situational Effects in Buyer Behavior,” Journal of Marketing as Signals: A Cross-Country Validation Study,” Journal of
Research, 11 (2), 156–63. Marketing, 70 (1), 34–49.
Feldman, Jack M. and John G. Lynch Jr. (1988), “Self-Generated McCrae, Robert R. and Paul T. Costa Jr. (1990), Personality in
Validity and Other Effects of Measurement on Belief, Adulthood, New York: Guilford.
Attitude, Intention, and Behavior,” Journal of Applied ———. (2008), “The Five-Factor Theory of Personality,” in
Psychology, 73 (3), 421–35. Handbook of Personality: Theory and Research, 3rd ed., ed.
Fischer, Marc, Franziska Völckner, and Henrik Sattler (2010), Oliver P. John, Richard W. Robins, and Lawrence A. Pervin,
“How Important Are Brands? A Cross-Category, Cross- New York: Guilford, 159–81.
Country Study,” Journal of Marketing Research, 47 (5), Menon, Geeta and Priya Raghubir (2003), “Ease-of-Retrieval as
823–39. an Automatic Input in Judgments: A Mere-Accessibility
Hays, Ron D., Toshi Hayashi, and Anita L. Stewart (1989), “A Framework?” Journal of Consumer Research, 30 (2),
Five-Item Measure of Socially Desirable Response Set,” 230–43.
Educational and Psychological Measurement, 49 (3), Milgrom, Paul and John Roberts (1986), “Price and Advertising
629–36. Signals of Product Quality,” Journal of Political Economy,
Herbst, Kenneth C., Eli J. Finkel, David Allan, and Grainne M. 94 (4), 796–821.
Fitzsimons (2011), “On the Dangers of Pulling a Fast One: Millward Brown (2016), “BrandZ Top 100 Most Valuable Global

