Cultural and Economic Impacts on Global Cultural Products: Evidence from U.S. Movies

Abstract

Existing international product diffusion studies have identified economic and cultural factors that influence consumers’ acceptance of new products, but they have not fully examined these factors’ roles in the international diffusion of global cultural products. The authors examine country-level economic and cultural factors that influence consumers’ acceptance of new global cultural products across countries. Using 846 recent U.S. movies’ box office performances in 48 national markets as the empirical context, the authors obtain the following key novel findings on product sales: (1) an inverse U-shaped impact of economic development status, (2) a positive impact of the cultural compatibility of the product and the market, and (3) a U-shaped impact of intercountry cultural distance in the presence of cultural compatibility and a decreasing linear impact of cultural distance in the absence of cultural compatibility.

Keywords

global product consumer culture text mining movie

Although scholars have examined the paradigm of standardization versus adaptation in international business for decades (Schilke, Reimann, and Thomas 2009; Strizhakova, Coulter, and Price 2008), the theoretical basis of the contrasting notions still seems incomplete (Ryans, Griffith, and White 2003). In particular, we note that emerging economies are inherently different in their consumer composition and respective cultures, compared with economically advanced Western countries (Sheth 2011). Currently, more promising growth opportunities exist outside of Western countries. Therefore, it is becoming increasingly important to understand rapidly growing economies with large populations but distinctive cultures, such as China and India.

This view suggests that national consumer preferences are driven by peculiar economic conditions and cultural interactions between product-producing and product-consuming countries (Ashraf, Thongpapanl, and Auh 2014; Griffith, Cavusgil, and Xu 2008). Accordingly, multinational companies want to raise their cultural intelligence in understanding local consumers’ preferences to develop their product design, launch, and promotion strategies. In particular, such subtle cultural nuances are better examined when considering cultural products—that is, products with embedded cultural content that serves as as a strong and relevant factor for consumers’ purchase decisions. Scott (2004, p. 462) defines cultural products as “service outputs that focus on entertainment, edification, and information (e.g., motion pictures, recorded music, print media, museums) and manufactured products through which consumers construct distinctive forms of individuality, self-affirmation, and social display (e.g., fashion clothing, jewelry).”

Prior research in international marketing has examined the influences of both cultural and economic factors in explaining the international diffusion of new products, but scholars have differed in determining which factors are primary. Tellis, Stremersch, and Yin (2003) find that cultural factors, but not economic factors, can robustly explain cross-country variations in new global product sales. Conversely, Van Everdingen, Fok, and Stremersch (2009) conclude that economic factors, but not cultural factors, are significant. In contrast to both of these previous studies, Chandrasekaran and Tellis (2008) demonstrate that both economic and cultural factors influence global product sales. With respect to global cultural products, we likewise theorize that both groups of factors simultaneously drive such product sales in international markets, but the patterns of their impacts are more complicated than what the literature has presented thus far.

Given the differing views on how to strategize global products, it is imperative to understand the country-specific economic and cultural characteristics that influence the success of global products. Understanding such relationships can also help marketers avoid costly mistakes and facilitate the development of promising global products. The consumer behavior literature addressing the cultural underpinnings of globalization has highlighted individual-level decisions in understanding how people react to various cultural undertones in a product (Markus and Kitayama 1991). Although investigation at the individual level provides information about the unique micro-level variables at play, we can also gain different valuable insights from the country-level questions posed previously (Arnould and Thompson 2005). This research is aimed at answering these macro-level questions by examining whether cultural and economic conditions play a role in the international success of cultural products. Such macro-level knowledge can help marketers choose appropriate country markets for global product introduction in international markets.

Specifically, on the economic side, most studies have suggested that developed countries tend to consume global products more than developing countries because of the former's greater purchasing power (Arndt 1999; Lizardo 2008). To purchase and enjoy cultural products, consumers need to have a certain level of discretionary income. Furthermore, Griffith, Yalcinkaya, and Rubera (2014) use national economic wealth as a moderator for the relationship between product sales and the number of countries for product introduction. On the cultural side, cultural discount theory, popular in media and cultural economics, has shown that a large cultural distance (CD) between the product-producing and consuming countries decreases the sales of the global product because consumers in the product-consuming country cannot fully appreciate the product's cultural content (e.g., American English slang and pop culture in Asian countries; Hoskins and Mirus 1988; Thompson and Chmura 2015). Thus, existing studies have supported a straightforward linear impact of both cultural and economic factors.

Contrary to these common findings, we theorize that economic and cultural factors can have more complicated impacts because there are other elements that play a critical role. On the economic side, as the national economy develops, domestic competitors improve their product quality to compete against imported global products. Specifically, in regard to cultural products, when improved domestic products can appeal to local residents with the customized local culture, they can effectively compete against even the best-quality global products (Moon and Song 2015). In brief, we find that the impact involving the economic development status of the market country on product sales in the country forms an inverse U-shaped curve instead of a simple, increasing linear line.

On the cultural side, as the intercountry CD widens, local consumers’ interest in novel, exotic, and unfamiliar foreign culture embedded in imported products can increase product sales (Wang et al. 2000). Therefore, the impact of intercountry CD on product sales forms a U-shaped curve instead of a simple, decreasing linear line, but only in the presence of the cultural compatibility of the product and the market. Clearly, this finding improves the cultural discount theory that is widely accepted in international movie research (Lee 2006; Moon and Song 2015).

We propose and test such hypotheses, summarizing these seemingly contradictory theories from both economic and cultural perspectives in the international diffusion of cultural products. Furthermore, cross-cultural studies have predominantly used Hofstede's national cultural dimensions to measure general cultural differences between countries but do not consider all of the important cultural aspects pertaining to the market and the product (Dawar and Parker 1994; Eisingerich and Rubera 2010; Griffith, Yalcinkaya, and Rubera 2014). Because this approach examines the match/mismatch of the general cultural environments of the product-producing and product-consuming countries, it does not grasp how local consumers react to the specific cultural content embedded in the product. For example, Japanese consumers will not accept all Hollywood movies similarly on the cultural dimension, because the cultural content of these movies varies greatly (e.g., The Last Samurai [2003] vs. A Million Ways to Die in the West [2014]).

