key: cord-0855065-1408e0qk
authors: Chandler, Jennifer D.; Graham, John L.
title: Relationship-Oriented Cultures, Corruption, and International Marketing Success
date: 2009-08-05
journal: J Bus Ethics
DOI: 10.1007/s10551-009-0152-7
sha: 93f2da2db640bc5cb223858477393f583a4c8b7f
doc_id: 855065
cord_uid: 1408e0qk

This study explores the general problems associated with marketing across international markets and focuses specifically on the role of corruption in deterring international marketing success. The authors do this by introducing a broader conceptualization of corruption. The dimensions of corruption and their importance in explaining the exporters’ successes in international markets are developed empirically. Partial Least Squares formative indicators are used in a comprehensive model including consumer resources (wealth and information resources), physical distance (kilometers and time zones), and cultural distance (linguistic and values differences) as alternative explanatory variables. Finally, differences in the model’s performance across data from three exporting countries (France, Japan, and the US) are delineated and discussed. For example, the successes of French and Japanese exporters in international markets are in part determined by the levels of corruption in target countries. Alternatively, corruption in target countries does not appear to affect the successes of American exporters in global markets. The conceptualization of corruption in this study extends the more narrow view of corruption solely as bribery.

As globalization continues, more firms face decisions about which foreign markets to enter. As a result, managers must assess unfamiliar business landscapes and scrutinize the economic, political, religious, cultural, and economic factors that often influence the successes of business ventures. Along with the obvious market potential, managers also recognize negative societal elements such as bribery or violence that can affect their overseas marketing activities. Questions arise such as: What are potential causes of business losses? What kind of adversities will the firm and its managers be forced to cope with in foreign markets? What fears do managers have in sending employees overseas?

We expect the answers to these questions to affect how successful firms are in international markets. This study generally explores the issues associated with marketing across international borders. We build on the work of Stern and Reve (1980) who provide a theoretical framework for assessing the marketing environment as individual ''political economies'' that can be compared across international markets. Wathne and Heide (2004) confirm the importance of this approach and reassert that ''greater attention must be directed to the larger networks in which relationships exist' ' (p. 73) .

Directing attention to the macro-level networks that affect international marketing, this study focuses specifically on corruption. As an environmental barrier to international marketing efforts, corruption continues to complicate firms' efforts to enter and be successful in particular international markets. Risks to financial performance and employee well-being are greater in markets where rule of law and transparency in decision making are little applied. Increasingly, home country laws and/or corporate ethics preclude commercial operations where corruption is rampant. Thus, managers, firms, and policy makers alike would benefit from unraveling the role that corruption plays in undermining their efforts. As the presence and mechanisms of corruption are difficult to understand, this study begins to explore and quantify corruption as an environmental barrier to international marketing efforts.

Toward these ends, a novel, higher order, multidimensional construct of corruption is introduced. By developing and testing a predictive model centered on an enriched concept of corruption, the analysis provides information about both its systemic and construct validities. Using the Partial Least Squares (PLS) form of structural equation modeling, the authors fit the same model for three datasets including market penetration statistics for exporters from France, Japan, and the United States. Additional novel aspects of the study include the use of new measures of linguistic distance and comparative violence levels as potential antecedents of market success. As it introduces new dimensions of corruption, the conceptualization of corruption in this paper is thus broader than the conventionally accepted notion of corruption solely as bribery.

The remainder of the paper is organized as follows: first, corruption is discussed in light of the political economy and international business literatures, and its influence on international marketing success is delineated. Second, a nomological net is developed by presenting hypotheses and a comprehensive model that include important control variables. Third, the research design, data, and analyses are described. Finally, the results and discussion sections are presented; the latter includes an interpretation of the findings, a comparison with the existing literature, and a description of implications for managers and policy makers.

The concept of corruption

The concept of corruption is centuries old and a variety of local rules and laws have long existed around the world to mitigate corruption's negative effects. In 1977, the US government specifically targeted bribery in foreign markets through passage of the Foreign Corrupt Practices Act (FCPA). The FCPA continues to prohibit American executives and their companies from bribing officials of foreign governments (US Department of Justice, 2004). As foreign governments control large portions of their industries in most of the world, the FCPA applies to a large percentage of export transactions for American firms. Immediately after passage of the law, American executives reported business lost to foreign competitors that were not subject to similar statutes in their home countries (Kaikati and Label, 1980) . Most recently many member countries of the Organization for Economic Cooperation and Development (OECD) and the United Nations have begun enforcing bribery restrictions similar to the FCPA. Thus, bribery of foreign government officials is currently prohibited by laws in all the three exporting countries considered in this study. That is, Japan and the United States in 1999 and France in 2000 joined the OECD Convention. Of course, the United States Congress had passed and President Carter has signed into law the FCPA in 1977.

Previous studies of corruption in foreign markets have focused on bribery as the main kind of corruption. Those studies ignore other kinds of corruption that make ''headlines'' both in the popular press and across the brows of corporate executives because they make international marketing efforts more difficult. Here, we consider four such aspects:

(1) As the value of American intellectual property (IP) burgeoned during the booming 1990s, an increasing number of American executives complained about the ''piracy'' of copyrighted and patented products and services in international markets.

