Purchasing and Supply Chain Management
National Culture, Economy, and Customer Lifetime Value: Assessing the Relative Impact of the Drivers of Customer Lifetime Value for a Global Retailer V. Kumar and Anita Pansari
ABSTRACT Customer lifetime value (CLV), a metric used in many industries, is based on the cumulated cash flow a customer accrues during his or her lifetime. Firms have used CLV as a basis for formulating and implementing customer-specific strategies; however, these can vary across countries because of each country’s cultural and economic influences. Typically, CLV is computed with three components: purchase frequency, contribution margin, and marketing costs. In this study, the authors demonstrate that national cultural dimensions affect the drivers of purchase frequency and contribution margin and that economic factors influence the components of CLV directly. They use customer-level transaction data from a global retailer for a random sample of customers in 30 representative countries over a six-year period. They estimate the model using a seemingly unrelated regression approach while accounting for the heterogeneity of customers across countries and the endogeneity of marketing costs incurred by the firm. The results indicate that global retailers should pay attention to the dimensions of national culture and economic conditions because of their differential impact on CLV.
Keywords: customer lifetime value, purchase frequency, national culture, economic growth, contribution margin
T he objective of multinational corporations (MNCs) is to establish sustainable profitable ventures around the world. Focusing on the best possible service for
customers, and understanding those customers’ needs,
can help firms stay ahead of the competition. It is im- possible to understand each customer through a direct (personal) interaction; therefore, firms are resorting to understanding the behavior of customers through trans- action data. This customer-level information can help firms define and service customer segments consisting of a single customer (Verhoef and Donkers 2001).
For example, firms can use this information to predict and understand the future profitability of customers through customer lifetime value (CLV) (Kumar 2008). The CLV metric refers to “the net present value of the future profit
V. Kumar (VK) is Regents’ Professor, Richard and Susan Lenny Dis- tinguished Chair Professor, and Executive Director, Center for Excellence in Brand and Customer Management, J. Mack Robinson College of Business, Georgia State University; Chang Jiang Scholar, HUST; Lee Kong Chian Fellow, Singapore Management University; and ISB Senior Fellow, Indian School of Business (e-mail: vk@gsu.edu). Anita Pansari is a doctoral student in marketing, Center for Excellence in Brand and Customer Management, J. Mack Robinson College of Business, Georgia State University (e-mail: apansari1@gsu.edu). The authors thank the review team for valuable suggestions during the revision of the manu- script. The authors thank the multinational firm for providing the data for this study. They also thank Alok Saboo and the participants of the 2015 Global fashion marketing conference for their valuable suggestions. Finally, they thank Renu for copyediting a previous version of the man- uscript. Bulent Menguc served as guest editor for this article.
Journal of International Marketing ©2016, American Marketing Association Vol. 24, No. 1, 2016, pp. 1–21 DOI: 10.1509/jim.15.0112 ISSN 1069-0031X (print) 1547-7215 (electronic)
National Culture, Economy, and Customer Lifetime Value 1
from a customer” (Kumar 2008, p. 4) and is a forward- looking metric that can serve as a basis for formulating and implementing customer-specific strategies (Kumar and Rajan 2012). Firms in various industries have ex- amined CLV; these include a catalog retailer (Reinartz and Kumar 2003), an entertainment product company (Drèze and Bonfrer 2003), a newspaper subscription service (Thomas, Blattberg, and Fox 2004), a music website (Fader, Hardie, and Lee 2005), a provider of on-demand Internet streaming media (Pfeiffer 2011), and a tech- nology and consulting corporation (Kumar et al. 2008). A firm can compute CLV by predicting its components, such as how often a customer buys (purchase frequency), how much he or she buys (gross contribution margin), and the cost incurred for the customer (marketing cost). Various drivers affect these three components of CLV. Specifically, these drivers define the nature of the relationship between the firm and its customers, help determine the level of profitability, and help increase CLV.
With the increasingly unified marketplace, it is important for MNCs to understand the attitudes and behaviors of customers worldwide. Firms can personalize their strate- gies for each segment of customers by evaluating the drivers that affect the two main components of CLV (purchase frequency and contribution margin) for a given segment. Firms typically set marketing costs according to what they expect of customers and their past behavior. At a global level, segments can be based on the different countries in which the firm operates. Countries differ in their cultural and economic factors, which affect the tastes and preferences of customers; that is, these factors can affect consumers’ choices, intentions, and behavior (Henry 1976). Therefore, we pose three key research questions in this study:
1. Does the relative importance of the drivers of purchase frequency and contribution margin vary across countries?
2. If yes, do national culture and economic factors play a role in explaining the differences in the relative importance of these drivers?
3. How can MNCs manage their customers across countries to increase future profitability?
Answering these questions will contribute not only to managers’ tool kits but also to the cross-cultural and strategy literature streams in marketing. According to Steenkamp (2001, p. 30), “The further advancement of marketing academic discipline requires that the validity of
our theories and models be examined in other cultural settings as well to identify their degree of generaliz- ability and to uncover boundary conditions.” Furthermore, Iyengar and Lepper (1999, p. 364) note that studying the role of national culture in marketing teaches us “the many ways in which our theories and paradigms are a reflection of the culture in which they were developed.”
Therefore, in this study we focus on understanding how national cultural differences moderate the importance of the drivers of purchase frequency and contribution margin and how a country’s economy affects all the CLV components for an MNC with operations in many countries. We obtain customer-level transaction data for a random sample of 1,000 customers in 30 repre- sentative countries, use Hofstede’s (1980) cultural di- mension scores for each country, and also use gross domestic product (GDP) per capita to measure the economic growth of each country. To our knowledge, this study is the first to comprehensively examine the empirical role of culture in moderating the effects of the traditional drivers of the CLV components.
The rest of this article is divided into five parts. We first discuss the motivation for the study and, second, offer a review of the popular press and the academic literature on the importance of national cultural and economic dif- ferences for firms. The third section focuses on the liter- ature review of CLV, national culture, and economic factors. Fourth, we discuss our conceptual framework, which comprises the national cultural and economic di- mensions and their impact on the drivers of purchase frequency and contribution margin. The fifth section provides details on the data, statistical analysis, and the results of the study. We conclude with a discussion of the study’s implications and limitations and the scope for further research.
MOTIVATION
Disney has six theme parks around the world. Although the parks all use the nickname “The Happiest Place on Earth,” they all differ in some key aspects depending on the culture of the country where each is located. For example, the Tokyo Disneyland serves steamed dessert buns, a specialty of Japan, and Hong Kong Disneyland’s design is based on traditional Chinese elements and feng shui.1 Robert Iger, Disney’s president and chief operating officer in 2004, stated: “We know if we’re too U.S.-centric, the products won’t be too relevant to those markets.” Clothing retailer C&A in Europe standardized buying in 1997. However, in June 2000, the company decided to close
2 Journal of International Marketing
all 109 stores in the United Kingdom and Ireland after substantial losses because the taste of British and Irish consumers differed from the taste of continental Europeans.2
These examples highlight the importance of understand- ing a country’s national culture before establishing a business there. Expanding its operations to countries with cultural values different from its own, without adapting to these differences, can negatively affect a firm’s revenue (Ricks 1993). A country’s culture is a key environmental force that shapes its people’s perceptions, dispositions, and behaviors (Triandis 1989). Culture leads to differential processing and evaluations of environmental information (Hofstede 1991) and influences the types of socially en- gaging and disengaging emotional processes people ex- perience. Although the outcome of the differences could be in the form of consumers’ attitudes and behaviors or the style of company operations, the source of the differences could be in the form of tastes and preferences, marketing practices, retail setup, and so on.
LITERATURE REVIEW CLV
CLV is based on the cumulated cash flow and value accruing directly or indirectly from a customer over his or her lifetime with the company. It can be used to identify the most and least profitable customers, classify them into various (high, medium, and low) segments, and direct marketing efforts at the most valuable of them. The CLV metric is a function of the purchase frequency, the pre- dicted contribution margin, and the marketing resources allocated to the customer.
Prior research (Venkatesan and Kumar 2004) has com- puted the established CLV metric as follows:
CLVit = � Ti
t=1
GCMit ð1 + rÞt=frequencyi
- � n
l =1
�mciml*Ximl ð1 + rÞl
,
(1)
where
CLVit = lifetime value of customer i in period t;
GCMit = predicted gross contribution margin from customer i in period t, measured in dollars;
r = discount rate for money;
ciml = unit marketing cost for customer i in channel m in year l;
ximl = number of contacts to customer i in channel m in year l;
frequencyi = predicted purchase frequency for cus- tomer i in each year;
n = number of years to forecast; and
Ti = predicted number of purchases made by customer i until the end of the planning period.
Components of CLV
Purchase Frequency. Purchase frequency is a count var- iable that measures the number of purchases a customer has made. A retailer’s objective is to ensure repeated future purchase activity, which generates future revenue. Mor- gan and Hunt (1994) argue that satisfactory interac- tions lead to greater trust, which in turn leads to a longer relationship and, thus, higher gross profits. However, Reinartz and Kumar (2003) show that a customer with moderate purchase frequency in a noncontractual setting stays longer with the firm. Nevertheless, in the immediate future, the goal is to increase the frequency of purchases.
Gross Contribution Margin. Gross contribution margin per purchase is calculated by deducting the cost of goods sold from the revenue received from a customer in each purchase. This component is affected by how many times the firm contacted the customer, the different discounts and promotions offered, and the other marketing activ- ities targeted to the customer.
Marketing Cost. The marketing cost is the cost of activities firms incur from increasing the value of the existing cus- tomer relationships, such as development and retention costs. A major component of these costs is typically the cost of marketing through various communication channels, such as direct mail, e-mail, telephone calls, and face-to-face interactions. These costs also vary depending on the country.