Therefore, we expand this currently popular approach by taking the multifaceted cultural approach of encompassing the market, the product, and the cultural compatibility of the market and the product (Moon and Song 2015). In particular, the international marketing literature has shown that the appropriateness of a specific strategy depends on its fit with the context in which it is applied. In the case of exporting global products, contingency theory suggests that product adaptation, standardization, or any combination of the two can upgrade product sales only if there is a fit between the strategy deployed and the context in which it is implemented (Hultman, Robson, and Katsikeas 2009; Katsikeas, Samiee, and Theodosiou 2006). Our cultural compatibility construct extends this strategy (global product) and context (local culture) fit theory to the cultural domain of international business. Because Hofstede's national cultural dimensions capture only the general market side, we develop a text-mining approach using consumers’ product reviews to elicit their cultural perceptions toward the product (Feldman and Sanger 2007). This approach enables us to examine all three cultural aspects.

Theoretical Conceptualization and Hypothesis Development

To better understand the different roles of globalization, glocalization, and localization in international marketing, we propose three hypotheses regarding the economic and cultural factors that affect global cultural products, using U.S. movies in international markets as our empirical context. First, we consider the economic effect on the commercial success of the global product, positing an inverse U-shaped effect of the economic development status of the product-consuming country to explain the competition between imported global products and domestic competitors. Next, we examine two cultural factors: (1) the positive impact of the cultural compatibility between the product-consuming country and the product and (2) the U-shaped effect of the CD between the product-producing and product-consuming countries that occurs only in the presence of cultural compatibility.

Economic Factors in International Diffusion

On the one hand, conventional wisdom says that developed countries consume more global products than developing countries because of the former's higher purchasing power (Lizardo 2008). On the other hand, economic development will produce stronger domestic competitors against global products as domestic competitors improve their product quality with better resources and technologies. Then, local consumers can have more alternatives, which can potentially reduce the sales of global products in a more competitive market. Therefore, we consider how the economic development status (measured by per capita gross domestic product [GDP]) of the product-consuming country is likely to influence consumer preferences as the country's development status moves from low, to medium, to high. First, when the economic development status of the product-consuming country is low, few consumers have enough discretionary income to enjoy global entertainment products. For this reason, we expect to find that most underdeveloped countries have only limited demand for global products due to their minority of wealthy consumers. The weak channel infrastructure of underdeveloped countries also undermines the distribution of imported global products (Kale 1986).

As per capita GDP increases from low to medium, consumers begin to use their discretionary income to buy global products from developed countries that are considered to be premium products in a medium-GDP country. At this stage, the domestic industries of such countries might not be developed enough to offer high-quality homemade local alternatives that can compete effectively against comparable premium global products. Accordingly, we expect the sales of global products to increase as GDP increases from low to medium. This sales growth may apply to movies whose ticket prices typically cost only a few dollars, as more local consumers seek additional leisure activities with their increased income. At the same time, improved channel distribution power may make finding such a product easier. With respect to movies, these sales growth patterns have appeared in many emerging countries (e.g., South Korea, China, India). Therefore, Cui and Liu (2001) maintain that multinational corporations need to adapt to the national market conditions in emerging countries.

However, as the GDP further increases from medium to high, we expect a decrease in global product sales because product-consuming countries can generate high-quality products that can effectively compete with global products (Dimofte, Johansson, and Ronkainen 2008; Guo 2013; Özsomer 2012). At the same time, local consumers’ positive product quality association of global products from developed countries weakens as the consuming country's economic status improves (Guo 2013). For example, the Korean movie market, for decades dominated by Hollywood movies, now has a balanced box office revenue ratio between domestic and imported movies (respectively, 52% vs. 48% in 2015) because of its improved and culturally customized domestic movies. From this supply-side perspective, as countries become more prosperous economically, their local products become comparable to global products in general quality.

In addition, local producers can even obtain a significant cultural competitive advantage by customizing their products to unique national tastes. Specifically, local producers are better qualified to create hybrid cultures, combining national and foreign cultural elements optimally to appeal to local consumers in the globalization era, whereas global producers would find it challenging to integrate customized local cultural features into their global product prototypes through glocalization because there are so many varied national cultures (Kraidy 2002). Moreover, as the developed national economy advances further, many industries become more concentrated; in Hollywood, a small number of domestic distribution companies dominate the domestic movie distribution system, which thus favors domestic movies because of the close relationships between the domestic production and distribution companies (Bakker 2005).

Furthermore, as economies become more developed, the status of many global products as aspiration goods weakens, because the consumers themselves belong to a prosperous country. Thus, national pride might kick in and work against the lure of global cultures. With an increase in economic prosperity, consumers would have greater purchasing power to indulge in local products that highlight their unique culture instead of buying global products common around the world. Owning premium local products might enable consumers not only to differentiate themselves from those who own ubiquitous comparable global products but also to convey national pride. Therefore, we propose the following:

H₁:

The impact of the economic development status of the product-consuming country on product sales is inversely U-shaped.

Cultural Factors in International Diffusion

Cultural meaning in a consumer society moves from the culturally constituted world, to consumer goods, to the individual consumer (McCracken 1986). There are fundamentally different perspectives on how consumers of different cultures receive global products. On the one hand, cultural imperialism suggests that consumers passively consume the dominant culture of their time, which would lead to the emergence of a global consumer with unified preferences (Levitt 1983; Zou and Cavusgil 2002). On the other hand, some cultural studies have made more nuanced predictions as to how various local cultures accept global products from economically advanced countries that are cultural leaders (Hoskins and Mirus 1988). This view of combining cultural nuances with changing economic prosperity across countries predicts that the increasing economic prosperity of countries with diverse cultures leads to a cultural mosaic, with cultural similarities versus differences playing nuanced roles. Similarly, research in international marketing has delved into the differences of standardization and adaptation strategies to understand how the strategic fit between the market environment and the imported product influences the commercial success of the product in international markets (Katsikeas, Samiee, and Theodosiou 2006).

Cross-cultural studies have focused on the CD of the general environments involving the product-producing and product-consuming countries (Hofstede, Hofstede, and Minkov 2010). This intercountry CD construct has been used in the international business literature to shed light on how delicate cultural nuances influence international business activities (Craig, Greene, and Douglas 2005; Tihanyi, Griffith, and Russell 2005). Because this CD construct examines the match/mismatch of the general cultural environments of the product-producing and product-consuming countries, it does not grasp how local consumers react to the specific cultural content embedded in the global product. For example, two U.S. animated movies, Kung Fu Panda (2008) and Dr. Seuss’ Horton Hears a Who! (2008), performed successfully in North America ($215 million and $154 million) but had drastically disparate box office numbers in Japan ($17 million and $.3 million) and Korea ($29 million vs. $3 million) (BoxOfficeMojo.com). The former movie highlighted a popular martial art in East Asia, kung fu, whereas the latter used the name of a famous American author in its title, who was hardly known outside of English-speaking countries.