(2) The tragic events of September 11, 2001 have ''terrorism'' and ''violence,'' at home and abroad, a major worry for all Americans. For example, controversial policies at the University of California precluded education abroad programs in Israel and other countries as travel bans are issued by the US State Department. (3) The prevalence of black market activities in foreign markets makes business transactions more risky from both legal and financial perspectives. (4) Finally, affluent countries have become more antiseptic while diseases have become more virulent in developing countries [e.g., the severe acute respiratory syndrome (SARS) outbreak in China and human immunodeficiency virus (HIV) worldwide], thus creating a biological barrier to 252

Jennifer D. Chandler and John L. Graham travel and international commerce (Cateora et al., 2009 ). The growing biological risk for international executives is well represented by the comments of one high-level manager at Shell regarding the necessity of carrying a set of personal syringes when traveling in Nigeria. All of these phenomena affect the attractiveness of overseas markets, can serve as barriers to successful international marketing, and can be viewed as kinds of corruption.

Recent focus has shifted to unraveling the entanglement of sociocultural, political, and economic effects of corruption. Scholars across disciplines acknowledge corruption's persistence in human societies over time and space (Aidt, 2003) , and agree on the nature of corruption as a very complex, pervasive, and intertwined social phenomenon (Park, 2003) . Most imply that corruption is endogenous to a society, usually one element in a cycle of activities that both empowers and obstructs the workings of social, political, and economic institutions. Everett et al. (2006, p. 1) comment on corruption, ''…its definition, consequences, and causes are more numerous than might be first imagined.' ' Aidt (2003, p. 632) concurs, ''Corruption is a many-faceted phenomenon and it is hard to give a precise and comprehensive definition…yet it is important to try.'' Much research has also been done on the determinants of corruption (Aidt, 2003; Davis and Ruhe, 2003; Sanyal, 2005) , the consequences of corruption (Aidt, 2003; Alam, 1995; Maor, 2004; Wei and Shleifer, 2000; Winters, 2004) , and ways to combat corruption (Ali and Isse, 2003; Brademas and Heimann, 1998; Husted, 1999) . It is no wonder that corruption has topped the research agenda of international institutions such as the World Bank, the International Monetary Fund, and the United Nations. Despite the esoteric criticism regarding the concept of corruption, western academics and practitioners have most often employed Transparency International's Corruption Perception Index (TICPI) as the best single measure of the theoretical construct across countries. The TICPI (Transparency International, 2007 ) is a poll of polls, drawing on 14 surveys from seven independent institutions worldwide combining perceptions of business people, academics, and country analysts alike. While it is an international poll, it is problematic because the seven institutions providing survey data are based mostly in western countries, where corruption is most often defined narrowly as bribery. In fact, only one of the reporting institutions is based in a non-western culture, specifically Asia. One study asserts that the attempt to understand corruption from the narrow perspective of bribery translates into an attempt to globalize the western definition of corruption (Sandholtz and Gray, 2004) . It is understandable that managers of various backgrounds and experiences would define corruption differently and most certainly, might not limit its definition to bribery.

The complexity of corruption results in current definitions which are fraught with subjectivity and cultural relativism (Houston and Graham, 2001) ; it is evident that people from different countries have various perceptions and tolerances toward bribery even though it is illegal in most nations (Jing and Graham, 2008) . Donleavy et al. (2008) suggest that in China, corruption has three faces: tanwu (malpractice), whui (bribery), and tequan (privileges). Thus, corruption encompasses more than simple bribery; in fact, it is apparent from the discussion above that a consensus definition eludes scholars. Webster's Dictionary (2007) defines corruption narrowly as ''bribery'' and broadly as ''impairment of integrity, depravity, decay, and decomposition.'' Synonyms listed in Roget's International Thesaurus (2001) include ''bribery, infection, decay, deterioration, and dishonesty.'' Taking cultural relativism to an extreme, it has been argued that rampant consumerism is corrupting India's newly opened markets or that individualism, at the expense of group solidarity, is corrupting Japan's social foundations (Houston and Graham, 2001) . Or, profits and currency speculation, not seen as forms of corruption in the US, are frowned upon by many in the emergent Russian nation.

In this study, corruption is broadly defined as those elements in a market that (1) signal disorganization, decomposition, or deterioration of a society and (2) deter efficient and safe commercial activities therein. This definition incorporates those elements that may Corruption Across International Markets deter international marketing managers from deciding to enter particular foreign markets. It extends the Webster's Dictionary meaning and acknowledges corruption as a complex notion resonating with the views of political, social, and international business scholars that corruption cannot be tightly and uniformly defined. Instead, its exact measurable definition is allowed to differ across national perspectives because different cultures often hold differing perceptions of ''that which contributes to the deterioration of society''. Most important, this definition allows us to empirically identify how international marketing managers evaluate corruption and offers insight on how managers may behave with reference to their assessment (Houston and Graham, 2001) .

The attractiveness of international markets bears heavily on decisions made by international managers regarding global market segmentation, new market entry, the persistence of effort, and overall international business strategies. Of course, company resources, home country conditions, and a variety of other external factors also influence success in foreign markets. But here the focus is on characteristics of and barriers to target countries. In particular, it is expected that more corrupt countries are less attractive to managers and firms interested in expanding overseas.