Advertisement Disclaimer Speed, Brand Trust, and Purchase Brands 2016,” https://bit.ly/2WOgdys.
Intention,” Journal of Consumer Research, 38 (5), 909–19. ——. (2018), “2018 BrandZ Top 100 Global Brands,” https://goo.
Herr, Paul M., Frank R. Kardes, and John Kim (1991), “Effects of gl/jiwvzJ.
Word-of-Mouth and Product-Attribute Information on Nanda, Rich (2014), “The 2014 American Pantry Study,” Deloitte,
Persuasion: An Accessibility-Diagnosticity Perspective,” https://tinyurl.com/y7yar3pz.
Journal of Consumer Research, 17 (4), 454–62. Nielsen (2015), “Looking to Achieve New Product Success?”
Holt, Douglas B. (2002), “Why Do Brands Cause Trouble? A June 23, https://goo.gl/U1d6Dj.
Dialectical Theory of Consumer Culture and Branding,” Pacini, Rosemary and Seymour Epstein (1999), “The Relation of
Journal of Consumer Research, 29 (1), 70–90. Rational and Experiential Information Processing Styles to
Hoyer, Wayne D. and Steven P. Brown (1990), “Effects of Brand Personality, Basic Beliefs, and the Ratio-Bias Phenomenon,”
Awareness on Choice for a Common, Repeat-Purchase Journal of Personality and Social Psychology, 76 (6),
Product,” Journal of Consumer Research, 17 (2), 141–8. 972–87.
Inglehart, Ronald and Wayne E. Baker (2000), “Modernization, Papatla, Purushottam and Lakshman Krishnamurthi (1996),
Cultural Change, and the Persistence of Traditional Values,” “Measuring the Dynamic Effects of Promotions on Brand
American Sociological Review, 65 (1), 19–51. Choice,” Journal of Marketing Research, 33 (1), 20–35.
Inglehart, Ronald and Christian Welzel (2005), Modernization, Park, C. Whan, Deborah J. MacInnis, Joseph Priester, Andreas B.
Cultural Change, and Democracy: The Human Development Eisingerich, and Dawn Iacobucci (2010), “Brand Attachment
Sequence, New York: Cambridge University Press. and Brand Attitude Strength: Conceptual and Empirical
Inman, J. Jeffrey, Leigh McAlister, and Wayne D. Hoyer (1990), Differentiation of Two Critical Brand Equity Drivers,”
“Promotion Signal: Proxy for a Price Cut?” Journal of Journal of Marketing, 74 (6), 1–17.
Consumer Research, 17 (1), 74–81. Pauwels, Koen, Selin Erguncu, and Gokhan Yildirim (2013),
Keller, Kevin Lane and Donald R. Lehmann (2003), “How Do “Winning Hearts, Minds and Sales: How Marketing
Brands Create Value?” Marketing Management, 12 (3), Communication Enters the Purchase Process in Emerging
26–31. and Mature Markets,” International Journal of Research in
———. (2006), “Brands and Branding: Research Findings and Marketing, 30 (1), 57–68.
Future Priorities,” Marketing Science, 25 (6), 740–59. Petrin, Amil and Kenneth Train (2010), “A Control Function
Kirmani, Amna (1990), “The Effect of Perceived Advertising Approach to Endogeneity in Consumer Choice Models,”
Costs on Brand Perceptions,” Journal of Consumer Research, Journal of Marketing Research, 47 (1), 3–13.
17 (2), 160–71. Petty, Richard E. and John T. Cacioppo (1986), “The Elaboration
Kirmani, Amna and Akshay R. Rao (2000), “No Pain, No Gain: A Likelihood Model of Persuasion,” in Advances in
Critical Review of the Literature on Signaling Unobservable Experimental Social Psychology, Vol. 19, ed. Leonard
Product Quality,” Journal of Marketing, 64 (2), 66–79. Berkowitz, New York: Academic Press, 123–205.
Kirmani, Amna and Peter Wright (1989), “Money Talks: Rao, Akshay R. and Humaira Mahi (2003), “The Price of
Perceived Advertising Expense and Expected Product Launching a New Product: Empirical Evidence on Factors
Quality,” Journal of Consumer Research, 16 (3), 344–53. Affecting the Relative Magnitude of Slotting Allowances,”
Klein, Benjamin and Keith B. Leffler (1981), “The Role of Market Marketing Science, 22 (2), 246–68.
Forces in Assuring Contractual Performance,” Journal of Rao, Akshay R. and Kent B. Monroe (1988), “The Moderating
Political Economy, 89 (4), 615–41. Effect of Prior Knowledge on Cue Utilization in Product
Kotler, Philip (2002), Marketing Management, Englewood Cliffs, Evaluations,” Journal of Consumer Research, 15 (2), 253–64.
NJ: Prentice-Hall. ———. (1989), “The Effect of Price, Brand Name, and Store Name
Li, Yan and Hongwei He (2013), “Evaluation of International on Buyers’ Perceptions of Product Quality: An Integrative
Brand Alliances: Brand Order and Consumer Review,” Journal of Marketing Research, 26 (3), 351–7.
Ethnocentrism,” Journal of Business Research, 66 (1), Riaz, Muhammad Naveed, Muhammad Akram Riaz, and Naila
89–97. Batool (2012), “Personality Types as Predictors of Decision
McCracken, Grant (1986), “Culture and Consumption: A Making Styles,” Journal of Behavioural Sciences, 22 (2),
Theoretical Account of the Structure and Movement of the 99–114.
Cultural Meaning of Consumer Goods,” Journal of Rindfleisch, Aric, James E. Burroughs, and Nancy Wong (2009),
Consumer Research, 13 (1), 71–84. “The Safety of Objects: Materialism, Existential Insecurity,
and Brand Connection,” Journal of Consumer Research, 36 Steenkamp, Jan-Benedict E. M., Harald van Heerde, and Inge
(1), 1–16. Geyskens (2010), “What Makes Consumers Willing to Pay a
Schmalensee, Richard (1978), “Entry Deterrence in the Ready-to- Price Premium for National Brands over Private Labels?”
Eat Breakfast Cereal Industry,” Bell Journal of Economics, 9 Journal of Marketing Research, 47 (6), 1011–24.
(2), 305–27. Sung, Yongjun and Jooyoung Kim (2010), “Effects of Brand
Scott, Susanne G. and Reginald A. Bruce (1995), “Decision- Personality on Brand Trust and Brand Affect,” Psychology &
Making Style: The Development and Assessment of a New Marketing, 27 (7), 639–61.
Measure,” Educational and Psychological Measurement, 55 Tan, Felix B. and Paul Sutherland (2004), “Online Consumer
(5), 818–31. Trust: A Multi-Dimensional Model,” Journal of Electronic
Sotgiu, Francesca and Katrijn Gielens (2015), “Suppliers Caught Commerce in Organizations, 2 (3), 40–58.
in Supermarket Price Wars: Victims or Victors? Insights Tse, David K., Kam-hon Lee, Ilan Vertinsky, and Donald A.
from a Dutch Price War,” Journal of Marketing Research, 52 Wehrung (1988), “Does Culture Matter? A Cross-Cultural
(6), 784–800. Study of Executives’ Choice, Decisiveness, and Risk
Srinivasan, Shuba, Koen Pauwels, Jorge Silva-Risso, and Adjustment in International Marketing,” Journal of