Prior research has established the impact of each of these components on profitable customer lifetime duration (Reinartz and Kumar 2000), which is similar in principle to CLV. Over the years, extensive research has examined various drivers that affect the revenue-generating com- ponents (purchase frequency and gross contribution margin), classifying them into exchange characteristics
National Culture, Economy, and Customer Lifetime Value 3
and customer heterogeneity. Exchange characteristics encompass the set of variables that define relationship activities in the broadest sense. Common exchange char- acteristics are the level of purchases, customer spending, cross-buying behavior, average interpurchase time, num- ber of product returns, usage of loyalty programs, and the firm’s mailing efforts (Reinartz and Kumar 2003). The exchange characteristics are mostly the same in business-to-business (B2B) and business-to-consumer (B2C) settings. Although these exchange characteristics can vary from industry to industry, the common exchange characteristics we observe in all the relevant studies are frequency of buying, multichannel buying behavior, cross-buying, enrollment in loyalty programs, number of product returns, and the firm’s mailing efforts. Table 1 provides the definitions and operationalization of these exchange characteristics. Customer heterogeneity refers to the demographic and psychographic indicators (in a B2C setting) and firmographics (in a B2B setting) that help a firm segment customers and manage customer–firm re- lationships. To identify the role of cultural and economic factors in influencing the drivers of purchase frequency and gross contribution margin, we first need to understand the impact of these two factors on buyer behavior.
Role of National Culture and Economy
National culture is one of the many constructs relevant to international marketing research (Cavusgil 1998) that can explain differences in marketing management de- cision making (Tse et al. 1988) in both B2B and B2C firms. Research has explained these differences in terms of global brand-image strategies (Roth 1995) and the effectiveness of emotional appeals in advertising (Aaker and Williams 1998). National culture can also influence consumer innovativeness (Steenkamp, Ter Hofstede, and Wedel 1999), new product development activity (Nakata and Sivakumar 1996), word-of-mouth behavior in in- dustrial services (Money, Gilly, and Graham 1998), and buyer behavior in tourism (Pizam and Reichel 1996).
A country’s culture is a key environmental characteristic underlying systematic differences in buying behavior. Cultural norms and beliefs are powerful forces that shape people’s perceptions, dispositions, and behaviors (Markus and Kitayama 1991). Culture is reflected in “general tendencies of persistent preference for particular state of affairs over others, persistent preferences for specific social processes over others, and general rules for selec- tive attention, interpretation of environmental cues, and responses” (Tse et al. 1988, p. 82). These tendencies of persistent preferences vary across countries and societies.
Hofstede (1980, 1991) and Schwartz (1994) developed the two most rigorous, comprehensive frameworks that define the differences among countries in the past two decades.
Hofstede (1980, 1991) developed by far the most in- fluential national cultural framework, using extensive empirical analyses. He distinguishes six dimensions of na- tional culture measured on a 100-point scale: individualism/ collectivism, masculinity/femininity, uncertainty avoidance, long-term/short-term orientation, power distance, and indulgence/restraint (the most recently developed measure). Schwartz (1994) proposes an alternative cultural frame- work based on human values. He identifies three basic societal issues that define seven national-cultural domains: conservatism/autonomy, hierarchy/egalitarianism, and mastery/harmony.
Hofstede’s classification pertains to national differences in motives for buying products and services, the degree of dependence on brands, adoption of new technology, and media use. Many consumption differences can also be predicted and explained by analyzing the relationship between consumption and scores on Hofstede’s di- mensions of national culture (Petersen, Kushwaha, and Kumar 2015). Therefore, we use Hofstede’s dimensions for our study. Hofstede’s power-distance dimension discusses power distribution (hierarchy) in a country, which affects behavior toward people (Hofstede and Bond 1988) and thus has a minimal impact on consumers’ purchase decisions. With our focus on understanding the effect of the cultural dimensions on the relationship be- tween CLV and its drivers, we do not include the effect of power distance on consumer behavior.
Culture dimensions highlight the differences across cul- tures that affect consumers’ desire and willingness to buy products. However, consumers’ ability to buy is also an important dimension to consider when evaluating customer profitability. Thus, we focus on the economic conditions of a country, which in turn reflect consumers’ ability to buy.
Economic Factors and Buying Behavior
Economic factors such as GDP per capita help determine the consumption pattern of a country. If a country has a high GDP and high purchasing power, its consumers will have more disposable income and therefore will be able to spend more. In emerging markets in which a large pro- portion of the population is middle class, the amount of disposable income is lower than that in developed countries. Companies cannot determine their strategies by taking into account only the current economic situation
4 Journal of International Marketing
Ta bl e 1.
D ef in iti on
an d Co nc ep tu al iz at io n of
Va ria bl es
U se d in th e M od el s
D ef in it io n
O p er at io n al iz at io n
M ea n (S D ) A cr o ss
C o u n tr ie s
E ff ec t o n
P u rc h as e
F re q u en cy
E ff ec t o n
C o n tr ib u ti o n
M ar gi n
V ar ia b le s U se d in
th e M o d el
C ro ss -b u yi n g (C
B )
T h e n u m b er
o f d ep ar tm
en ts
(r el at ed
o r u n re la te d p ro d u ct
ca te go
ri es ) fr o m
w h ic h
cu st o m er s p u rc h as e
p ro d u ct s o r se rv ic es
T o ta l n u m b er
o f d ep ar tm
en ts
in w h ic h th e cu st o m er
m ad
e p u rc h as es
in a ye ar
5 .2
(2 .1 )
+ +
P ro d u ct
re tu rn s (R
E T )
T h e n u m b er
o f p ro d u ct s th e
cu st o m er
re tu rn s b et w ee n
th e tw
o p u rc h as e p er io d s
T o ta l va
lu e o f re tu rn s cu st o m er
m ad
e in
a ye ar
(2 0 0 8 U SD
) 9 3 .8
(5 6 .4 )
+ ᴖ
M u lt ic h an
n el
b u yi n g
b eh av
io r (M
C B )
Sh op
pi ng
m ul tip
le ch an ne ls of
th e fir m
(e .g ., st or e, on
lin e, ca ta lo g)
N u m b er
o f ch an
n el s fr o m
w h ic h
th e cu st o m er
p u rc h as ed
in a ye ar
2 .7
(. 6 4 )
+ +
L o ya
lt y p ro gr am
(L P )
T h e st at u s o f th e cu st o m er
w it h
th e fi rm
in te rm
s o f p o ss es si n g
a lo ya
lt y ca rd
an d th e le ve l o f
lo ya
lt y ca rd
E q u al
to 1 if th e cu st o m er
h as
en ro ll ed
in th e lo ya
lt y p ro gr am
, 0 if th e cu st o m er
h as
n o t en ro ll ed
in th e lo ya
lt y p ro gr am
.5 2 (. 2 2 )
+ +
P u rc h as e fr eq u en cy
(P F R E Q )
T h e n u m b er
o f ti m es
th e
cu st o m er
b u ys
fr o m
th e b ra n d
N u m b er
o f ti m es
th e cu st o m er
m ad
e a p u rc h as e
4 .8
(1 .7 )
+ ᴖ
C o n tr ib u ti o n m ar gi n (C
M )
T h e p ro fi t ea rn ed
p er
p u rc h as e
(R ev en u e m in u s co st
o f go
o d s
so ld ) p er
p u rc h as e (2 0 0 8 U SD
) 5 7 .6
(2 2 .4 )
+ +
D ir ec t m ar k et in g (D
M )
T h e to ta l co st
o f m ar k et in g
d ir ec tl y to
th e cu st o m er ,
in cl u d in g e- m ai ls , co u p o n s,
an d d is co u n ts
p ro vi d ed
to ea ch
cu st o m er
T o ta l co st
o f d ir ec t m ar k et in g to
a cu st o m er
in a ye ar
(2 0 0 8 U SD
) 1 8 .7
(8 .4 )
+ ᴖ
A d ve rt is in g co st
(A D V )
T h e ag
gr eg at e co st
o f
b ro ad
ca st
m ed ia
T o ta l b ro ad
ca st
m ed ia
co st
in a ye ar
(2 0 0 8 U SD
) 1 0 .6
(7 .5 )
+ +
G D P p er
ca p it a (G
D P )
T h e m ea su re
o f ec o n o m ic
gr o w th
o f a co u n tr y
G D P p er
ca p it a (2 0 0 8 U SD
) 3 2 ,4 1 9 .6 3 (2 2 ,5 7 2 .8 6 )
C o n tr o l V ar ia b le s
H o u se h o ld
in co m e (H
H I)
T h e to ta l in co m e o f th e
h o u se h o ld
C o m b in ed
in co m e o f h o u se h o ld
(2 0 0 8 U SD
) 5 8 ,6 0 0 (2 1 ,4 7 1 )
N .A .
N .A .
A ge
T h e ag
e o f th e h ea d o f th e
h o u se h o ld
A ge
o f th e h ea d o f th e
h o u se h o ld
(y ea rs )
3 4 (1 1 )
N .A .
N .A .
N o te s: Sy m b o ls : + = p o si ti ve
re la ti o n sh ip ; \ = in ve rs e U -s h ap
ed re la ti o n sh ip . N .A . = n o t ap
p li ca b le .
National Culture, Economy, and Customer Lifetime Value 5
of a country; they must also consider the changes occurring in the world economy, because differences in the relative ability to buy products will have an impact on the ultimate customer buying behavior. It is important to understand how national culture and the economy affect the drivers of purchase frequency and contribution margin. Un- derstanding this will help managers strategize their marketing communications and maximize firm profits globally.