Deciding which markets to enter (and remain committed to) are important initial decisions in the marketing planning process and the firm must consider a market's overall attractiveness (Kotler, 2000) by performing a comprehensive scan of the external environment in which marketing takes place. Market segmentation receives more attention in the literature because it is important to developing, positioning, and selling products across national borders (Tripodo and Dazzi, 1995) . However, a firm must decide first whether to even enter an international market before it can consider these other sorts of marketing variables.

These conditions, directly and indirectly, represent threats or opportunities to businesses (Tripodo and Dazzi, 1995) . A market is attractive to the degree that the firm perceives it can be successful there; and an indicator of market attractiveness will be the success exporters have in penetrating that market. For example, how attractive Brazil is as a market for Japanese products and companies will be indicated by the ''market share'' of Japanese imports to Brazil vis-à-vis imports from all other countries.

Toward understanding the place and meaning of corruption in international markets, a nomological net has been developed (Cronbach and Meehl, 1995) , as presented in Figure 1 . As suggested in the literature, the model includes constructs often described as antecedents and consequences of corruption and other factors that can be expected to influence market attractiveness (Diamantopoulos and Winklhofer, 2001) .

Here, we begin with a key antecedent. Scholars generally concur that corruption is clearly influenced in part by cultural values (Kaikati and Label, 1980; Sanyal, 2005; Scholtens and Dam, 2007) . Most also agree that the cultural values closely related to corruption are empirically represented by two of Hofstede's cultural dimensions: Individualism/Collectivism and Power Distance (Davis and Ruhe, 2003; Houston and Graham, 2001; Husted, 1999) . ''Individualism (IDV) pertains to societies in which the ties between individuals are loose: everyone is expected to look after himself or herself and his or her immediate family'' (Hofstede, 1991) . Its opposite is collectivism, where group membership and cooperation are paramount. Similarly, power distance (PDI) is defined as the extent to which less powerful members of institutions and organizations within a country expect and accept that power is distributed unequally. In high power distance countries people tend to accept authority and dependence. Hofstede (1980) notes the high correlation between IDV and PDI in his own data (r = 0.67) suggesting that collectivism and hierarchy usually coincide. Conceptually, these notions can be integrated into a concept that Cateora et al. (2009) refer to as relationship-oriented (ROCs) versus information-oriented cultures (IOCs). That is, in 254

Jennifer D. Chandler and John L. Graham some cultures, particularly the vertical and collectivistic ones, social relationships are salient for the interpretation of signs and symbols, including interpersonal communication. Alternatively, in other cultures, information is more salient. Thus, Cateora et al. (2009, p. 155 ) label cultures along a continuum as relationship-oriented versus information-oriented:

For example, American culture is low-context, individualistic, low power distance, and obviously [linguistically] close to English. Bribery is less common and Americans are monochromic time oriented, linguistically direct, foreground focused, and they achieve efficiency through competition…Alternatively, Japanese culture is high-context, collectivistic, high power distance, linguistically distant from English, bribery is more common, 1 polychronic (in part), linguistically indirect, and background focused. Japanese culture achieves efficiency through reduction in transactions costs… These authors classify America as an IOC and Japan as a ROC -very different culturally while recognizing that both are highly industrialized and affluent countries.

In his latest book, Hofstede (2001) reports high correlations with the Transparency International Corruption Perception Index (CPI) and GNP/per capita, and both his Power Distance and Individualism scales, as do Davis and Ruhe (2003) and Houston and Graham (2001) . Essentially, in ROCs, personal relationships, harmony, and hierarchy are primary, while rules, institutions, economics, and the marketplace are secondary. These descriptions of ROCs are consistent with the ''nepotism, cronyism, and corrupt politics'' often ascribed to Asian cultures. Fukuyama (1995) defines culture as ''inherited ethical habit,'' suggesting its influence on behavior and associated ethics. For example, patronage is seen more in the highly networked ROCs even though these countries have laws discouraging patronage and bribery in both political and commercial settings. Such statutes are less important than the maintenance of strong personal relationships in ROCs and rules are more easily ignored when they conflict with personal relationships. Therefore, we propose:

Hypothesis 1: Countries whose cultures are more relationship-oriented (ROCs) will be more corrupt.

Corruption in business systems creates additional risks for commercial operations. Argandoña (2007) argues, ''It is a problem that affects companies in particular, especially in international commerce, finance, and technology transfer.'' For example, markets wherein bribery is common add the business expense of the bribes and potential legal issues with regard to both local and home government restrictions for companies. Piracy of intellectual property directly subtracts from exporters' revenues generated in many developing countries. Black market activities cause a murkiness and legal risk associated with financial transactions in foreign markets. Finally, violence and disease both cause companies losses in terms of lost manpower, medical, security, and insurance costs. Thus, international marketers will have more difficulties penetrating markets in such countries.

To demonstrate, the FCPA has been changing American business policies and processes by essentially forcing US firms to ''walk away'' from questionable foreign business (Perry, 1992; Sheffet and Calantone, 1993) . Beginning in 1978, some foreign markets became less attractive to American firms for two reasons: (1) the competition against other firms from countries without anti-bribery laws became tougher and (2) the FCPA itself brought with it the risk of fines and jail time for involved American executives. It is therefore reasonable to hypothesize:

Hypothesis 2: Exporters will be less successful in penetrating foreign markets with higher levels of corruption.