Cultural compatibility (or cultural congruence or cultural familiarity) is known to increase category knowledge and learning (Newman and Nollen 1996). Because local consumers find familiar cultural elements (e.g., local celebrities, local places, local music) in a global cultural product, such elements will raise their interest and can even create significant online word of mouth about the product, which will eventually increase product sales in the local market. Similarly, the adaptation/standardization research has suggested that if consumers have similar beliefs and attitudes or similar product usage patterns, this will lead to greater strategy standardization (Katsikeas, Samiee, and Theodosiou 2006). The cultural compatibility construct extends the CD construct to capture the multifaceted dimensions of national cultures more completely. In summary, when a global product is introduced into a foreign market, we can observe three entities: (1) the product-producing country, (2) the product-consuming market, and (3) the product itself. The CD construct utilized in prior research is helpful in examining the interaction between (1) and (2). Importantly, our compatibility construct captures the interaction between (2) and (3). Thus, we advance the existing findings in this domain by considering the following relationship:

H₂:

The cultural compatibility of the product-consuming country and the global cultural product increases the sales of the global product in the product-consuming country.

To examine the impact of CD moderated by cultural compatibility, we consider two cases—the presence and absence of cultural compatibility. First, in the presence of cultural compatibility, as CD increases from small, to medium, to large, we predict a U-shaped pattern of product sales (Moon and Song 2015). With the presence of cultural compatibility, when CD is small, we draw on cultural categorization theory to hypothesize that consumers can understand the culture of the global product as a result of cultural familiarity (Applbaum and Jordt 1996; McCracken 1986). This would lead to greater acceptance of the culturally understandable product. When CD increases from small to medium, with cultural compatibility present, we are likely to observe the cultural discount phenomenon (Hoskins and Mirus 1988). Consumers of the product-consuming country would have a limited understanding of the global cultural product because of the increased CD (Carvalho et al. 2008). Furthermore, although the culture of the product may still be generally understandable to consumers, the unfamiliar culture of the product-producing country could appear threatening. For instance, Slumdog Millionaire, a U.S. movie about India, was not well received in India because Indians questioned the validity of the movie. (In this example, there is a medium level of CD between the United States and India, but there is cultural compatibility between the movie and India because the movie is situated in India.) Such a threat factor is accentuated in a situation in which the cultural implications are neither so similar as to be liked nor so dissimilar as to seem novel.

Finally, when CD becomes large enough, the culture implied by the product does not appear to be a threat any longer because it is quite alien or exotic to the consuming country. When a culturally distant country generates a global product whose cultural content is novel enough to arouse curiosity from consumers, they tend to see the global product as culturally novel, which motivates them to accept it more positively. Consumers tend to buy such culturally novel products from culturally distant countries when they intend to enjoy a variety of cultures (Wang et al. 2000). This behavior is supported by previous research showing that consumers often seek novelty in their enjoyment when they consume cultural products such as travel, food, fashion, and movies (Jeong and Park 1997; Tse and Crotts 2005).

Conversely, when there is no cultural compatibility, we do not expect to see the U-shaped impact of CD on product sales, which can be observed in the presence of cultural compatibility. As most cultural research has shown, using the CD measure to capture the general cultural match between the product-producing and product-consuming countries, we predict a downward slope as CD increases. In this case, the cultural discount effect becomes dominant without being hindered by the cultural novelty–seeking or cultural threat effects that occur in the presence of cultural compatibility. In summary, we propose the following:

H₃:

The impact of CD on product sales is moderated by the cultural compatibility between the culture of the product-consuming country and the culture embedded in the global product.

H_3a:

In the presence of cultural compatibility, the impact of CD on product sales is U-shaped.

H_3b:

In the absence of cultural compatibility, the impact of CD on product sales decreases linearly.

Empirical Application

This section presents our research methods and data analyses to test our hypotheses. To do so, we chose the movie industry as a representative global cultural product for the following reasons. First, movies generate enormous sales and diffuse well in international markets. Second, a movie's cultural content plays a major role in its commercial success in international markets (Lee 2006). Accordingly, cultural roles are easy to examine objectively because movies have apparent cultural meanings that consumers in product-consuming countries can perceive in terms of a cultural match versus mismatch.

Third, the industry provides a wealth of public data through online movie forums (e.g., IMDb.com, BoxOfficeMojo.com). These forums publicly release information on both quantitative movie characteristics and qualitative text reviews. Importantly, the recent popularity of movie forums as data sources and text analytics as a tool to analyze such qualitative information enable us to analyze the cultural content of movie reviews in a systematic and rigorous fashion. Recently, researchers have applied this approach to conduct in-depth analyses of various products ranging from digital cameras to hotels (Lee and Bradlow 2011). Whereas these studies use text analytics to infer product attributes and consumers’ attribute evaluations, our application of text analytics is innovative by using consumers’ movie reviews to infer the cultural content of each movie.

Data and Variables

To examine the roles of U.S. movies’ cultural content, we collected both quantitative (movie characteristics and foreign country characteristics) and qualitative (consumers’ text reviews on movies) information on U.S. movies (see Table 1), using web-scraping techniques. We chose 846 U.S. movies released between 2008 and 2015. This process removed some movies (mostly movies with low box office performance) that lacked the required data. We obtained the movies’ key quantitative characteristic information, such as box office revenues and consumer ratings (Moon, Bergey, and Iacobucci 2010). The countries were distributed in Europe, Asia, and other regions. To examine how consumers evaluate movies, we collected a large amount of consumers’ reviews available on the Internet Movie Database (IMDb) with web scraping. The total number of reviews collected for analysis was 128,444. Our information on the reviews included accompanying quantitative variables, such as consumers’ numeric movie ratings. These reviews were written by voluntary contributors from around the globe and have been used by other studies (Boskem and Smidts 2015; Hennig-Thurau, Houston, and Heitjans 2009). Although they are written in English, anyone can access them through the Internet. It is assumed that some reviewers can find good or bad movies and act as opinion leaders to spread the word to the entire movie-watching community.