PROPOSED CONCEPTUAL FRAMEWORK
Every industry features its own bandwidth of trans- actional and relational exchanges (Anderson and Narus 1991) that help form customer attitudes and behavior toward the brand. These behaviors can be in the form of purchases, loyalty program enrollment, cross-buying,
and multichannel buying. Such behaviors vary depend- ing on the level of trust and commitment the customer has established with the firm. The factors that determine future exchanges for short- versus long-term-oriented customers differ (Garbarino and Johnson 1999). Cus- tomers who are more committed are also likely to seek greater relationship expansion and enhancement (Bendapudi and Berry 1997); thus, these customers are likely to invest more in cross-buying and multichannel buying. All these customer behaviors contribute to the revenue-generating components of CLV (contribution margin and purchase frequency). We illustrate our con- ceptual model in Figure 1.
The purchase behavior of customers depends not only on their relationship with the firm but also on the firm’s actions toward the customers. Customized marketing messages relevant to customers’ wants and needs will
Figure 1. CLV in a Global Context: A Conceptual Framework
Past Relationship with the Firm
Trust
Multichannel Buying
Commitment Loyalty
Past Commitment to Customer
Marketing Effort/Cost
Direct Marketing
Advertising
Cultural Factors Individualism Uncertainty Avoidance Masculinity
Indulgence
Customer Behavior
Commitment to Customer Marketing Effort/Cost
Past Customer Behavior Purchase Frequency Contribution Margin
Purchase Frequency
Contribution Margin
Economic Factors
Long-Term Orientation
Cross-Buying Product Returns
Customer Lifetime Value
6 Journal of International Marketing
motivate them to visit a store more frequently. Therefore, firms’ marketing actions are important to enhance CLV components. Marketing that focuses on building trust and commitment is more effective for long-term relationships (Garbarino and Johnson 1999). Ninety percent of the cumulative effect of ad- vertising for mature, frequently purchased, low-priced products occurs within three to nine months of the advertisement (Clarke 1976). Thus, because the results of firms’ marketing efforts come to fruition only over time, customers’ present purchase frequency and con- tribution margin reflect the impact of past marketing efforts.
Research has extensively discussed the relationship among past customer behavior, past marketing efforts, and current purchase frequency and contribution margin in the context of the U.S. market (e.g., Kumar, Shah, and Venkatesan 2006). However, it is important for MNCs to understand the changes in the magnitude of these relationships, which are caused by the national cultural and economic factors of each country. The cultural factors affect the variance in the magnitude of the relationship between drivers of purchase frequency and contribution margin (and, therefore, CLV) across countries, and the economic factors have a direct impact on purchase frequency and contribution mar- gin. However, not all cultural factors affect all the drivers of the components of CLV. In the following subsections, we discuss the cultural dimensions and the impact of the relevant dimensions on the drivers of two CLV components (purchase frequency and contribu- tion margin).
Relationship of Individualism with Purchase Frequency and Contribution Margin
The cultural dimension of individualism/collectivism has the strongest variation across cultures (Hofstede 1980). In individualist cultures, the needs, values, and goals of in- dividuals take precedence over those of the group, whereas in collectivist cultures, the needs, values, and goals of the group take precedence over those of the individuals (Gudykunst 1997). Trust levels are higher in individualist (vs. collectivist) countries because individualism promotes a trusting stance; for example, better outcomes might arise from assuming that others are reliable. Thus, in- dividualists are more likely to trust others until they are given reasons not to do so. In turn, this trust affects their buying behavior. Next, we discuss how individualism affects the drivers of purchase frequency and contri- bution margin.
Multichannel Buying. Consumers in individualist coun- tries not only buy from stores but also trust companies to deliver the right products at the right time when they order online or through a catalog. By contrast, collec- tivist consumers are more likely to base their trust on relationships in which they have firsthand knowledge, thus prompting them to buy products in the actual stores. Interactions with store employees and being able to see, feel, and test products help them establish trust. Therefore, consumers from individualist rather than collectivist societies can reap more benefits because they shop in multiple channels. These customers are also more profitable to the firm because they tend to spend more, and the cost of handling these customers is dis- tributed over channels. Thus:
H1a: The higher the individualism in a country, the greater is the positive impact of multichannel buying on purchase frequency.
H1b: The higher the individualism in a country, the greater is the positive impact of multichannel buying on the contribution margin.
Cross-Buying. In individualist societies, consumers think in terms of “I” rather than “we.” They focus on indi- vidual goals and tend not to follow in-group behavior. As such, these consumers may buy from different de- partments, but these departments are restricted to their own needs and goals. However, in a collectivist society, individuals are encouraged to suppress their hedonic desires in favor of group interests and goals (Kacen and Lee 2002). They tend to be concerned with affiliating closely with others, maintaining connectedness, and blending the self–other boundary (Aaker and Williams 1998). As such, these individuals not only buy for themselves but also shop for others in their groups. Al- though both types of consumers cross-buy, such behavior is more frequent in collectivist cultures. Because collec- tivist consumers buy for more people, they tend to visit stores more frequently and spend more money. Such cross-buying is beneficial to firms in terms of cost, because they do not need to market separately to individuals. Thus:
H1c: The lower the individualism in a country, the greater is the positive impact of cross-buying on purchase frequency.
H1d: The lower the individualism in a country, the greater is the positive impact of cross-buying on contribution margin.
National Culture, Economy, and Customer Lifetime Value 7
Loyalty Programs. In collectivist countries, people tend not to trust out-group members, and thus it is easier for firms to enroll customers in loyalty programs and offer company credit cards because customers want to establish trust with retailers. Furthermore, people in high collec- tivist countries have strong beliefs that loyalty is rewarded through improved well-being (Matthyssens and Wursten 2003; Patterson and Smith 2003). After customers be- come loyal to the firm, they will likely buy more fre- quently and spend more money in the store. Thus:
H1e: The lower the individualism in a country, the greater is the positive impact of owning a loyalty card on purchase frequency.
H1f: The lower the individualism in a country, the greater is the positive impact of owning a loyalty card on contribution margin.
Product Returns. In individualist countries, the level of trust is high and perceived risk is low. As such, many firms have lenient return policies because of the high level of trust established between them and their cus- tomers. However, collectivist customers trust only their in-group members, and therefore many firms have more stringent return policies because the trust between them and their customers is low. Lenient return policies en- courage customers to return products and therefore provide the firm an opportunity for interaction. In turn, such polices encourage consumers to buy more fre- quently, because they are able to return the product if necessary. Thus:
H1g: The higher the individualism in a country, the greater is the positive impact of returns on purchase frequency.
H1h: The higher the individualism in a country, the greater is impact of the inverse U-shaped re- lationship of returns on contribution margin.
Relationship of Uncertainty Avoidance with Purchase Frequency and Contribution Margin
Uncertainty avoidance captures the risk dimension of culture. It measures the degree to which societies feel threatened by uncertain, risky, ambiguous, or undefined situations and the extent to which they try to avoid such situations by adopting stricter codes of behavior. Countries low in uncertainty avoidance have a greater tolerance for risk and a higher “willingness to take unknown risks” (Hofstede and Hofstede 2001, p. 161). Next, we discuss
how uncertainty avoidance affects the drivers of purchase frequency and contribution margin.
Multichannel Buying. People in countries with low un- certainty avoidance accept uncertainty and want variety in their lives (Dwyer, Mesak, and Hsu 2005). As Hofstede (1991, p. 118) notes, people in low-uncertainty-avoidance cultures are “more prepared to give the benefit of the doubt to unknown situations, people, and ideas.” As such, consumers in countries low in uncertainty avoidance are willing to explore firms’ multiple channels. Thus:
H2a: The lower the uncertainty avoidance in a coun- try, the greater is the positive impact of multi- channel buying on purchase frequency.
H2b: The lower the uncertainty avoidance in a country, the greater is the positive impact of multichannel buying on contribution margin.
Cross-Buying. Buying in different departments of the same stores is an indication of greater trust in the company and lower perceived risk. Therefore, consumers with lower tolerance of risk engage in more cross-buying. Such low levels of risk are evident in countries high in uncertainty avoidance, in which consumers are hesitant to switch brands, especially if they have established trust with the retailer. In such cultures, consumers may buy all their products from the same firm and thus provide more rev- enue. In low-uncertainty-avoidance cultures, consumers have a greater tolerance for risk and thus are willing to try new products from different brands. In addition, these consumers prefer more variety in their lives (Dwyer, Mesak, and Hsu 2005) and therefore are willing to explore alternative brands. Thus:
H2c: The higher the uncertainty avoidance in a country, the greater is the positive impact of cross-buying on purchase frequency.
H2d: The higher the uncertainty avoidance in a country, the greater is positive impact of cross-buying on contribution margin.
Loyalty Programs. Loyalty instruments are useful to cus- tomers who regularly buy from the same store; therefore, they are more effective in high-uncertainty-avoidance countries because customers avoid risk and are hesitant to try new things. In contrast with low-uncertainty- avoidance countries, these consumers largely believe that “what is different is harmful.” Therefore, they rely on stores with which they have established trust over time and
8 Journal of International Marketing
with which they are familiar. Conversely, consumers in low-uncertainty-avoidance countries tend to follow mar- ket trends and try new things. Therefore, the loyalty in- strument in these countries may not be as affective, unless the store has a frequent in-flow of new styles and different offerings. Thus:
H2e: The higher the uncertainty avoidance in a country, the greater is the positive impact of owning a loyalty card on purchase frequency.
H2f: The higher the uncertainty avoidance in a country, the greater is the positive impact of owning a loyalty card on contribution margin.
Product Returns. In high-uncertainty-avoidance cultures, consumers are resistant to change from established pat- terns, and focus on risk avoidance and reduction. How- ever, those in low-uncertainty-avoidance cultures have a greater tolerance for risk and are willing to take unknown risks (Hofstede and Hofstede 2001); that is, they accept uncertainty and want variety in their lives (Dwyer, Mesak, and Hsu 2005). Consequently, stores in countries low in uncertainty avoidance are willing to take higher risks and therefore have lenient return policies. Furthermore, in addition to trying out new products, consumers in these countries use multiple channels to shop because of the ease of returns. Thus:
H2g: The lower the uncertainty avoidance in a coun- try, the greater is the impact of returns on purchase frequency.