The relative importance of corruption as a factor in international marketing success is necessarily evaluated relative to other plausible explanatory variables. Three are included in our analyses as controls: (1) consumer resources in target countries and (2) the physical and (3) cultural distances between home and target countries.

Markets with greater resources available to consumers have traditionally been seen to be more attractive to firms in exporting countries. Alternatively, Prahalad (2005) provides a strong, relatively new argument to the contrary. That is, the fastest growing new markets are at what he calls the ''bottom of the wealth pyramid,'' and it can be worthwhile for specific firms to target groups of consumers with lower incomes, etc. with specific products designed for low-income market segments. However, looking across the diverse set of product categories of the three exporting countries considered in this study, it is expected that markets with greater buying power will provide better potential for success on average. In this study, we consider two kinds of consumer resources to affect the buying power of consumers: per capita income and computer ownership. Increasingly the latter as an indicator of technological infrastructure is becoming a key element in defining the potential success within particular markets. Thus:

Control 1: Exporters will be more successful in markets with higher levels of consumer resources.

Several studies show evidence of proximity-biased trade growth, or trade growth that is lower between countries more distant from one another (Freund and Weinhold, 2004) . Furthermore, Redding and Schott (2003) argue that the world's geographically peripheral countries are becoming increasingly economically remote over time. For this paper, physical distance is measured in two ways -difference in (1) time zones and (2) kilometers between buyers and sellers. Working across large numbers of time zones is daunting. For example, the business day is just beginning at 8 a.m. in Los Angeles while it has already ended at 11 p.m. in the Philippines. The costs of disrupted family lives resulting from inconvenient conference calls and jet lag are substantial, if not quantifiable. Trade relationships between countries that are farther apart in kilometers will also be more costly to maintain because of added transportation, communication, and business travel costs. Thus, it is hypothesized that distance plays a role in determining market attractiveness.

Control 2: Exporters will be more successful in foreign markets that are less physically distant from exporting nations.

Hofstede defines culture as the collective programming of the mind that distinguishes the members of one group or category of people from another (Hofstede, 2001) . As such, cultural values differ among societies and are the underlying mechanisms that drive environmental factors such as corruption. For instance, France is known for its paternalistic conception of prerogative power toward regulation 256 Jennifer D. Chandler and John L. Graham (Hayward, 1983) compared with the Dutch corporatist tradition toward coping with passionate social interests. Or, the belief in liberalism gives the United States a more pluralistic perspective, while in Britain the philosophy of self-regulation exercises more control over advertising, financial services, and various other professional activities (Ogus, 1995) . Therefore, national culture plays an important role in determining how markets and institutions in different countries operate. All international business relationships depend on communications (mass-media and face-to-face) and negotiations (Brahm et al., 1992) between people of different cultures. The literature is quite clear that greater cultural differences or distances (sometimes referred to as psychic distances) will make transactions less efficient (Adler and Graham, 1989; Kogut and Singh, 1988; Mitra and Golder, 2002) . That is, a foreign markets' perceived cultural homophily may drive the preferences expressed by managers in the exporting countries (Shenkar, 2001) .

Control 3: Exporters will be more successful in foreign markets that are less distant culturally.

While the literature is clear about an association between cultural values and affluence, researchers and theorists continue to debate the causality involved.

That is, affluence and information-orientation coincide, but can it be determined which causes which, or is there mutual causation, or does a separate third factor cause both? Among the most important sentences ever written is Smith's (1776, p. 485 ) epiphany about one of Man's oldest conundrums, choosing between self and group interests: ''By pursuing his own interests he frequently promotes that of society more effectually than when he really intends to promote it.'' 2 That is, the invisible hand of self-interest leads to the maximal wealth of society. Individualism and competition yield benefits for the greater society. Indeed, consistent with Weber's reasoning (1930) , Hostede finds individualism is strongly correlated with GNP per capita (Hofstede, 2001, p. 251) . He explains that monopolies are more common in collectivistic countries, while in more individualistic countries competition goes hand-in-hand with greater economic freedom and the associated better economic performance. Finally, in a series of tests of all the potential relationships among cultural values, economic performance, and corruption Jing and Graham (2008) report that the most analytical credence accrues to cultural values causing both economic performance and corruption. The last go on to explain that innovation, a key driver of affluence, is more difficult in relationship-oriented cultures, because innovations tend to disrupt long lasting interpersonal bonds and are therefore resisted. Thus, to be complete we include the following relationship as an aspect of the nomological net.

Control 4: Relationship-oriented cultures will have lower levels of consumer resources.

Finally, many studies have reported economic development and corruption to be related. Usually the relationship is represented as causal, although there is disagreement about its direction. Some studies show that higher stages of economic development cause lower levels of corruption (Alam, 1995; Houston and Graham, 2001; Pavarala, 1996) . On the other hand, a serious argument can be made that corruption impedes economic development, which then may provoke even more corruption (Brademas and Heimann, 1998) . Pavarala (1996) provides a concise summary of the mechanisms explaining how stage of economic development might influence corruption levels. Generally, such studies suggest that corruption stems from a lack of capitalism and inadequate pay for government employees. However, consistent with the findings of Jing and Graham (2008) we have excluded a connection between consumer resources and corruption levels in our model specification. That is, they found that the correlation between corruption and economic development disappeared when cultural values were modeled as influencing both.