Table 1

Variable Description

Variable	Measure	M	SD	Source
Movie-Specific Variables
Running time	Minutes	110	19	IMDb.com
Budget	U.S. dollars	65.1 million	60.0 million	BoxOfficeMojo.com
Star power	Average box office revenues of all movies in which the actors/actresses were cast members in the five years before the release of the movie; in USD; inflation adjusted	4.6 billion	4.2 billion	BoxOfficeMojo.com
Consumers’ rating	Ten-point scale	6.55	.94	IMDb
Genre	A movie can have multiple genre categories	Action/Adventure/Sci-Fi: 48%; Animation/Family: 14%; Comedy: 36%; Drama: 40%; Romance: 17%; Thriller/ Mystery/Crime/Horror: 36%		IMDb
Sequel	Sequel = 1; No sequel = 0	Sequel: 19%		IMDb
Motion Picture Association of America rating	R = 1; PG-13, PG, and G = 0	R: 38%; PG-13: 43%; PG: 17%; G: 2%		IMDb
Country-Specific Variables
English use proportion	The proportion of the country's population who can speak English	36%	34%	Wikipedia, NationMaster
Population	Number of people (in 2014)	51.8 million	109.5 million	World Bank
Trade openness^a	Sum of exports and imports of goods and services measured as a share of GDP	91%	78%	World Bank
Internet penetration^a	Percentage of individuals who have used the Internet in the last 12 months	59%	22%	World Bank
Gini coefficient^b	Extent to which income is distributed	38%	10%	World Bank
Per capita GDP^a	US$	28,917	21,222	World Bank
CD	Based on the Hofstede's four cultural dimensions of power distance, individualism, masculinity, and uncertainty avoidance	2.30	1.31	geert-hofstede.com
Geographical region	A total of 47 countries (excluding the United States)	Europe: 25; Asia: 10; other countries: 12 (base category)		Google
Country group	K-means clustering based on the Hofstede's four cultural dimensions of power distance, individualism, masculinity, and uncertainty avoidance	United States: 1; Germany-Australia: 13; Italy-Japan: 4; Spain-Brazil: 24; India-Indonesia: 6 (base category)		geert-hofstede.com
Movie and Country Combination Variables
Box office revenues	Sales for the movie in the country in U.S. dollars; dependent variable	2.7 million	7.0 million	BoxOfficeMojo.com
Number of U.S. movies in country	Of the 846 U.S. movies in our data, the number released in the country in 2008–2015	688	150	BoxOfficeMojo.com
Release day difference	The number of days between the U.S. release and the foreign country's release	49	103	BoxOfficeMojo.com
Number of screens in country	The number of screens used for the movie in each country	553	965	BoxOfficeMojo.com
Cultural category: United States	Text categorization based on movie reviews	.008	.022	IMDb
Germany-Australia	on IMDb	.010	.021
Italy-Japan		.012	.028
Spain-Brazil		.018	.030
India-Indonesia		.008	.024
Cultural compatibility	The degree of match of the cultural content of the movie (cultural category) and the culture of the country market (country group)	.54	1.89	geert-hofstede.com, IMDb
Release season	Four seasons	January–March: 29%; April–June: 23%; July–September: 24% (base category); October–December: 25%		BoxOfficeMojo.com

Measure is from 2014.

Measure is from 2012.

Moreover, for each of the foreign countries, we used economic measures such as per capita GDP and population. Unlike many developed Western nations, emerging markets are still growing, and, subsequently, their economic status is constantly changing. To measure such changing economic conditions, we used the per capita GDP of the product-consuming country as our main economic measure. Next, we explain the two key cultural variables: CD and cultural compatibility.

CD between the Product-Producing and Product-Consuming Countries

We measured the CD between the product-producing country (i.e., the United States) and the product-consuming country using Hofstede's CD measurement. This measure indicates the level of similarity in the general cultures of the two countries. A larger distance indicates less cultural similarity, while a smaller distance indicates more cultural similarity. Although this measure has been criticized, it remains the most common framework to measure the compatibility of general cultural environments with respect to multiple national markets (Soares, Farhangmehr, and Shoham 2007). Therefore, this measure has been commonly used by media economists and cultural economists to address the influences of intercountry cultural differences on U.S. movies’ international box office performance (Fu and Lee 2008).

To calculate each product-consuming country's CD with the United States as the product-producing country, this measure utilizes the four cultural dimensions of individualism, masculinity, power distance, and uncertainty avoidance (Hofstede, Hofstede, and Minkov 2010), using the following formula (Kogut and Singh 1988):

{CD}_{j} = \sum_{i = 1}^{4} [{(I_{ij} - I_{iA})}^{2} / V_{i}] / 4,

(1)

where I_ij stands for the index for the ith cultural dimension and the jth product-consuming country, V_i is the variance of the index of the ith dimension, A indicates the United States, and CD_j is the calculated cultural distance of the United States and the jth country.

The Product's Cultural Content and the Cultural Compatibility between the Product and the Country Market

Traditionally, cross-cultural studies have predominantly used Hofstede's CD measure as a general cultural indicator of the fit of two involved countries’ business environments (Eisingerich and Rubera 2010). However, this research practice reveals a significant void in the international diffusion of global products from a cross-cultural perspective, in that this measure cannot capture the specific cultural content embedded in individual cultural products. In other words, although the product should be evaluated and received by the market, the popular CD measure misses the product's cultural content entirely by examining only the general cultural environment of the market. To resolve this issue, we develop a procedure of eliciting the cultural content embedded in each product on the basis of what consumers say about it, which eventually enables us to measure the cultural compatibility between the product-consuming country and the product (e.g., how the Japanese enjoy specific U.S. movies).

One way to measure cultural compatibility is to develop a procedure for eliciting the cultural content of the product from consumers’ online product reviews using text-mining techniques (Feldman and Sanger 2007). Although text reviews contain a wealth of information, they have been largely ignored because of their unstructured nature. However, recent studies have demonstrated that, with text-mining analysis, such text reviews can contribute to explaining and predicting new product sales (Archak, Ghose, and Ipeirotis 2011; Ghose, Ipeirotis, and Li 2012), beyond consumers’ numeric product ratings (Chevalier and Mayzlin 2006), which can tell us only how much consumers like products. In contrast, text reviews can provide more insights into why certain products succeed because such reviews have the potential to reveal consumers’ intimate experiences and detailed thoughts. Furthermore, obtaining information directly from a variety of consumers provides the true voice of the market to the researcher (Decker and Trusov 2010; Moon, Park, and Kim 2014). Next, we offer a detailed explanation of this four-step procedure of creating the cultural compatibility variable.

Step 1: Country Segmentation in Terms of the National Culture

To reduce the cultural variability of international markets composed of several national markets (Hsieh 2002), we grouped the 48 countries in our data (including the United States) into five cultural country groups (see Table 2). First, we classified the United States as a separate country group (the product-producing country) in our application. Furthermore, the United States is the dominant country in the movie industry and the most influential leader in the world economy and culture. Next, we divided the 47 foreign countries into four culturally distinct groups on the basis of Hofstede's four national cultural dimensions, using K-means clustering analysis. To determine the optimal number of clusters, we used the elbow method, which generated four additional cultural country groups.