H2h: The lower the uncertainty avoidance in a country, the greater is the impact of the inverse U-shaped relationship of returns on contribu- tion margin.
Relationship of Long-Term Orientation with Purchase Frequency and Contribution Margin
Long-term orientation is the extent to which a society exhibits a pragmatic future-oriented perspective rather than a conventional historic or short-term point of view (Hofstede 1991). Differences in long-term orientation may lead to differences in consumers’ frugality, frequency of credit abuse, and propensity to plan purchases (Lastovicka et al. 1999). In countries with long-term orientation, consumers tend to establish long-term relationships and are cautious about new products. Therefore, they are more comfortable buying the brand with which they have
established trust. This dimension is a time-oriented mea- sure and therefore does not affect the return behavior of customers; thus, we discuss how long-term orientation affects only the relevant drivers of multichannel buying, cross-buying, and loyalty programs.
Multichannel Buying. In long-term-oriented cultures, people prefer using cash or debit cards to credit cards (De Mooij and Hofstede 2002). Consumers in long-term- oriented cultures use resources sparingly and prefer going to a store to pick up merchandise rather than having merchandise delivered to their homes. As such, they likely purchase less online or through web catalogs. Thus:
H3a: The higher the long-term orientation in a country, the less is the positive effect of multi- channel buying on purchase frequency.
H3b: The higher the long-term orientation in a country, the less is the positive effect of multi- channel buying on contribution margin.
Cross-Buying. Consumers in a long-term-oriented cul- ture are frugal with their resources and emphasize savings. Thus, they shop at stores whose prices they trust and rarely try new stores because of their uncertainty about prices and variety. Conversely, consumers from short-term-oriented countries experience materialist consumption pressures, “even if it means overspending” (Hofstede 1991, p. 173). They tend to be sensitive to social trends and are willing to try new products at different stores with the latest fashions. They may or may not buy in different departments of the same store, because they are not driven by deals, long-term re- lationships, or savings but, rather, by social trends. Thus:
H3c: The higher the long-term orientation in a coun- try, the greater is the positive impact of cross- buying on purchase frequency.
H3d: The higher the long-term orientation in a country, the greater is the positive impact of cross-buying on contribution margin.
Loyalty Programs. Loyalty cards, by which customers accumulate points to gain rewards and discounts, are successful in countries with a long-term orientation. Because consumers in these countries plan for the future, they also often use the rewards from these loyalty cards. Furthermore, they are frugal insofar as they acquire and
National Culture, Economy, and Customer Lifetime Value 9
resourcefully use goods and services to achieve their long- term goals (Lastovicka et al. 1999). Thus:
H3e: The higher the long-term orientation in a coun- try, the greater is the positive impact of owning a loyalty card on purchase frequency.
H3f: The higher the long-term orientation in a coun- try, the greater is the positive impact of owning a loyalty card on contribution margin.
Relationship of Masculinity with Purchase Frequency and Contribution Margin
Masculinity is the degree to which a society is charac- terized by assertiveness versus nurturance. More mas- culine societies place greater emphasis on wealth, success, ambition, material things, and achievement, whereas more feminine societies place greater value on people, helping others, preserving the environment, and equality (Hofstede 1980). Purchasing new items is one way for a person to assert his or her interests and to show (off) wealth and success (Rogers 1983). In a masculine society, little can be inferred from multichannel buying, loyalty programs, or product returns because these dimensions do not focus on “showing off” or assertive- ness; thus, we discuss only how masculinity affects cross- buying.
Wearing the latest fashions and buying expensive prod- ucts are quite evident in masculine cultures. These cul- tures focus on conspicuous consumption and encourage buying expensive products to show success. Members of high-masculinity cultures tend to buy the latest fashion products from different specialty stores, frequently to maintain their status. As such, consumers in masculine cultures tend to purchase from multiple stores and not from multiple departments of the same stores. Thus:
H4a: The lower the masculinity in a country, the greater is the positive impact of cross-buying on purchase frequency.
H4b: The lower the masculinity in a country, the greater is the positive impact of cross-buying on contribution margin.
Relationship of Indulgence with Purchase Frequency and Contribution Margin
The latest dimension added to Hofstede’s cultural di- mension is indulgence versus restraint. Indulgent societies
allow for gratification of basic and natural human drives, such as enjoying life and having fun (Hofstede, Hofstede, and Minkov 2010), whereas more restrained societies tend to adhere to relatively strict social norms that curb such gratifications. Thus, indulgence reflects the value underlying consumers’ attitudes toward the role of products and products’ relationship to individuals and society (Griffith and Rubera 2014). Because the indulgence measure captures customer spending for gratification, it does not affect their return behavior; thus, we discuss how in- dulgence affects only the relevant drivers of multichannel buying, cross-buying, and loyalty programs.
Multichannel Buying. In a high-indulgence society, con- sumers focus on instant gratification and therefore want to purchase items such as apparel in stores and outlet malls. However, they may not be active shoppers in online retail because gratification is realized only on receipt of the apparel. This suggests that consumers in a high-indulgence country purchase less online. Thus:
H5a: The higher the indulgence in a country, the less is the positive effect of multichannel buying on purchase frequency.
H5b: The higher the indulgence in a country, the less is the positive effect of multichannel buying on contribution margin.
Cross-Buying. Consumers in indulgent cultures do not restrict their purchases and buy different categories of products. As such, they likely shop from any store that has a wide variety of goods from which to choose. Be- cause indulgence signifies instant gratification, con- sumers in indulgent cultures are willing to try new and creative things. This behavior can lead them to buy a wider variety of products in a wider variety of stores. Thus:
H5c: The higher the indulgence in a country, the greater is the positive impact of cross-buying on purchase frequency.
H5d: The higher the indulgence in a country, the greater is the positive impact of cross-buying on contribution margin.
Loyalty Programs. Possessing loyalty cards may cause consumers to shop only in stores at which they hold these cards, because of the various benefits. Therefore, con- sumers in indulgent countries may not tie themselves to loyalty programs. By contrast, loyalty cards, which help
10 Journal of International Marketing
customers accumulate points, are more successful in low- indulgence countries. Consumers in these countries are restrained from possessing goods/services. Thus:
H5e: The lower the indulgence in a country, the greater is the positive impact of owning a loyalty card on purchase frequency.
H5f: The lower the indulgence in a country, the greater is the positive impact of owning a loyalty card on contribution margin.
Economic Factors
In our study, we expect that customers in emerging markets, which are characterized by higher population density and lower disposable income, will have fewer options in choosing their most preferred store and less ability to buy from high-end, luxurious retailers because of the smaller number of stores in these markets. Fur- thermore, only a small percentage of the population buys luxurious brands, and this group mostly belongs to the upper- and middle-class economic statuses in the country. However, the middle-class group in emerging markets has been growing in the past few decades; these customers now have more disposable income to pur- chase luxury goods. Although there is growth in these segments of society, the amount of money they spend on luxury goods is less than that in the developed markets, which have higher GDPs per capita. Furthermore, the cost of setting up stores in emerging markets is higher because of higher real estate costs, fewer customers, and legal regulations. Thus:
H6a: The higher the country’s GDP per capita, the higher is the purchase frequency.
H6b: The higher the country’s GDP per capita, the higher is the contribution margin.
RESEARCH METHODOLOGY Data Description
We used customer data from a global fashion retailer operating in multiple countries that sells apparel, shoes, jewelry, cosmetics, and accessories for both men and women. We obtained data on a sample of 1,000 cus- tomers for the 2008–2013 period from 30 representative countries in which the retailer operates. This resulted in a sample size of 30,000 customers for the six-year period. Thus, we had six observations for each customer,
resulting in 180,000 total observations. The countries span multiple stages of cultural and economic devel- opment. In addition, our data form a balanced panel; we had annual data for all customers’ demographics and purchase frequency, the products they bought, any returns, their membership in loyalty card programs, the number of different departments they shopped in, and the number of channels they used to shop. The company also provided its marketing data, which include data on advertising and direct marketing costs. The direct marketing costs include the costs of e-mails, telephone calls, and mailed catalogs. Because the study spans countries with different currencies, we pegged the value of all our monetary variables to the 2008 U.S. dollar value for ease of comparison. We use the data to generate model variables to understand customer behavior and also to estimate the CLV components of the customers in each country. These variables include multichannel buying, cross-buying, usage of loyalty card programs, purchase frequency, and returns. Furthermore, to con- trol for brand-building efforts in countries, we also use advertising dollars spent as a variable to model customer behavior. Because only direct marketing costs vary by customers, we model these to predict CLV (see Table 1).
Model Specification
As mentioned previously, CLV consists of three main components: purchase frequency, contribution margin, and direct marketing cost. In line with prior research (Kumar, Shah, and Venkatesan 2006), we also provide the nature of the relationship observed between the model variables and the two revenue generating com- ponents of CLV (i.e., purchase frequency and contri- bution margin). For accurate measurement of CLV, we estimate the three components for each customer using models documented in the literature (Reinartz and Kumar 2003) and then combine the predictions from the three models to arrive at a single value representing the CLV in dollar terms.
Our panel data enable us to predict how often each cus- tomer will buy, how much he or she will buy, and how much the firm has to spend on marketing to each customer in each of the six years. However, many customers will not have bought anything in a particular year and thus will have a zero purchase frequency. To account for the con- centration of zeros in our model, we model those who will purchase in each year by using a choice model (probit). For example, our dependent variable is equal to 1 if the cus- tomer purchased in 2008 and 0 otherwise. We use SAS 9.4 to implement this model. We first use PROC logistic with a
National Culture, Economy, and Customer Lifetime Value 11
probit link function to estimate the model of the number of active customers (those who purchased in 2008) and then use the inverse Mills ratio from the probit model as an independent variable to model purchase frequency and contribution margin in 2008. Following the same pro- cedure for every year, we then use the inverse Mills ratio from the probit model of each year as an independent variable in the equation for purchase frequency and contribution margin.