Traditional measurement theory does not allow for a simultaneous determination of the nature of corruption and its relationships with other pertinent constructs. However, the use of a PLS algorithm, including a formative indicator approach to the measurement of the constructs in the study, does. A formative indicator measurement approach does not assume that separate measures of constructs necessarily coincide (correlate). Rather, the contribution of each indicator to the definition of each construct is specifically determined simultaneously with the testing of the theoretical relationships among constructs (Diamantopoulos and Winklhofer, 2001; Fornell and Bookstein, 1982; Jarvis et al., 2003) . This approach is particularly appropriate as corruption is a concept expected to vary across cultures of both of the exporting and the importing countries involved in international marketing. Thus, the measurement model consists of six constructs, five of which are quantified by at least two measures. Descriptive statistics and correlation matrices are provided in Table I . The use of PLS makes two specific contributions. First, this approach allows for clarification of relationships among vague, esoteric constructs (such as corruption) that are present in the external environment of global exchange relationships. The measurement of these constructs would not be possible with classical measurement theory. However, to gauge each construct's reliability and validity, each construct (along with its indicators) is embedded in a structure of theoretically important relationships. Additionally, PLS is particularly appropriate when different levels of measurement (e.g., the ordinal measure of linguistic distance) and small sample sizes (i.e., n = 78) pertain (Falk and Miller, 1992) .

Second, the research design fits the model to three datasets -the focal exporting countries are France, Japan, and the US. In this way, it considers predictors of international marketing success (including corruption) and how they will differ from exporting country to exporting country. These three countries were selected because they are generally equivalent with respect to industrialization and economic development; however, they differ significantly in their social, linguistic, religious, and cultural foundations. Examining the data as focused on the export successes of these three varying countries provides a more robust test of the construct of corruption because managers based in these countries are expected to have varying perspectives on corruption.

As a result, all data are secondary, covering 78 countries in or around 2001, a year in which we should be able to observe some of the first effects of the previously described OECD anti-bribery convention particularly in France and Japan. Sources of data include The World Bank, The International Monetary Fund, and The World Health Organization, to name a few. The exact sources are detailed below.

Relationship-versus information-oriented cultures Scores for Individualism (IDV) and Power Distance (PDI) were obtained directly from Hofstede (2001) and were combined using a formative indicator approach to measure ROCs.

Measures of the prevalence of bribery, intellectual property piracy, violence, black market activities, and disease constitute the construct of corruption. The Corruption Perceptions Index created by Transparency International (TICPI), the most commonly used indicator for measuring the level of corruption across countries, reflects the impressions and perceptions of corruption in more than 100 countries based on surveys of business people, risk analysts, and the general public. The scores can vary from ''0 = highly corrupt'' to ''10 = highly clean''. As most of these surveys equate corruption with bribery, this is the measure for bribery. Piracy is measured by the Business Software Alliance (2005), which annually reports ''the difference between software applications installed (demand) and software applications legally shipped (supply)''. The measure of Violence is the number deaths by violence per 100,000 people (UCIrvine, 2006) and is derived from the World Health Organization (2004). The Black Market (specifically related to currency transactions) scores were taken from the World Bank Development Indicators Database (2000) . The percentage of adults infected with HIV was taken from the CIA Factbook (2005) . These five measures were combined using a formative indicator approach to define corruption.

International marketing success One measure of marketing success and competitiveness is market penetration or market share. Here, it is calculated for each of the three countries separately as the proportion of a country's total imports that originate from the specified exporting country. For example, Spanish merchandise imports from Japan are divided by Spain's total merchandise imports to determine Japan's share of imports by Spain. These data are gathered from the International Monetary Fund World Export/Import database.

Consumer Resources is measured by combining GNI per capita (World Bank, 2004) and number of computers per 1000 people (World Bank, 2004) using a formative indicator approach.

Physical distance is measured simply by kilometers and time zones between the capitals of the exporting countries and their target market capitals. The only exception is data from the US, in which the distance taken is that which is shorter between Los Angeles and New York. Data on time zones were obtained from the National Geographic Mapmachine (2004), while kilometers between market capitals was obtained from the National Imagery and Mapping Agency (Swartz, 2005) .

For this study, the comprehensiveness of Hofstede's (2001) value scores yields the most appropriate measure of cultural distance. Widely cited, Hofstede reports four types of cultural values that are measured across countries; and all four are used in measuring cultural distance. Power distance (PDI) is defined as the extent to which the less powerful members in a society expect and accept the unequal distribution of power. Individualism (IDV) is high in societies where the ties between individuals are loose and group membership is less important. Uncertainty Avoidance (UAI) relates to the degree that people are concerned for security in their lives and, therefore, (Hofstede, 2001) . Given these various measures, cultural distance is the sum of the differences between the exporting country and the importing country across each of Hofstede's cultural dimensions.