Table 2

Country Segmentation in Terms of National Culture

Group Label	N	PDI	IDV	MAS	UAI	CD with United States	Cultural Properties
United States	1	40	91	62	46	.00	High IDV
Germany-Australia	13	30	71	45	52	.45	High IDV, low PDI
Italy-Japan	4	64	64	91	75	1.68	High MAS, high UAI
Spain-Brazil	24	69	35	49	83	2.55	Low IDV, high UAI
India-Indonesia	6	83	28	54	34	2.56	Low IDV, high PDI

Notes: N = number of member countries; PDI = power distance index; IDV = individualism (vs. collectivism); MAS = masculinity (vs. femininity); UAI = uncertainty avoidance index.

Step 2: Defining Each Country Group's Culture by a Google Search

To elicit the cultural content of each of the 846 movies in terms of each of the five cultural country groups, we text mined and analyzed those movies’ qualitative cultural content using consumers’ movie reviews collected on IMDb (Moon and Song 2015). To define the culture of each country group, we searched for each country's cultural terms using a Google search (see the Appendix). For example, to collect cultural terms for the United States, we used “US culture” as a search keyword on Google. Then, three research assistants and an author went through the pages provided by Google to select relevant terms, using their own judgments. Next, the authors combined all of the cultural terms for the particular country and selected the most relevant 150 terms for the country. Subsequently, all of the terms for each member country were combined into the term collection defining the country group's culture. This task of collecting the relevant cultural terms to define each country group resulted in 7,050 terms (150 terms × 47 countries).

Step 3: Text Categorization to Elicit the Cultural Content of Individual Movies

We analyzed movie consumers’ text reviews to determine the degree of the cultural content of each movie in the five cultural definitions—(1) United States, (2) India-Indonesia, (3) Germany-Australia, (4) Italy-Japan, and (5) Spain-Brazil. To that end, we applied text categorization (TC) to quantify and categorize product reviews, focusing on the cultural content in the reviews. Text categorization classifies documents (i.e., movie reviews in our application) into a fixed number of predefined categories (i.e., the five cultural definitions) given the terms used in movie reviews (Joachims 1998). How relevant the term usage is for each document, given the term definitions of each predefined category (as we show in the Appendix), determines the posterior probability (PP) that each document belongs to one of the categories. Conceptually, if a certain movie review frequently mentions some of the terms used to define the Italy-Japan cultural terms in the Appendix, the movie review's PP will turn out to be high in that cultural category. In addition, some movies can be highly cultural in multiple categories, whereas others can be culturally irrelevant to all categories. For example, The Last Samurai (2003), in which the American main character (played by Tom Cruise) traveled to Japan, is high in its cultural content in both the U.S. and Italy-Japan categories. In contrast, Alien vs. Predator (2004) turns out to be culturally irrelevant, given that its background was purposely created in an imaginary setting as a science-fiction movie. Feldman and Sanger (2007) provide the technical details of our TC task.

Step 4: Developing the Cultural Compatibility Variable

Drawing on the estimated PPs from Step 3, we developed the cultural compatibility variable that measures the degree of the cultural match of the country market and the product (see Table 3). To generate this interaction term of the country and the movie, we multiplied the continuous cultural group variables (i.e., the United States [culture], Germany-Australia [culture], Italy-Japan [culture], Spain-Brazil [culture], and India-Indonesia [culture]) and the binary geographical country group variables (i.e., Germany-Australia, Italy-Japan, Spain-Brazil, and India-Indonesia). Then, for the cultural compatibility variable of the market and the product, we used the continuous cultural categorical variable that matches the country market. For example, if the movie was released in Spain, we used the continuous value of the Spain-Brazil group (culture) variable. Because each of the five cultural country group variables has a different mean and standard deviation, we used the standardized values for this cultural compatibility variable to equalize the impacts among the five cultural country groups.

Table 3

Parameter Estimation of the Country-Level Movie Box Office Model

Variable	Full Model	Compatibility Model	No Compatibility Model
Intercept	-15.7326^***	-15.4889^***	-16.1492^***
Movie-Specific Variables
Consumers’ rating	.2726^***	.2700^***	.2548^***
Genre: action	−.07751	−.08219	−.04190
Genre: animation	.04060	−.06817	.3611^*
Genre: comedy	−.09907	.02678	−.2631^**
Genre: drama	−.3160^***	−.3174^***	−.3042^***
Genre: romance	−.07229	−.09713	.2512^**
Genre: thriller	.07525	.1185	−.04382
Sequel	.5836^***	.6612^***	.4990^***
Running time (minutes)	−.00240	−.00270	−.00191
Motion Picture Association of America rating (R = 1)	−.08686	−.1286	−.00987
Budget (natural log)	.7074^***	.7321^***	.7049^***
Star power (natural log)	−.01075	−.03094	.000355
Country-Specific Variables
English use proportion	−.00602^***	−.00629^***	−.00557^***
Population (natural log)	.6375^***	.6206^***	.6617^***
Trade openness	−.00410^***	−.00385^***	−.00436^***
Internet penetration	.01107^***	.01073^***	.01202^***
Gini coefficient	.04330^***	.04408^***	.04231^***
Per capita GDP (linear)	.000079^***	.000079^***	.000078^***
Per capita GDP (quadratic)	-535E-12^***	-541E-12^***	-519E-12^***
CD (linear)	−.1523^***	−.1693^***	−.0576^***
CD (quadratic)	.0520^***	.0650^***	.0036
Geographical region: Europe	.5425^***	.5585^***	.5200^***
Geographical region: Asia	.4697^***	.3471^***	.5818^***
Country group: Germany-Australia	−.1540^***	−.1722^**	−.2452^***
Country group: Italy-Japan	−.09310^*	−.1028	−.1720^**
Country group: Spain-Brazil	−.1025^**	−.1210^*	−.1792^**
Movie and Country Combination Variables
Number of screens in country (100)	−.0178	−.0182	−.0196
Release day difference	−.00523^***	−.00466^***	−.00601^***
Release: January–March	−.00437	.06270	−.08206^*
Release: April–June	−.1306^***	−.1155^***	−.1525^***
Release: October–December	−.02161	−.00062	−.05989
Number of U.S. movies in country	.001807^***	.001818^***	.001823^***
Cultural category: United States	-3.8134^**	-1.3259	-4.5010^***
Cultural category: Germany-Australia	-4.3499^***	-2.5136	-3.8724^**
Cultural category: Italy-Japan	-2.8818^*	-4.6884^*	-2.3051
Cultural category: Spain-Brazil	3.4399^**	6.0162^**	3.1048^*
Cultural category: India-Indonesia	-3.7777^**	-2.9835	-3.7893^**
Cultural compatibility	.01863^***	N.A.	N.A.
Compatibility × CD × GDP	9.55E-10^***	N.A.	N.A.
σ_α²	.7149	.7336	.6289
σ_ε²	.7454	.7137	.7222
Akaike information criterion	49852.1	28135.7	21829.9
Sample size	25905	13917	11988

Significant at 10%.