Our conceptual model of the moderating impact of the cultural characteristics of customer i from country k in period t on purchase frequency and contribution margin involves jointly estimating the two equations of the components of CLV and also modeling the marketing cost. The first equation models purchase frequency (Equation 7), the second equation models contribution margin (Equation 8), and the third equation models direct marketing cost (Equation 9).
Modeling Purchase Frequency. Purchase frequency (PFREQ) for the current period is a function of multi- channel buying behavior (MCB), cross-buying (CB), loy- alty, product returns (RET), direct marketing cost (DMC), past purchase frequency, age, household income (HHI), advertising (ADV), and GDP per capita in the previous period. Purchase frequency varies across countries (k), consumers (i), and time (t). The model specification for purchase frequency is as follows:
PFREQikt = a0 + a1kMCBik,t-1 + a2kCBik,t-1 + a3kLoyaltyik,t-1 + a4kRETik,t-1 + a5kDMCik,t-1 + a6kPFREQik,t-1 + a7kAgeik,t-1 + a8kHHIik,t-1 + ADVik,t-1 + a9kIMRikt + a10kGDPk,t-1 + 2ikt.
(2)
As we hypothesized, several cultural dimensions of a country can affect the impact of the drivers of purchase frequency and contribution margin. To test our hypoth- eses, we use the following model specification. For ex- ample, a1k can explain the effect of multichannel buying on purchase frequency; however, a1k can be specified in line with the hypothesized relationship as follows:
a1k = b0 + b1INDIk + b2UCAk + b3LTOk + b4INDUk + e1k.
(3)
Similarly,
a2k = g0 + g1INDIk + g2UCAk + g3LTOk + g4MASk + g5INDUk + e2k;
(4)
a3k = d0 + d1INDIk + d2UCAk + d3LTOk + d4INDUk + e3k;
(5)
and
a4k = q0 + q1INDIk + q2UCAk + e4k,(6)
where INDIk is the level of individualism in a country, UCAk is the level of uncertainty avoidance in a country, LTOk is the level of long-term orientation in a coun- try, MASk is the level of masculinity in a country, INDUk is the level of indulgence in a country, and GDPkt is the GDP per capita of country k. Substituting Equations 3–6 in Equation 2 yields the following model:
(7) PFREQikt = a0 + b0MCBik,t-1 + b1INDIk · MCBik,t-1 + b2UCAk · MCBik,t-1 + b3LTOk · MCBik,t-1 + b4INDUk · MCBik,t-1 + g0CBik,t-1 + g1INDIk · CBik,t-1 + g2UCAk · CBik,t-1 + g3LTOk · CBik,t-1 + g4MASk · CBik,t-1 + g4MASk · CBik,t-1 + g4INDUk · CBik,t-1 + d0Loyaltyik,t-1 + d1INDIk · Loyaltyik,t-1 + d2UCAk · Loyaltyik,t-1 + d3LTOk · Loyaltyik,t-1 + d4INDUk · Loyaltyik,t-1 + q0RETik,t-1 + q1INDIk · RETik,t-1 + q2UCAk · RETik,t-1 + a5kDMCik,t-1 + a6kPFREQik,t-1 + a7kAgeik,t-1 + a8kHHIik,t-1 + a9kIMRikt + a10kGDPk,t-1 + a11kADVik,t-1 + 2ikt.
Modeling Contribution Margin. As mentioned pre- viously, we define the gross contribution margin as the revenue the firm receives from the customers whenever they make a purchase, less the cost of goods sold. Al- though the cost of goods sold does not change much over time, large variances in the revenue a customer provides over time are possible depending on purchase frequency and type of purchase. Thus, we operation- alize the contribution margin variable as contribution margin per purchase. Contribution margin in the cur- rent year is a function of purchase frequency, cross- buying, returns, marketing cost, gross contribution margin per purchase in the previous year, cultural factors, and economic factors. The step-by-step process to arrive at Equation 8 is similar to that in modeling purchase fre- quency (Equation 7):
12 Journal of International Marketing
(8) GCMikt = r0 + p0MCBik,t-1 + p1INDIk · MCBik,t-1 + p2UCAk · MCBik,t-1 + p3LTOk · MCBik,t-1 + p4INDUk · MCBik,t-1 + s0CBik,t-1 + s1INDIk · CBik,t-1 + s2UCAk · CBik,t-1 + s3LTOk · CBik,t-1 + s4MASk · CBik,t-1 + s5INDUk · CBik,t-1 + q0Loyaltyik,t-1 + q1INDIk · Loyaltyik,t-1 + q2UCAk · Loyaltyik,t-1 + q3LTOk · Loyaltyik,t-1 + q4INDUk · Loyaltyik,t-1 + w0RETik,t-1 + w1INDIk · RETik,t-1 + w2UCAk · RETik,t-1 + w3LTOk · RETik,t-1 + w4MASk · RETik,t-1 + j0PFREQik,t-1 + j1INDIk · PFREQik,t-1 + j2UCAk · PFREQik,t-1 + j3LTOk · PFREQik,t-1 + j4MASk · PFREQik,t-1 + j5INDUk · PFREQik,t-1 + r6kAgeik,t-1 + r7kHHIik,t-1 + r8kGDPk,t-1 + r9kDMCik,t-1 + r11kADVik,t-1 + r10kGCMik,t-1 + r11kIMRikt + Eikt.
Modeling the Direct Marketing Cost. We compute the direct marketing cost for each customer on the basis of the marketing activities directed to each customer. The term DMCikt in Equation 9 is the marketing cost for customer i in country k at period t. The marketing efforts of a firm are endogenously determined by the firm’s marketing team according to customers’ past (and expected) be- haviors. A good instrument in this context is one that influences the firm’s decision to implement marketing efforts but not the customer’s financial decision making (Petersen, Kushwaha, and Kumar 2015). Therefore, the level of marketing efforts a firm expends currently is dependent on past contribution margin and past purchase frequency, along with the economic condition of the country. We also add the lagged marketing cost to avoid any omitted variable effects.
DMCikt = to + t1iCMik,t-1 + t2iDMCik,t-1 + t3PFREQik,t-1 + t4GDPk,t-1 + eikt.
(9)
Modeling Challenges
Heterogeneity. In different countries, consumers respond differently to marketing messages. Because our data set contains transaction data of consumers across the world, there will be differences across countries in consumer responses to marketing messages. Therefore, accounting for customer heterogeneity across countries is warranted in our study. We use the GDP per capita as an indicator of
the economic climate of a country, but we do not observe political tensions or short-term changes in the economy, which can lead to unobserved heterogeneity.
Simultaneity and Endogeneity. In our study, we consider differences among the countries by accounting for the national culture and economic factors. Furthermore, we examine the impact of the drivers on the components of CLV (purchase frequency, contribution margin, and marketing cost). The drivers affect purchase frequency and contribution margin, and purchase frequency affects contribution margin; thus, there is a simultaneity bias. The same is true for the marketing cost, which affects both purchase frequency and contribution margin, and we estimate it as a function of past frequency, past contri- bution margin, and past marketing cost. Therefore, the model needs to be estimated simultaneously. However, the effect of a customer’s past purchase frequency and past contribution margin affects his or her current pur- chase frequency and contribution margin. Therefore, to correct for endogeneity, we use Arellano and Bond’s (1991) procedure, which can help derive a consistent generalized method of moments estimator for the pa- rameters of this model. After identifying the instrumental variables from Arellano and Bond’s approach, we specify the variables in the respective models of purchase fre- quency, contribution margin, and direct marketing cost.
Model Estimation
In Equations 7, 8, and 9, the errors associated with the dependent variables may be correlated because contri- bution margin is a function of purchase frequency, lagged contribution margin and lagged purchase frequency are used as predictors for estimating marketing cost, and lagged marketing cost is used to predict purchase fre- quency and contribution margin. We need to account for the correlated errors when we estimate the model. Therefore, we use a seemingly unrelated regression (SUR) model; SUR can be estimated in SAS or by the CMP procedure in Stata. We use both SAS 9.4 and Stata 14.0 to estimate our models; we attain similar results using both programs.
RESULTS Main Effect
In our analysis, we first estimate the relationship of the drivers of CLV with each of the CLV components. The overall model fits for purchase frequency, contribution margin, and marketing cost are Radj2 = .38, Radj2 = .39,
National Culture, Economy, and Customer Lifetime Value 13
and Radj2 = .64, respectively. Table 2 provides the stan- dardized estimates for all the parameters of the purchase frequency, contribution margin, and direct marketing models. As expected, there was a positive relationship between all the drivers of CLV and the CLV components, and an inverse U-shaped relationship between product returns and contribution margin. We find a positive effect of multichannel buying on purchase frequency (.304,
p < .01) and contribution margin (.231, p < .01). Simi- larly, we find a positive effect of cross-buying on purchase frequency (.261, p < .01) and contribution margin (.202, p < .01) and a positive effect of loyalty program enroll- ment on purchase frequency (.242, p < .00) and contri- bution margin (.189, p < .01). Product returns have a positive relationship to purchase frequency (.209, p < .01) and an inverse U-shaped relationship to contribution
Table 2. Parameter Estimates for the Purchase Frequency, Contribution Margin, and Direct Marketing Cost Models
Lagged Variables
Standardized Coefficients
Purchase Frequency
Hypothesis Tested
Hypothesis Supported?
Contribution Margint
Marketing Costt
Hypothesis Tested
Hypothesis Supported?