To augment the more traditional approach of using Hofstede's scores to calculate cultural distance, an alternative measure of linguistic distance is included. Triandis' (1977) hierarchy of subjective culture also proposed a measure of culture that stems from language, or linguistic distance. Based on his multicountry empirical study, Triandis proposed that values are derived from elemental cognitive structures, which in turn are derived from lower-level abstractions of language: words, morphemes, and phonemes. Language is also one of several proximal antecedents to various cognitive processes, which in turn are the antecedents of values in his subjective culture model. Other researchers (Hofstede, 2001; Usunier, 1998 ) support these ideas. More specifically, Nisbett (2003) argues that the observed divergence in thinking begins with language acquisition. That is, American parents focus on teaching kids clear, context-independent definitions of words. Alternatively, definitions of words in Japanese are often ambiguous and highly context-dependent. The structures of the respective languages learned thus influence the foci and processes of thinking, or culture. Thus, another measure of culture used in this study is linguistic distance as measured in West and Graham (2004) .

In this section, results from both the measurement model and the theoretical model are reported for each dataset, as seen in Table II . Estimating the measurement model reveals the contribution of individual measures in defining their associated construct, while estimating the theoretical model provides parameter estimates that indicate the strength of predictive relationships among constructs in the system.

Relationship-oriented cultures versus information-oriented cultures An examination of the latent variable weights reveals that individualism and power distance are equally important in defining the construct ROCs with all six latent variable weights varying between 0.54 and 0.57.

The latent variable weights for each of the five measures of corruption listed in Table II provide a basis for inferring the relative importance of each measure in defining corruption for each country's dataset. Based on our analyses the French definition of corruption corresponds most closely with piracy (latent variable weight = 0.69), bribery (0.24), and violence (0.20); whereas black market (0.03) and HIV (-0.15) are clearly less important. The Japanese equate corruption most with bribery (0.59) and piracy (0.41), while violence, black market, and HIV are less important. Finally, the American dataset shows piracy (0.62) and bribery (0.41) as the most important aspects of corruption, while it de-emphasizes violence, black market, and HIV.

As seen in Table II , computers (as a proxy for technological infrastructure) proved to be the more important consideration when assessing the effects of consumer resources on corruption and market attractiveness. This effect is consistent across all three exporting countries (France = 0.85, Japan = 0.65, US = 0.67). Surprisingly, this technology measure contributes more than country wealth, or GNI per capita (France = 0.16, Japan = 0.36, US = 0.35) in defining consumer resources.

For all three datasets, ''time zone difference'' dominates ''distance in kilometers'' as measures of physical distance. That is, the French, Japanese, and Americans all appear to have more trouble conducting trade across time zones than mere miles.

For both the French and the American datasets, cultural distance, as measured by Hofstede (France = 0.93, US = -1.08), plays a larger role than linguistic distance (France = -0.12, US = 0.62). However, for the Japanese, linguistic distance (Japan = 0.93) dominates cultural distance (Japan = 0.17).

Although the model is tested separately for each exporting country, there is an aspect of the model identical across the three datasets. This ''nested model'' is the set of relationships among relationship orientation, corruption, and consumer resources. As indicated in Table II , the nested model fit (variance explained) is high across the three datasets [that is, corruption R 2 for France = 0.62, Japan = 0.62, US = 0.63 (all p < 0.05)]. For the international marketing success model, the fit is statistically significant (p < 0.05) for all the three datasets, but shows the strongest explanatory power for Japan (R 2 = 0.72). The variance explained is also substantial for the US (R 2 = 0.46) and France (R 2 = 0.39). The 

Jennifer D. Chandler and John L. Graham overall fit for all the three datasets is good with model RMS Cov (E,U) all less than 0.11. Hypothesis 1 is strongly supported across the three datasets (ROC fi CORR = 0.78-0.79, all p < 0.05). That is, the more ROCs demonstrated higher levels of corruption.

Hypothesis 2 was supported for the French and Japanese datasets [CORR fi IMS (FR) = -0.28 and (JP) = -0.19, both p < 0.05], but was not supported for the American dataset. Thus, the French and Japanese are more successful in countries with lower levels of corruption. Corruption appears not to affect international marketing success for the American exporters.

Control 1 is unsupported across all three datasets. Indeed, the parameter estimates for consumer resources fi international marketing success relationship were negative or zero, and not statistically significant. In no case was international marketing success found to be influenced by consumer resources.

Control 2 (physical distance) is supported across all three datasets (p < 0.05), although it is strongest for the US [PHD fi IMS (US) = -0.71], followed by Japan (-0.48), and then France (-0.36). Exporters from all three exporting countries appear to be less successful in physically distant markets.

Control 3 (cultural distance) is supported for the French [CD fi IMS (FR) = -0.33, p < 0.05]. Greater cultural distance affects international marketing success negatively for exporters from France. Cultural distance is unrelated to market attractiveness for the American data. Unexpectedly, for the Japanese, a strong positive relationship was found between cultural distance and international marketing success [CD fi IMS (JP) = 0.58, p < 0.05].

Control 4, a relationship between relationship-oriented cultures and consumer resources was included in the analyses. For all three datasets relationshiporiented cultures were found to have lower consumer resources (ROC fi CONRES ranged between -0.73 and -0.74, p < 0.05).