Significant at 5%.

***

Significant at 1%.

Notes: N.A. = not applicable. Boldfaced cells refer to the key variables used in hypothesis testing.

Country-Level Movie Box Office Model

To test the impacts of the cultural and economic variables on foreign box office performance, we developed a country-level movie box office model. This movie and country-level analysis used a combination of 846 movies and 47 countries for its observations. Because no single movie was released in all 47 countries, the actual observation size of this analysis was 25,905, indicating that a movie was released in 31 countries, on average.

First, for this analysis, we integrated the inferred text review categorization results into a log-linear regression model that can explain which factors influence new movies’ commercial success. To begin with, our base model is expressed as follows:

Y_{pn} = α + β_{p} x_{p} + β_{n} x_{n} + β_{pn} x_{pn} + ε_{pn},

(2)

where y_pn indicates the natural log of the sales dollar amount for product (movie) p in national market n (e.g., Italy). We took the natural log to better fit the model by normalizing the distribution of the sales dollar amount. α denotes the intercept; β is a vector of the parameters to be estimated; x_p is a vector of multiple product-specific characteristics (e.g., movie budget, movie genre) as a set of interdependent variables; x_n is a vector of multiple nation-specific characteristics (e.g., population, Internet penetration) as a set of interdependent variables; x_pn is a vector of multiple product and nation combination characteristics (e.g., the number of U.S. movies in a country, the number of screens in a country) as a set of independent variables; and the distribution term ε_pn follows the normal distribution.

A primary question would be how to integrate the review categorization information, which resulted in the aforementioned five cultural variables of the United States, Germany-Australia, Italy-Japan, Spain-Brazil, and India-Indonesia into the base model in Equation (2). Because the posterior membership probabilities from our review categorization can be computed for each review, we convert the posterior membership probabilities at the review level into movie-level membership probabilities by taking the average of the probabilities across all of the reviews on product p. That is,

{PP}_{p} = Σ_{r} {pp}^{(p)} / R^{(P),}

(3)

where PP indicates the continuous posterior probability of belonging to a specific category for product p. The specific category is one of the five cultural country groups. Subscript p denotes the product, and subscript r denotes the review. Therefore, Σ_rPP_r^(p) covers all of the PPs of all reviews (r) on product p from the text-clustering procedure, and R^(p) is the total number of reviews on product p. In brief, PP_p is the simple average of the posterior probabilities of all reviews on product p. Conceptually, the average cultural content revealed by the reviews on the movie represents the average perception by consumers.

When we combine Equation (3) with the base model in Equation (2), the model is extended as follows:

Y_{pn} = {α + β}_{p} x_{p} + β_{n} x_{n} + β_{pn} x_{pn} + β_{p1} {pp}_{p1} + β_{p2} {pp}_{p2} + … + β_{pC} {pp}_{pC} + ε_{pn},

(4)

where PP_pc denotes the product-level PP of product p for each cultural variable c. When there are a total of C classifications, we have C of PP variables for product p.

We note that the model has unobserved heterogeneity across nations and movies. Because the country membership variables can substantially account for country-related heterogeneity, we account for unobserved movie-related heterogeneity by extending the model in Equation (4) into the following random-effects model (Greene 2003):

Y_{pn} = ({α + α}_{p} {) + β}_{p} x_{p} + β_{n} x_{n} + β_{pn} x_{pn} + β_{p1} {pp}_{p1} + β_{p2} {pp}_{p2} + … + β_{pC} {pp}_{pC} + ε_{pn},

(5)

E [α_{p} | X] = σ_{α}^{2}, and

(6)

E [ε_{pn} | X] = σ_{ε}^{2},

(7)

where α_p indicates the product-level intercept for product p. As Equation (6) shows, these intercepts are allowed to vary across products. However, to account for observed variations across countries and products, we include several country- and product-specific variables. Whereas α_p accounts for unobserved variations across products, ε_pn accounts for unobserved variations across products and countries.

Finally, in our application, one movie characteristic variable is the number of screens, which can be determined by the movie distributor's anticipation of box office revenues. To reduce this endogeneity problem, we use the distribution by eight major studios as an instrumental variable for the number of screens variable. We select this particular instrumental variable because although these major studios can exert a larger influence on number of screens than nonmajor studios, they still do not have a direct influence on box office revenues. Because of this instrumental variable approach, we estimate the proposed model using the two-stage least squares method.

Empirical Results of the Country-Level Movie Box Office Model

Table 3 shows the complete list of the independent variables and parameter estimates for each of three regression models: (1) the Full Model, (2) the Compatibility Model, and (3) the No Compatibility Model. The complete sample for the Full Model was divided into two subsamples: (1) a subsample with only compatibility cases and (2) a subsample with no compatibility cases, with which to test H₃. Notably, the aforementioned instrumental variable approach rendered the otherwise significant number of screens variable insignificant, which supports our suspicion that the endogeneity effect pertains to this variable.

We focus on our testing results for the proposed hypotheses, using the empirical results in Table 3. First, as the major economic factor, we examine whether the relationship between per capita GDP and box office revenues follows an inverted U-shaped pattern. The combination of the significantly positive linear term and the significantly negative term of per capita GDP in Table 3 shows an inverted U-shaped pattern, as indicated in Figure 1 (H₁ supported).

Figure 1

The Curvilinear Effects of Per Capita GDP on Box Office

Next, the cultural compatibility factor between the movie-consuming country and the movie involves four country-group variables (Germany-Australia, Italy-Japan, Spain-Brazil, and India-Indonesia) and five cultural category variables (United States, Germany-Australia, Italy-Japan, Spain-Brazil, and India-Indonesia). The four country group variables indicate each country's membership, determined by the country segmentation task in Table 2. For example, because Brazil belongs to the Spain-Brazil country group, the variable is 1 for Brazil, whereas the other three country variables are 0 for Brazil. Because these four variables are dummy coded, as in the case of Brazil, we designate the India-Indonesia country-group variable as the reference category while keeping the other three variables in the regression. The empirical results indicate the following order in box office revenues while holding all of the other factors constant: India-Indonesia > Italy-Japan > Spain-Brazil > Germany-Australia. To further reduce the unobserved variability across countries, the model includes the geographical region variables that divided the 47 countries into three regions: Europe, Asia, and other countries (the reference category).