Direct Effect Multichannel buying (MCB) .304 .231 Cross-buying (CB) .261 .202 Loyalty programs (LP) .242 .189 Product returns (RET) .209 .161 Product returns R2 N.A. −.02 Purchase frequency (PFREQ) .276 .257 .421 Inverse Mills ratio (IMRt) .092 .072 Contribution margin (CM) N.A. .216 .301 Marketing cost (MC) .198 .177 .578 Advertising spend (ADV) .096 .109 .268
Moderating Effect With Individualism (INDI) MCB × INDI .142 1a Yes .128 1b Yes CB × INDI −.106 1c Yes −.126 1d Yes LP × INDI −.116 1e Yes −.099 1f Yes RET × INDI .128 1g Yes .122 1h Yes
With Uncertainty Avoidance (UCA) MCB × UCA −.089 2a Yes −.066 2b Yes CB × UCA .112 2c Yes .107 2d Yes LP × UCA .263 2e Yes .217 2f Yes RET × UCA −.096 2g Yes −.077 2h Yes
With Long-Term Orientation (LTO) MCB × LTO −.106 3e Yes −.113 3f Yes CB × LTO .136 3a Yes .129 3b Yes LP × LTO .176 3c Yes .152 3d Yes
With Masculinity (MAS) CB × MAS −.063 4a Yes −.081 4b Yes
With Indulgence (INDU) MCB × INDU −.091 5e Yes −.083 5f Yes CB × INDU .079 5a Yes .094 5b Yes LP × INDU −.108 5c Yes −.117 5d Yes
GDP .231 6a Yes .241 .162 6b Yes
Notes: All parameter estimates are significant at least at the .05 level. N.A. = not applicable.
14 Journal of International Marketing
margin (.161 for the linear term, p < .01; –.02 for the quadratic term). Past purchase frequency has a positive impact on current purchase frequency (.276, p < .01) and a positive impact on contribution margin (.216, p < .01). To incorporate the company’s marketing cost, we included the advertising and direct marketing costs of the company in our equation. The results indicate a positive effect of advertising on purchase frequency (.096, p < .01) and contribution margin (.109, p < .01). We also find a positive relationship between direct marketing cost and purchase frequency (.198, p < .01) and between direct marketing cost and contribution margin (.177, p < .01). These results are in line with those of Reinartz and Kumar (2003), who examine CLV in the B2B context, and with those of Kumar, Shah, and Venkatesan (2006), who focus on the B2C context.
After establishing the relationship between the drivers and the components of CLV, we wanted to understand how national culture moderates this relationship. Therefore, we modeled the interaction of the national cultural dimensions with the respective drivers of pur- chase frequency and contribution margin. The overall model fit improved from an Radj2 of .38 to an Radj2 of .61 for purchase frequency and from an Radj2 of .39 to an Radj2 of .59 for contribution margin when we included the cultural dimensions. This indicates that the cultural di- mensions moderate the relationship of the drivers of purchase frequency and contribution margin.
Moderating Impact of Individualism
Effect of Multichannel Buying on Purchase Frequency and Contribution Margin. We find that the effect of multichannel buying on purchase frequency and contri- bution margin is greater in countries high in individualism than in countries low in individualism. The parameter estimates for multichannel buying × individualism on purchase frequency (.142, p < .01) and contribution margin (.128, p < .01) are positive and significant, thus providing empirical support for the hypothesized re- lationship (H1a–b).
Effect of Cross-Buying on Purchase Frequency and Contribution Margin. In countries low in individualism, consumers buy not only for themselves but also for their families, and thus they are willing to try new de- partments of the same store. The negative coefficients of cross-buying × individualism on purchase frequency (–.106, p < .01) and contribution margin (–.126, p < .01) indicate that the magnitude of the effect is lower in individualist countries. This finding provides support
for our hypotheses (H1c–d) that the lower the indi- vidualism in a country, the greater is the positive impact of cross-buying on purchase frequency and contribution margin.
Effect of Loyalty Programs on Purchase Frequency and Contribution Margin. As discussed, consumers in in- dividualist societies are fashion conscious and therefore are loyal to fashion trends more than to a specific retailer. Consequently, the impact of individualism on the mag- nitude of the effect of loyalty programs on purchase frequency (–.116, p < .01) and contribution margin (–.099, p < .01) is negative. This implies that the higher the individualism in a country, the lesser is the effect of the relationship of loyalty programs on purchase frequency and contribution margin, in support of H1e–f.
Effect of Product Returns on Purchase Frequency and Contribution Margin. In countries high in individualism, the level of trust is higher and perceived risk is lower. Therefore, these countries tend to have lenient return policies, enhancing the relationship of returns with pur- chase frequency and contribution margin. The parameter estimates of returns × individualism on purchase fre- quency (.128, p < .01) and contribution margin (.122, p < .01) provide support for H1g–h.
Moderating Impact of Uncertainty Avoidance
Effect of Multichannel Buying on Purchase Frequency and Contribution Margin. Consumers in countries high in uncertainty avoidance are prone to reduce risk, and thus they choose to shop through traditional channels rather than buying online or from catalogs. This activity is evident from the parameter estimates of multichannel buying × uncertainty avoidance on purchase frequency (–.089, p < .01) and contribution margin (–.066, p < .01). The resultimpliesthat the magnitude of the positiverelationship between multichannel buying × purchase frequency and multichannel buying × contribution margin is lower in countries high in uncertainty avoidance (higher in countries low in uncertainty avoidance), providing support for H2a–b.
Effect of Cross-Buying on Purchase Frequency and Contribution Margin. In countries high in uncertainty avoidance, consumers avoid risk and rely on established sources they trust. Therefore, they buy across de- partments from their trusted retailer. We find that the magnitude of the effect of cross-buying × uncertainty avoidance on purchase frequency (.112, p < .01) and contribution margin (.107, p < .01) is positive, in support of H2c.
National Culture, Economy, and Customer Lifetime Value 15
Effect of Loyalty Programs on Purchase Frequency and Contribution Margin. Enrolling in a loyalty program is an indication of a consumer’s trust in the firm. We find support for this impact in the higher coefficient of the interaction term of loyalty program × uncertainty avoidance on purchase frequency (.263, p < .01) and contribution margin (.217, p < .01), confirming the hy- pothesized relationship (H2e–f) that the higher the un- certainty avoidance in a country, the greater is the positive impact of owning a loyalty card on purchase frequency and contribution margin.
Effect of Product Returns on Purchase Frequency and Contribution Margin. Stores in low-uncertainty-avoidance countries have lenient return policies. This is reflected in the strength of the interactive relationship of product return with uncertainty avoidance on purchase frequency (–.096, p < .01) and contribution margin (–.077, p < .01) in countries low in uncertainty avoidance. Thus, there is support for H2g–h.
Moderating Impact of Long-Term Orientation
Effect of Multichannel Buying on Purchase Frequency and Contribution Margin. In long-term-oriented cultures, people prefer using cash or debit cards versus credit cards (De Mooij and Hofstede 2002). This preference may affect the different channels they can buy from because not all channels support cash transactions. Thus, the magnitude of the effect of long-term orientation on the relationship between multichannel buying and purchase frequency and contribution margin is lower. That is, the interaction effect of long-term orientation on multichannel buying is negative. The results are in line with H3a–b, with parameter estimates of −.106 (p < .01) for purchase fre- quency and –.113 (p < .01) for contribution margin.
Effect of Cross-Buying on Purchase Frequency and Contribution Margin. Future customer profitability in- creases when customers purchase more product cate- gories from the retailer (Kumar, Shah, and Venkatesan 2006). Our findings confirm this relationship in different cultures. Cross-buying is an indication of consumers’ trust in the firm. Consumers in countries with a long-term orientation trust well-established resources and plan to build lasting relationships. Our parameter estimates for the interaction effect of cross-buying × long-term orien- tation on purchase frequency (.136, p < .01) and con- tribution margin (.129, p < .01) indicate that the relationship between cross-buying and purchase fre- quency and contribution margin is enhanced in long- term-oriented countries, in support of H3c–d.
Effect of Loyalty Programs on Purchase Frequency and Contribution Margin. Consumers in countries with long- term orientation are ideal for loyalty programs because they plan for the future and thus likely plan on how to use the benefits from these programs. Therefore, the effect of loyalty programs on purchase frequency and contribution margin is magnified in countries with long-term orien- tation. The parameter estimates of loyalty programs × long-term orientation on purchase frequency (.176, p < .01) and contribution margin (.152, p < .01) provide support for H3e–f.
Moderating Impact of Masculinity
Effect of Cross-Buying on Purchase Frequency and Contribution Margin. Consumers who are conscious about marketplace trends do not restrict themselves to one store but, rather, shop across stores to buy the trendiest products. Therefore, the higher the mascu- linity in a country, the lower is the number of de- partments (of the same store) from which consumers shop. Therefore, the magnitude of the interaction effect of cross-buying × masculinity on purchase frequency (–.063, p < .01) and contribution margin (–.081, p < .01) is lower in countries high in masculinity, in support of H4a–b.
Moderating Impact of Indulgence
Effect of Multichannel Buying on Purchase Frequency and Contribution Margin. Consumers in high- indulgence countries are prone to seek immediate gratification and thus choose to shop through tradi- tional channels to instantly experience the product. This finding is evident from the parameter estimates of multichannel buying × indulgence on purchase fre- quency (–.091, p < .01) and contribution margin (–.083, p < .01). This result implies that the magnitude of the positive relationship between multichannel buying × purchase frequency and multichannel buying × con- tribution margin is lower in countries high in in- dulgence (higher in countries low in indulgence), providing support for H5a–b.