Within the scope of exploring issues associated with international marketing, the salient finding of this study is the predictive and nomological validity of the corruption construct, and its role as an important determinant of international marketing success for both French and Japanese exporters. That is, French and Japanese exporters, different from their American counterparts, appear to be deterred by corruption in international markets. The construct of corruption also demonstrates nomological validity through its strong relationship to relationship-oriented cultures in the nested model. Further, by using a formative indicator approach we can discern the key aspects of corruption that are meaningful to exporters from the three countries included in the study. That is, across the datasets, bribery and intellectual property piracy have proven salient while black markets and disease were of relatively little concern. However, for French exporters relative violence levels also proved to be of importance based on the correlation coefficient and the latent variable weight listed in Tables I and II, respectively. As mentioned above, corruption is insignificant in determining market attractiveness for the Americans. This is so despite, or perhaps because of, the threedecade old US Foreign Corrupt Practices Act (FCPA). Perhaps American exporters are undeterred by the prevalence of bribery in target countries because they have had 30 years of experience in using the ''excuse'' of the FCPA as a reason not to pay bribes. Alternatively, because Japan and France have only recently joined the OECD and United Nations anti-bribery conventions, executives in those countries are at least temporarily more reluctant to trade in corrupt countries. Indeed, other studies confirm such a reluctance among US exporters immediately following passage of the FCPA (Houston and Graham, 2001) .

The reader might also reasonably predict that piracy of intellectual property might be a prime concern of American exporters given the predominance of software production, etc. in that country during the last decade. Indeed, frequent public complaints about piracy are the mantra of particularly California executives (i.e., Hollywood and Silicon Valley). However, the Americans' apparent lack of concern evident in our results may be explained by their use of alternative business strategies. For example, Bill Gates inadvertently revealed Microsoft's ''marketing strategy'' in piracy-prone countries (cf. Lam and Graham, 2007, p. 342 ):

Although about 3 million computers get sold every year in China, people don't pay for the software. Someday they will, though. And as long as they're going to steal it, we want them to steal ours. They'll get sort of addicted, and then we'll somehow figure out how to collect something in the next decade.

Apparently, it did not take a decade for Gates' strategy to work. On April 18, 2006, one day ahead of Chinese President Hu Jintao's arrival in Redmond, WA, for dinner at Gates' home on his way to a meeting with George W. Bush, Mr. Gates signed a deal with Lenovo for $1.2 billion of software for the Chinese firm's computers. Physical distance hampers marketing success and attractiveness for exporters from all three countries. Moreover, temporal distance is a much more important obstacle than mere miles. The findings clearly demonstrate that crossing time zones limits marketing success, yielding a strong preference for north-south trade, or at least trade with close neighbors. Perhaps jet travel, the Internet, and other communications advances are ameliorating physical distances, but time zone differences still appear to be a substantial impediment to trade between countries.

Consistent with predictions for the French exporters, cultural distance, particularly with regard to cultural values, limits marketing success in international markets. However, perhaps the most surprising finding of the study is the contrary one regarding cultural distance and market attractiveness for the Japanese. For both measures of cultural distance, particularly linguistic distance, the correlation coefficients in Table I and the parameter estimates in  Table II show strong negative relationships with international marketing success. A closer look at the data yields the following list of countries most distant from Japan linguistically -Indonesia, Kenya, Malaysia, the Philippines, Singapore, and Thailand. Five of the six are important South East Asian trading partners for Japan despite the differences in values and language. Part of the explanation is due to the oddity of the small Altaic language group that includes Japanese, Korean, Mongolian, and Turkish. Certainly the Japanese are well known for taking great care in adjusting marketing efforts to host country cultures (Adler and Graham, 1989; Hodgson et al., 2008) . Perhaps this is so because they have had to make the greatest adjustments. However, certainly this issue deserves more attention in future research.

International managers in the three countries were not asked directly about corruption, physical distance, or cultural distance and their effects on business success. Instead, the results of the study only provide bases for inferences about these obstacles to global marketing. Thus, parallel survey methods will be appropriate for continuing work in the area.

Although we test hypotheses nested in a rather comprehensive model of the success in international markets, our study is still exploratory particularly with regard to the development of the notion of corruption. Partial Least Squares and formative indicators allow for glimpses at and initial statements about the relative importance of the indicators forming the constructs. Our results lay important ground work for future, more deterministic measures and analyses of similar data. Similarly, the use of novel measures of violence levels and linguistic distance proved worthwhile in this study, and therefore these concepts deserve scrutiny in future work.

The use of formative indicators assumes at its base that all relevant aspects are included. However, completeness of indicator lists is a goal always sought, but never really achieved. For example, the Physical Distance construct might have included the shipping costs, or the Cultural Distance construct might include differences in religions. And more to the point, other kinds of corruption may be included in future work such as relative crime rates, corporate fraud, etc.

Finally, longitudinal studies will be important as well. Indeed, based on this study we can only speculate about causal connections among Japan's joining of the OECD anti-bribery convention, recent reforms in Japan represented by their quick ascendance in the TICPI rankings to a spot above the United States (Ishii et al., 2006) , and their future successes in foreign markets. Further, one might hypothesize that disease and violence may become more important issues as the virulence of germs, viruses, and terrorism grows. Indeed, the trade data employed in this study were collected in close 264

Jennifer D. Chandler and John L. Graham temporal proximity to both the terrorist attacks on the World Trade Center in New York and the SARS outbreak in China.