The five cultural variables measure the degree of cultural content related to the specific cultural category as a continuous variable. A variable is not restricted by the others in the five cultural variables because the same movie can be high (or low) in multiple cultural categories. We observe that only the Spain-Brazil (culture) variable is positively significant, which means that a movie rich in that cultural category tends to do well in terms of box office revenues, after accounting for the other factors included in the regression. Notably, Table 2 indicates that this country group has the second-largest CD with the United States among the four country groups. In other words, in certain situations, consumers prefer to experience novel and exotic cultural elements, contrary to the cultural discount phenomenon. Next, the cultural compatibility variable involving these cultural category factors measures the degree of the cultural match between the movie and the country group. This variable shows a significantly positive impact on box office revenues. In other words, movies that contain certain cultural categories can increase their sales in countries that can understand such cultures (H₂ supported).

The factor of the product-consuming country's CD with the United States has both linear and quadratic terms, which can test whether the relationship between CD and box office revenues is linear or U-shaped. To test H₃, we divided the entire sample into cultural compatibility and no cultural compatibility cases, as shown in Table 3. This division generated Figure 2, in which we can see the relationship between CD and box office revenues in the two cases, respectively. First, in the presence of cultural compatibility, the combination of the significantly negative linear term and the significantly positive term results in a U-shaped relationship between CD and box office revenues (H_3a supported). By contrast, in the absence of cultural compatibility, the relationship between CD and box office revenues is reduced to a negatively linear line (H_3b supported), as cross-cultural studies have traditionally explained. Finally, the three-way interaction term of cultural compatibility × CD × per capita GDP measures the most complex effect arising from the interactions of the two cultural measures (cultural compatibility and CD) and the economic measure (GDP). The positively significant estimation result indicates that box office performance can be enhanced in the presence of cultural compatibility, large CD, and high per capita GDP.

Figure 2

The Curvilinear Effects of CD on Box Office

Discussion

The main purpose of this research is to examine the influence of economic and cultural factors on global product acceptance in international markets. We measured acceptance in terms of the revenue generated by the global cultural product (U.S. movies in our empirical application; Griffith, Yalcinkaya, and Rubera 2014). First, on the economic side, we found that per capita GDP has an inverse U-shaped impact on global product sales (H₁ and Figure 1), which runs counter to conventional wisdom that would suggest a simple, positively linear impact (Lizardo 2008). Consumers in a low–per capita GDP country may not have the money to buy the global product. In a medium-GDP country, when consumers have some money, they may show a preference for the product, which increases revenues. Surprisingly, however, at a high GDP, we observe a drop in the preference for global products, because consumers have more competitive domestic products to choose from. This finding helps us answer our questions of whether a global consumer will emerge in different countries with common tastes and whether consumer preferences change as a result of economic prosperity. In our data, we do observe a change, which indicates that preferences at low, medium, and high GDP differ from one another.

Next, when we considered the CD between the product-producing and product-consuming countries in the presence of the cultural compatibility of the product and the market, we observed a U-shaped curve (H_3a and Figure 2) owing to the surprising positive effect of cultural novelty seeking in the case of high CD. This finding contrasts with the previously reported linearly decreasing relationship based on cultural discount theory (Hoskins and Mirus 1988; Lee 2006), which happens only in the absence of cultural compatibility. In the case of the cultural compatibility of the product and the market, when the cultural proximity between the product-producing and product-consuming countries decreases, we find that product sales decrease in line with the cultural discount. However, as the intercountry CD becomes very large, product sales begin to increase as consumers find distant culture embedded in the product to be novel and interesting.

Theoretical Implications

For the adaptation versus standardization debate in the context of international business, Dawar and Parker (1994) draw the conclusion that certain behaviors are likely to be universal, whereas others are not. Understanding such differences is essential to designing international marketing strategies. In this context of international business, the current research includes several theoretical advancements. First, this research helps us understand different economic and cultural variables that influence product acceptance from a macro (country-level) rather than a micro (consumer-level) perspective. Furthermore, in this research we examine important variables at the macro level and find that it is worthwhile to consider cultural and economic factors while introducing a product to different countries. Second, this research uses consumers’ product reviews to empirically measure how consumers react to the culture embedded in the product. Third, despite all of the increasing trade and communication among countries, this research suggests that consumers will not converge toward the “global consumer” concept (i.e., a situation in which consumers across the world begin to reflect the same tastes). The influence of local cultures will always persist, which supports the cultural mosaic account and the effectiveness of glocalization (Kjeldgaard and Askegaard 2006).

Fourth, this research theorizes and empirically supports the notion that the impact of the economic development status of the product-consuming country on product sales is inversely U-shaped (H₁; Elliott and Cameron 1994; Kraidy 2002) versus the prevailing notion that the relationship is increasingly linear. Finally, this research also proposes and empirically proves that the impact of intercountry CD on product sales is U-shaped only in the presence of cultural compatibility of the product and the market (H₃; Jeong and Park 1997; Tse and Crotts 2005; Wang, Chen, Chan, and Zheng 2000). This finding contradicts cultural discount theory, which would indicate that the relationship is decreasingly linear.

This research is within the stream of examining cultural and economic factors in new movies’ international diffusion. In particular, two notable studies share this common research theme: Moon and Song (2015; MS) and Griffith, Yalcinkaya and Rubera (2014; GYR). We acknowledge that there are similarities between MS and the current research because MS's study was the main motivator for this article. However, there are significant differences between the two studies. First, MS's study focuses on the cultural aspect, whereas we focus on both the cultural and economic aspects. To the best of our knowledge, the current research is the first to propose and empirically support the curvilinear impact of the economic development status of the product-consumer country on product sales (H₁ and Figure 1).

Second, both studies use similar text-mining techniques—specifically, TC techniques—to generate cultural categories for the measurement of cultural compatibility of the product and the country. In doing so, however, our cultural compatibility measurement procedure has a significant improvement. Whereas MS use two subjectively predetermined cultural categories (i.e., American and un-American), we use five data-driven cultural country groups, based on the definition of each country group's culture by a Google search (Tables 2 and 3). In brief, our procedure is a data-driven, more sophisticated way to measure the cultural content of the product and cultural compatibility. Finally, both studies consider the U-shaped influence of CD on product sales but do so in different contexts. MS show that the curvilinear effect can take place only in a certain movie segment and in a certain country segment in its movie and country segmentation models. By contrast, our model uses cultural compatibility as a moderator in the curvilinear effect. In other words, we show that such a curvilinear effect involving CD is not universal and is effective only in the presence of cultural compatibility (H₃ and Figure 2).