Effect of Cross-Buying on Purchase Frequency and Contribution Margin. In highly indulgent countries, consumers explore new things and thus are open to buying from multiple departments. They buy across de- partments to enjoy the widespread choice. Therefore, the magnitude of the effect of cross-buying × indulgence on purchase frequency (.079, p < .01) and contribution margin (.094, p < .01) is positive, in support of H5c–d.
16 Journal of International Marketing
Effect of Loyalty Programs on Purchase Frequency and Contribution Margin. Enrolling in a loyalty program constrains consumers to buy from the same store and use the benefits of the loyalty program. Therefore, consumers in indulgent societies do not prefer loyalty programs. This is evident in the negative coefficient of the interaction term of loyalty program × indulgence on purchase frequency (–.108, p < .01) and contribution margin (–.117, p < .01). This finding provides support for our hypothesized re- lationship (H5e–f) that the lower the indulgence in a country, the greater is the positive impact of owning a loyalty card on purchase frequency and contribution margin.
Economic Growth on Purchase Frequency and Contribution Margin
Although a country’s cultural dimensions affect consumer behavior, its economic condition also plays an important role in consumers’ ability to purchase goods. Therefore, we estimate our model with an economic growth variable (GDP per capita). The overall model fit improves from an Radj2 of .61 to an Radj2 of .66 for purchase frequency and from an Radj2 of .59 to an Radj2 of .64 for contribution margin. The higher the economic growth of the country, the higher is consumers’ disposable income, which pos- itively affects their ability to buy. Therefore, the effect of economic growth in countries with a higher GDP per capita is greater on purchase frequency (vs. countries with lower GDP per capita; .231, p < .01) and contribution margin (.241, p < .01), in support of H6a–b.
Computing CLV
The metric CLV is the sum of cumulated cash flows, discounted with the weighted average cost of capital of a customer during his or her lifetime with a company (Kumar 2008). According to Reinartz and Kumar (2003), the lifetime of the customer at a given time is three years. As we show in Equation 1, CLV is a function of the predicted contribution margin, predicted purchase fre- quency, and predicted marketing resources allocated to the customer, adjusted for time. We provide the mean CLV of customers in all the countries used in our data in Table 3.
IMPLICATIONS
To successfully maximize CLV around the world, man- agers need to understand the differences in the cultural and economic dimensions across countries. They also must recognize the impact of national culture on the
magnitude of the relationship between the drivers of the components of CLV and the effect of the economy on purchase frequency and contribution margin. Table 4 summarizes the effects of the cultural dimensions on the drivers of purchase frequency and contribution margin.
In addition, managers should strategically allocate their resources between the drivers of purchase frequency and contribution margin according to the differences in the cultural dimensions across countries. Countries high in individualism have strengthened relationships of multi- channel buying, cross-buying, and product returns with purchase frequency and contribution margin. However, the effect of loyalty program enrollment on purchase frequency and contribution margin is weak in countries high in individualism. This implies that firms in in- dividualist countries should focus on promoting multi- channel buying and buying across departments and implement a lenient return policy, with minimal focus on loyalty cards.
Similarly, the relationships of multichannel buying, cross- buying, and loyalty programs enrollment with purchase frequency and contribution margin is strengthened in high-uncertainty-avoidance countries. In this situation, managers of firms involved in global operations may face conflicts in the effect of their loyalty programs on pur- chase frequency and contribution margin if a country is high in both individualism and uncertainty avoidance. How can managers resolve this dilemma? If there were only one driver to consider, managers could evaluate the relative effect of each cultural dimension on this driver of purchase frequency and contribution margin. However, there are multiple drivers, and the relative effect of these cultural dimensions on the drivers of the CLV compo- nents shows a contrast. For example, Country C1’s high score on individualism reflects a higher impact on mul- tichannel buying, and its high score on uncertainty avoidance reflects a lower impact on multichannel buying (Table 3).
To solve this dilemma, managers could analyze Equations 7 and 8 for every country to understand the impact of the interaction between the cultural dimension and the drivers of CLV. For example, for country C1, the co- efficient of multichannel buying × individualism is .142, and the coefficient of cross-buying × individualism is –.106. This indicates that an increase in one additional channel would increase purchase frequency, and one more in- stance of cross-buying would have a negative impact on purchase frequency. Therefore, managers should focus on improving their firms’ multichannel activities in
National Culture, Economy, and Customer Lifetime Value 17
individualist countries. However, analyzing the equation with just one cultural dimension would again raise a dilemma if the coefficient for the interaction between high uncertainty avoidance and multichannel buying was lower than the coefficient of the interaction between high uncertainty avoidance and cross-buying. This leads to the
question of which cultural dimension is more important. Because all dimensions are representative of the charac- teristics of consumers in the country, managers cannot disregard any. Therefore, they should evaluate the net effect of the cultural dimensions on the drivers of pur- chase frequency and contribution margin and focus on
Table 3. Average Predicted CLV at the End of the Third Year, by Country
Country Code
Levels of Cultural Dimensions Predicted CLV for the Next Three Years (in 2008 USD)Individualism
Uncertainty Avoidance
Long-Term Orientation Masculinity Indulgence
C1 Highest High Low High High 389 (114)
C2 High High High Highest High 207 (98)
C3 High Highest Highest High High 326 (107)
C4 Medium Highest High Medium Low 168 (71)
C5 Low Highest Medium Medium Low 499 (216)
C6 Low Highest Low Medium Low 261 (131)
C7 High Highest High Medium Medium 422 (206)
C8 High High Highest High Medium 187 (111)
C9 Low Medium High High Low 218 (134)
C10 High Medium Low High High 266 (99)
C11 High Highest Medium Medium Low 249 (108)
C12 Highest High High High Medium 192 (71)
C13 Medium Highest Highest Highest Medium 284 (112)
C14 Medium Medium Medium Medium High 307 (105)
C15 Medium High Low High Low 393 (138)
C16 Highest High High Low High 454 (156)
C17 High Medium Medium Low High 465 (142)
C18 Medium Medium Medium High Medium 212 (86)
C19 Medium Highest Highest Medium Low 194 (93)
C20 Low Highest Medium High High 255 (116)
C21 Low Low High Medium Medium 343 (143)
C22 High Highest Medium Medium Medium 457 (158)
C23 High Medium High Low Highest 479 (152)
C24 High High High High High 365 (128)
C25 Low High Highest Medium Medium 286 (106)
C26 Low High Medium Medium Medium 266 (113)
C27 Medium Highest Medium Medium Medium 169 (72)
C28 Highest Medium Medium High High 294 (126)
C29 Highest Medium High High High 309 (133)
C30 Low Medium High Medium Medium 418 (218)
Notes: Standard deviations for CLV predictions are in parentheses.
18 Journal of International Marketing
the drivers whose net effect is the highest, after accounting for all the relevant cultural dimensions. The net effect can be computed by using Equation 10, which includes the parameter estimate from Table 2 and the standardized value of the cultural dimensions:
Net Effect of MCB = ðb0 + b1INDIk + b2UCAk + b3LTOk + b4INDUkÞ
(10)
(11) = :304 + ð:142 · -1:2387Þ + ð -:089 · -2:28Þ + ð -:106 · :88Þ + ð -:091 · :056Þ
= :2320.
Similarly, we can compute the net effect of cross-buying as follows:
(12) Net Effect of CB = :261 + ð-:106 · -1:23Þ + ð:112 · -2:28Þ + ð:136 · :88Þ + ð-:079 · :56Þ + ð-:063 · -:09Þ
= :2469.
In Equations 11 and 12, we demonstrate the net effect of multichannel buying and cross-buying on all the cultural dimensions and the coefficients of the interaction of the cultural dimensions with the drivers of purchase fre- quency. The net effect of multichannel buying is .2320, and the net effect of cross-buying is .2469. These numbers indicate that the overall relationship of multichannel buying and cross-buying with purchase frequency changes when all the cultural dimensions are considered, as com- pared with evaluating only one cultural dimension. Thus, in country C1 managers should focus resources on pro- moting cross-buying. This example helps highlight the importance of analyzing all the cultural dimensions when making strategic decisions. Resource allocation decisions across countries can be efficient if managers compute the
net effect of all the drivers of purchase frequency and contribution margin for all countries and use the net effect values for decision making.
Global companies likely expect that the relationship of all the drivers of purchase frequency and contribution margin is strengthened in developed countries because consumers have more disposable income and thus likely shop across channels and departments, return fewer products, and enroll in loyalty programs. However, in considering cultural factors, such firms may draw a dif- ferent picture. For example, managers might expect a higher purchase frequency from consumers in China by examining the impact of all the drivers. However, China is a collectivist country, and the magnitude of the re- lationship between multichannel buying and purchase frequency tends to be lower in such countries. Therefore, even if firms plan to invest across channels because of a growing economy or increased Internet usage, this strategy might not work because of the differential be- havior of consumers.
CONCLUSION AND LIMITATION
This study highlights the importance of both the cultural and the economic dimensions of a country for maxi- mizing firm profits. We demonstrate the impact of cul- tural and economic factors on the drivers of purchase frequency and contribution margin by estimating our model using SUR. The results indicate that all the cultural dimensions moderate the relationship between the drivers of purchase frequency and contribution margin and that the economy of a country has a direct impact on purchase frequency and contribution margin.
We restricted our study to 30 countries and one retailer only, and therefore, the scope of the study is limited.
Table 4. Effect of Cultural Dimensions on the Drivers of Purchase Frequency and Contribution Margin
Cultural Dimension
Relative Effect of Driver with High (vs. Low) Level of Cultural Dimension
Multichannel Buying Cross-Buying Enrollment in Loyalty Program Product Returns
Individualism Greater Less Less Greater
Uncertainty avoidance Less Greater Greater Less
Long-term orientation Less Greater Greater N.A.
Masculinity N.A. Less N.A. N.A.
Individualism Less Greater Less N.A.
Notes: The higher the individualism of a country, the greater is the effect of MCB on purchase frequency and contribution margin. N.A. = not applicable.