Managers from all three countries, and by implication all other countries, should recognize that views of international market attractiveness vary substantially across countries. As exporting countries have unique geographical locations and cultures, target market A may be the most attractive for country X, and likewise B for Y, etc. And, in the three exporting nations studied, market success is not statistically affected by income per capita or information systems of target countries. This supports Prahalad's (2005) findings regarding the attractiveness of the ''bottom of the pyramid.'' All these inferences hold important insights for international marketing strategy formulation and entry decisions. Policy makers should also be encouraged by our findings. While culture's influence is pervasive, it is apparently also malleable. Witness the sequential adoption of the so-called western intellectual property legal regime in Japan, South Korea, and Taiwan in recent decades (Lam and Graham, 2007) . Or, consider the successes of Singapore and Hong Kong in fighting corruption in the business systems in those relationship-oriented cultures as evinced their continuing high rankings in the TICPI. Perhaps the British colonial influence has operated there? However, now also we have the success story of Japan as well. Alternatively, perhaps the most troubling datum is the decline in America's ranking in the 2007 TICPI. Yes, culture and levels of corruption can change.

This study applies a political economic perspective to understanding international marketing strategy (Stern and Reve, 1980; Wathne and Heide, 2004) by broadening the conceptualization of corruption. In this study, two new concepts -violence levels and linguistic distance -were introduced and proved pertinent in the analyses. A comprehensive model refining the relationship between corruption and international marketing success was tested for three major world exporters, and the model was found to work differently across the three. Finally, by looking simultaneously at the results from the three exporting countries' trade relationships, this study yields a rich scan of the external environment in which international marketing takes place. These results raise important issues to be considered with international market entry decisions as well as international policy making with regard to corruption.

1 The reader will note the error here. As indicated in the Appendix, c. 2008 bribery is now more prevalent in the United States than in Japan, at least according to Transparency International. 2 The reader will note Smith's hedge: he says ''frequently,'' not ''always'' or even ''most of the time.'' Thus, he left room for regulation.

Cross-Cultural Interaction -the International Comparison Fallacy

Economic Analysis of Corruption: A Survey

A Theory of Limits on Corruption and Some Applications

Determinants of Economic Corruption: A Cross-Country Comparison

The United Nations Convention Against Corruption and Its Impact on International Companies

Tackling International Corruption: No Longer Taboo

Business Negotiations in the Peoples Republic of China

Business Software Alliance

Construct Validity in Psychological Tests

Perceptions of Country Corruption: Antecedents and Outcomes

Index Construction with Formative Indicators: An Alternative to Scale Development

Does East Meet West in Business Ethics: An Introduction to the Special Issue

The Global Fight Against Corruption: A Foucaultian, Virtues-Ethics Framing

Miller: 1992, A Primer for Soft Modeling

Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory

The Effect of the Internet on International Trade

Value Change in Global Perspective

The European Economy -Growth and Crisis

Doing Business with the New Japan (Rowman & Littlefield

Culture's Consequence: International Differences in Work-Related Values

Culture and Organizations: Software of the Mind

Culture and Organizations: Comparing Values, Behavior, and Organizational Across Nations

Culture and Corruption in International Markets: Implications for Policy Makers and Managers

Wealth, Culture, and Corruption

Transparency International: Berlin)

A Critical Review of Construct Indicators and Measurement Model Misspecification in Marketing and Consumer Research

Values vs. Regulations: How Culture Plays Its Role

American Bribery Legislation -an Obstacle to International Marketing

The Effect of National Culture on the Choice of Entry Mode

Marketing Management

China now, Doing Business in the World's Most Dynamic Market

Feeling the Heat? Anticorruption Mechanisms in Comparative Perspective

Whose Culture Matters? Near-Market Knowledge and Its Impact on Foreign Market Entry Timing

The Geography of Thought

Law and Economics in the United Kingdom -Past, Present, and Future

Determinants of Corruption: A Cross-National Analysis

Interpreting Corruption, Elite Perspectives in India

The Evolution of US Trade Intermediaries, the Changing International Environment

The Fortune at the Bottom of the Pyramid

Distance, Skill Deepening and Development: Will Peripheral Countries Ever Get Rich?

International Integration and National Corruption

Determinants of Bribery in International Business: The Cultural and Economic Factors

Cultural Values and International Differences in Business Ethics

Reviewing the Foreign Corrupt Practices Act: Did US Firms Change Their Behavior?

Cultural Distance Revisited: Towards a More Rigorous Conceptualization and Measurement of Cultural Differences

The Wealth of Nations, Book IV (Modern Library

Distribution Channels as Political Economies -a Framework for Comparative Analysis

The Analysis of Subjective Culture

Market Attractiveness of Developing Countries

Citizen Peacebuilding Project: Peace Monitor, Countries', www.cpbp.org

International and Cross-Cultural Management Research

Relationship Governance in a Supply Chain Network

The Protestant Work Ethic and Spirit of Capitalism

Eleventh New Collegiate Dictionary

Local Corruption and Global Capital Flows

A Linguistics-Based Measure of Culture and Its Relationship to Managerial Values

Trade Liberalisation and Economic Performance: An Overview

World Bank Development Indicators Online Database', www.Worldbank.org