Dominique M. Hanssens (2009), “Product Innovations, Marketing, 52 (4), 81–95.
Advertising, and Stock Returns,” Journal of Marketing, 73 Vinken, Henk, Joseph Soeters, and Peter Ester (2004), Comparing
(1), 24–43. Cultures: Dimensions of Culture in a Comparative
Srinivasan, Shuba, Marc Vanhuele, and Koen Pauwels (2010), Perspective, Boston: Brill.
“Mind-Set Metrics in Market Response Models: An Wedel, Michel and Wagner A. Kamakura (1998), Market
Integrative Approach,” Journal of Marketing Research, 47 Segmentation: Conceptual and Methodological Foundations,
(4), 672–84. Boston: Kluwer Academic Publishers
Steenkamp, Jan-Benedict E. M. and Hans Baumgartner (1998), Wiener, Joshua Lyle (1985), “Are Warranties Accurate Signals of
“Assessing Measurement Invariance in Cross-National Product Reliability?” Journal of Consumer Research, 12 (2),
Consumer Research,” Journal of Consumer Research, 25 (1), 245–50.
78–90. Wilkie, William L. and Edgar A. Pessemier (1973), “Issues in
Steenkamp, Jan-Benedict E. M. and Martijn G. de Jong (2010), “A Marketing’s Use of Multi-Attribute Attitude Models,”
Global Investigation into the Constellation of Consumer Journal of Marketing Research, 10 (4), 428–41.
Attitudes toward Global and Local Products,” Journal of Yi, Youjae and Jaemee Yoo (2011), “The Long-Term Effects of
Marketing, 74 (6), 18–40. Sales Promotions on Brand Attitude across Monetary and
Steenkamp, Jan-Benedict E. M. and Inge Geyskens (2012), Non-Monetary Promotions,” Psychology & Marketing, 28
“Cultural Boundedness of Transaction Costs Economics: A (9), 879–96.
Test of Hypotheses Based on Inglehart and Hofstede,” Yoo, Boonghee, Naveen Donthu, and Sungho Lee (2000), “An
Journal of the Academy of Marketing Science, 40 (2), Examination of Selected Marketing-Mix Elements and Brand
252–70. Equity,” Journal of the Academy of Marketing Science, 28
———. (2014), “Manufacturer and Retailer Strategies to Impact (2), 195–211.
Store Brand Share: Global Integration, Local Adaptation, Young & Rubicam (2017), “The Decline of Trust,” October 9,
and Worldwide Learning,” Marketing Science, 33 (1), https://tinyurl.com/ya4d6uuu.
6–26. Zielke, Stephan and Marcin Komor (2015), “Cross-National
Steenkamp, Jan-Benedict E. M. and Alberto Maydeu-Olivares Differences in Price–Role Orientation and Their Impact on
(2015), “Stability and Change in Consumer Traits: Evidence Retail Markets,” Journal of the Academy of Marketing
from a 12-Year Longitudinal Study, 2002–2013,” Journal of Science, 43 (2), 159–80.
Marketing Research, 52 (3), 287–308.
Steenkamp, Jan-Benedict E. M. and Laurens Sloot (2019), Retail
Disruptors: The Spectacular Rise and Impact of the Hard
Discounters, London: Kogan Page.

In Brands We Trust? A Multicategory, Multicountry Investigation of Sensitivity of Consumers' Trust in Brands To Marketing-Mix Activities

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

In Brands We Trust? A Multicategory, Multicountry Investigation of Sensitivity of Consumers' Trust in Brands To Marketing-Mix Activities

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

In Brands We Trust? A Multicategory, Multicountry Investigation of Sensitivity of Consumers' Trust in Brands To Marketing-Mix Activities

Uploaded by

Copyright:

Available Formats

In Brands We Trust?

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

The essence of a brand is that it delivers on its promises. However, consumers’

Keywords: consumer trust, brand trust, marketing mix, branding, information

Koushyar Rajavi (Koushyar.rajavi@scheller.gatech.edu) is an as-

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

EXPANDED BRAND VALUE CHAIN

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

SUMMARY OF PAST RESEARCH ON CTB

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

materialistic secularism (Inglehart and Welzel 2005, 26, TABLE 2

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

measured using items developed by Donnellan et al. Model and Estimation

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

rized in table 3. þ dq02 MKTMIXqijk SELFEXPRk

to construct our final data set. Our final sample consisted X

VARIABLES AND DESCRIPTIONS

Variable Operationalization Reference Source

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

MEAN CTB ACROSS THE COUNTRIES

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Covariate Parameter Expected sign Estimate p-value

Intercept c000 5.347 < .001

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

MEAN CTB AT VARYING LEVELS OF MARKETING-MIX ACTIVITIES

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

TABLE 5 In our theorizing, we discussed rival predictions for

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

Downloaded from https://academic.oup.com/jcr/article/46/4/651/5519122 by guest on 14 January 2021

You might also like