National Culture, Economy, and Customer Lifetime Value 19
Further research could extend this study to more than 30 countries and across industries, which would help de- termine the relative importance of culture and economic dimensions across industries. Furthermore, a comparison across industries would shed light on which industries are influenced the most by which culture; this would help a business conglomerate allocate its resources efficiently across businesses. Companies have long localized their marketing content according to the culture of the country and the trends being followed there. What is the effect of this varying marketing content on an MNC’s profits? Research could also examine the change in the effects of cultural dimensions due to the increased presence of social media and its influence on firm profits. Whereas our study is positioned as an empirical inquiry, further research could develop a stronger theoretical rationale if the current findings are consistent across industries.
NOTES
1. See http://www.bbc.com/travel/story/20121213-different- disneylands-around-the-world.
2. See http://www.fundinguniverse.com/company-histories/ c-a-history/.
REFERENCES Aaker, Jennifer L. and Patti Williams (1998), “Empathy Versus Pride: The Influence of Emotional Appeals Across Cultures,” Journal of Consumer Research, 25 (3), 241–61.
Anderson, James C. and James A. Narus (1991), “Partnering as a Focused Market Strategy,” California Management Re- view, 33 (3), 95–113.
Arellano, Manuel and Stephen Bond (1991), “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations,” Review of Economic Studies, 58 (2), 277–97.
Bendapudi, Neeli and Leonard L. Berry (1997), “Customers’ Motivations for Maintaining Relationships with Service Providers,” Journal of Retailing, 73 (1), 15–37.
Cavusgil, S. Tamer (1998), “Knowledge Development in In- ternational Marketing,” Journal of International Marketing, 6 (2), 103–12.
Clarke, Darral G. (1976), “Econometric Measurement of the Duration of Advertising Effect on Sales,” Journal of Mar- keting Research, 13 (November), 345–57.
De Mooij, Marieke and Geert Hofstede (2002), “Conver- gence and Divergence in Consumer Behavior: Implications
for International Retailing,” Journal of Retailing, 78 (1), 61–69.
Drèze, Xavier and Andre Bonfrer (2003), “To Pester or Leave Alone: Lifetime Value Maximization Through Optimal Communication Timing,” working paper, Research Collec- tion Lee Kong Chian School of Business.
Dwyer, Sean, Hani Mesak, and Maxwell Hsu (2005), “An Exploratory Examination of the Influence of National Culture on Cross-National Product Diffusion,” Journal of In- ternational Marketing, 13 (2), 1–27.
Fader, Peter S., Bruce G.S. Hardie, and Ka Lok Lee (2005), “RFM And CLV: Using Iso-Value Curves For Customer Base Analysis,” Journal of Marketing Research, 42 (November), 415–30.
Garbarino, Ellen and Mark S. Johnson (1999), “The Different Roles of Satisfaction, Trust, and Commitment in Customer Relationships,” Journal of Marketing, 63 (April), 70–87.
Griffith, David and Gaia Rubera (2014), “A Cross-Cultural Investigation of New Product Strategies for Technological and Design Innovations,” Journal of International Marketing, 22 (1), 5–20.
Gudykunst, William B. (1997), “Cultural Variability in Com- munication: An Introduction,” Communication Research, 24 (4), 327–48.
Henry, Walter A. (1976), “Cultural Values Do Correlate with Consumer Behavior,” Journal of Marketing Research, 13 (May), 121–27.
Hofstede, Geert (1980), “Motivation, Leadership, and Orga- nization: Do American Theories Apply Abroad?” Organiza- tional Dynamics, 9 (1), 42–63.
——— (1991), Cultures and Organisations—Software of the Mind: Intercultural Cooperation and Its Importance for Survival. London: McGraw-Hill.
——— and Michael Harris Bond (1988), “The Confucius Connection: From Cultural Roots to Economic Growth,” Organizational Dynamics, 16 (4), 5–21.
——— and Gert Jan Hofstede (2001), Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organiza- tions Across Nations, 2d ed. Thousand Oaks, CA: Sage Publications.
———, ———, and Michael Minkov (2010)). Cultures and Organizations. New York: McGraw-Hill.
Iyengar, Sheena S. and Mark R. Lepper (1999), “Rethinking the Value of Choice: A Cultural Perspective on Intrinsic Moti- vation,” Journal of Personality and Social Psychology, 76 (3), 349–66.
20 Journal of International Marketing
Kacen, Jacqueline J. and Julie Anne Lee (2002), “The Influence of Culture on Consumer Impulsive Buying Behavior,” Journal of Consumer Psychology, 12 (2), 163–76.
Kumar, V. (2008), Managing Customers for Profit: Strategies to Increase Profits and Build Loyalty. Upper Saddle River, NJ: Prentice Hall.
——— and Bharath Rajan (2012), “Social Coupons as a Marketing Strategy: A Multifaceted Perspective,” Journal of the Academy of Marketing Science, 40 (1), 120–36.
———, Denish Shah, and Rajkumar Venkatesan (2006), “Managing Retailer Profitability—One Customer at a Time!” Journal of Retailing, 82 (4), 277–94.
———, Rajkumar Venkatesan, Tim Bohling, and Denise Beckmann (2008), “Practice Prize Report—The Power of CLV: Managing Customer Lifetime Value at IBM,” Mar- keting Science, 27 (4), 585–99.
Lastovicka, John L., Lance A. Bettencourt, Renee Shaw Hughner, and Ronald J. Kuntze (1999), “Lifestyle of the Tight and Frugal: Theory and Measurement,” Journal of Consumer Research, 26 (1), 85–98.
Markus, Hazel R. and Shinobu Kitayama (1991), “Culture and the Self: Implications for Cognition, Emotion, and Motiva- tion,” Psychological Review, 98 (2), 224–53.
Matthyssens, Paul and Huib Wursten (2003), Internal Mar- keting: Cross-Cultural Marketing. South Melbourne, Aus- tralia: Thomson Learning, 243–56.
Money, R. Bruce, Mary C. Gilly, and John L. Graham (1998), “Explorations of National Culture and Word-of-Mouth Referral Behavior in the Purchase of Industrial Services in the United States and Japan,” Journal of Marketing, 62 (October), 76–87.
Morgan, Robert M. and Shelby D. Hunt (1994), “The Commitment–Trust Theory of Relationship Marketing,” Journal of Marketing, 58 (July), 20–38.
Nakata, Cheryl and K. Sivakumar (1996), “National Culture and New Product Development: An Integrative Review,” Journal of Marketing, 60 (January), 61–72.
Patterson, Paul G. and Tasman Smith (2003), “A Cross-Cultural Study of Switching Barriers and Propensity to Stay with Service Providers,” Journal of Retailing, 79 (2), 107–20.
Petersen, J. Andrew, Tarun Kushwaha, and V. Kumar (2015), “Marketing Communication Strategies and Consumer Fi- nancial Decision Making: The Role of National Culture,” Journal of Marketing, 79 (January), 44–63.
Pfeiffer, Philip E. (2011), “On Estimating Current-Customer Equity Using Company Summary Data,” Journal of In- teractive Marketing, 25 (1), 1–14.
Pizam, Abraham and Arie Reichel (1996), “The Effect of Na- tionality on Tourist Behavior: Israeli Tour-Guides’ Percep- tions,” Journal of Hospitality & Leisure Marketing, 4 (1), 23–49.
Reinartz, Werner J. and V. Kumar (2000), “On the Profitability of Long-Life Customers in A Noncontractual Setting: An Empirical Investigation and Implications for Marketing,” Journal of Marketing, 64 (October), 17–35.
——— and ——— (2003), “The Impact of Customer Re- lationship Characteristics on Profitable Lifetime Duration,” Journal of Marketing, 67 (January), 77–99.
Ricks, David A. (1993), Blunders in International Business. Malden, MA: Blackwell.
Rogers, Everett M. (1983), Diffusion of Innovations, 3rd ed. New York: The Free Press.
Roth, Martin S. (1995), “The Effects of Culture and Socioeco- nomics on the Performance of Global Brand Image Strategies,” Journal of Marketing Research, 32 (May), 163–75.
Schwartz, Shalom H. (1994), Beyond Individualism/Collectivism: New Cultural Dimensions of Values. Thousand Oaks, CA: Sage Publications.
Steenkamp, Jan-Benedict E.M. (2001), “The Role of National Culture in International Marketing Research,” International Marketing Review, 18 (1), 30–44.
———, Frenkel ter Hofstede, and Michel Wedel (1999), “A Cross-National Investigation into the Individual and National Cultural Antecedents of Consumer Innovativeness,” Journal of Marketing, 63 (April), 55–69.
Thomas, Jacquelyn S., Robert C. Blattberg, and Edward J. Fox (2004), “Recapturing Lost Customers,” Journal of Marketing Research, 41 (February), 31–45.
Triandis, Harry C. (1989), “The Self and Social Behavior in Differing Cultural Contexts,” Psychological Review, 96 (3), 506–20.
Tse, David K., Lee Kam-hon, Ilan Vertinsky, and Donald A. Wehrung (1988), “Does Culture Matter? A Cross-Cultural Study of Executives’ Choice, Decisiveness, and Risk Adjust- ment in International Marketing,” Journal of Marketing, 52 (October), 81–95.
Venkatesan, Rajkumar and V. Kumar (2004), “A Customer Lifetime Value Framework for Customer Selection and Resource Allocation Strategy,” Journal of Marketing, 68 (October), 106–25.
Verhoef, Peter C. and Bas Donkers (2001), “Predicting Cus- tomer Potential Value: An Application in the Insurance In- dustry,” Decision Support Systems, 32 (2), 189–99.
National Culture, Economy, and Customer Lifetime Value 21
Copyright of Journal of International Marketing is the property of American Marketing Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.