Next, we acknowledge that GYR and the current study highlight both economic and cultural factors. However, the two studies examine different factors in various ways. On the economic side, GYR examine economic wealth as a main effect and a moderator for two relationships (the relationship between the time lag and the country-level performance of a new product and the relationship between the number of counties and the same country-level performance variable). By contrast, our research examines the inverse U-shaped impact of the economic development status of the product-consuming country on product sales, which is a novel finding (H₁ and Figure 1). On the cultural side, GYR examine the roles of Hofstede's traditional cultural dimensions. Compared with GYR's cultural findings, our approach has two notable points: we use (1) the cultural compatibility of the product and the market (H₂) and (2) the curvilinear effect of CD on movie sales (H₃ and Figure 2).

Managerial Implications

Rapid globalization heightens the importance of global business, which should affect various cultural entertainment products such as music, performances, and fashion. Our research presents several practical implications on that front. With many Hollywood movies having their international release weeks ahead of their release in the United States (e.g., Skyfall [2012]), the manner in which audiences in various countries respond to movies has significant managerial implications (Hennig-Thurau, Houston, and Walsh 2006; Kalish, Mahajan, and Muller 1995). In the era of social media, when reviews are easily available, movie studios need to carefully consider in which countries to release their movies first. Our research suggests that they should keep cultural and economic factors in mind while making that decision.

Indeed, media economists have examined how U.S. movies’ international box office performance varies across countries (Jayakar and Waterman 2000), primarily drawing on the CD between the United States and foreign countries as a general measure of the cultural fit of the product-producing country and product-consuming countries (Kogut and Singh 1988). However, they have not investigated the issue in terms of the variability of U.S. movies’ cultural content as a specific measure of the product purchased and consumed. Some U.S. movies (e.g., Garden State [2004]) contain strong cultural references specific to U.S. audiences, whereas others (e.g., Harry Potter and the Prisoner of Azkaban [2004]) are relatively more universal in a cultural sense. It is known that less culturally American movies tend to perform better outside the United States (Lee 2006). Therefore, measuring individual movies’ specific cultural content and understanding the role of this content in international box office performance can help U.S. movie studios produce and promote their movies more effectively. This problem is becoming more important because international movie markets, particularly Asian markets, are growing quickly, while U.S. movie ticket sales have been declining since their peak in 2002 (www.the-numbers.com/market; see also Schuker 2010).

When movie studios are selecting countries in which particular U.S. movies are to be introduced, our research suggests that they should consider countries whose cultures match the cultural content of the movie. For example, if movie studios want to release a film such as King Arthur (2004), they should consider European countries first. For a film such as Sinbad: Legend of the Seven Seas (2003), they may want to consider Middle Eastern countries first. Furthermore, in countries whose culture generally matches the cultural content of the movie, movie studios may still want to consider either culturally close or culturally distant countries to the United States rather than culturally medium-distance countries.

In addition, our results suggest that movie studios should first consider countries in the medium stage of economic development rather than the wealthiest or poorest countries because medium-stage countries tend to have enough purchasing power without strong domestic competitors. Finally, increasingly more movies are being produced with modified versions created for different countries. For example, Iron Man 3 (2013) has an exclusive Chinese version featuring extra scenes with more Chinese actors. This strategy may mean increased production costs, but it presents increased revenue potential in foreign markets. When consumers’ cultural orientations and preferences are illuminated through online movie reviews, the multiple-version strategy can be more effective.

Limitations and Future Research Avenues

Pointing to the numerous nuances inherent in cultural products, Holbrook and Schindler (1994) suggest that cultural products are so multifaceted as to discourage the application of common methods suitable for other products. It is worth noting that, therefore, our hypotheses are more applicable to global cultural products and less so to utilitarian or high-tech products (Dhar and Wertenbroch 2000). Even within global cultural products, the empirical results from testing the proposed hypotheses can vary to some extent. In addition, we focus on the offerings of global cultural products from economically developed countries (particularly, U.S. movies) rather than economically developing countries because most popular cultural global products are produced in developed countries (e.g., U.S. movies, French wines, Italian fashion). Moreover, because the market mechanism and consumer reactions toward global cultural products from developing countries would be drastically different (Guo 2013), this specific subject would be more effectively examined in a separate study. Because of these details pertaining to the product category and the product-producing country (specifically, U.S. movies), it remains to be seen whether the proposed hypotheses hold for other product categories and product-producing countries.

In obtaining the measure of cultural compatibility between the product and the market, we used IMDb as the single information source that generated the cultural content of the product. Although this website presents deep and detailed views from a variety of reviewers around the globe, it may not reflect all of the cultural thoughts from those who have limited access to the website because of linguistic or technological barriers. Moreover, because reviewers on the website post their reviews voluntarily, there may be self-selection bias. We leave these potential sources of bias as a potential future research topic. Furthermore, we tested the three-way interaction effect of cultural compatibility × CD × per capita GDP as a way of shedding light on the potentially complicated interactions of multiple cultural and economic factors. Because examining this issue requires more sophisticated modeling efforts, we leave this notable but challenging topic for further research.

Finally, while this research focuses on the impacts of cultural and economic factors regarding country-level demand for U.S. movies, a future study could draw on new institutional economics to develop a model concerning the “contractual details” between product-consuming and product-producing countries to significantly expand the scope of this research. Such a model may include governance structure, contracting theory, transaction cost, and path dependence in the international diffusion of global cultural products (North 1995; Swati and Divya 2011).

Footnotes

Elicited Terms for Five Cultural Categories

Country Group	Country	Representative Cultural Terms
United States	United States	American dream, baseball, blue jeans, California, Civil War, diversity
Germany-Australia	Australia	Aboriginal, cricket, crocodile, English, swimming, tennis, wine
	Austria	Alps, beer, cheese, classical music, German, ski, tourism, Viennese
	Denmark	Castles, Christian, Danish, handball, monarchy, same-sex marriage
	Finland	Baltic, European, Finnish, Nordic, oatmeal, reindeer, Russian, skiing
	Germany	Beer, classical music, efficient, engineers, sausage, soccer, tennis
	Israel	Alienated, Arab Israelis, education, Jews, Hebrew, Holocaust, Zionist
	Netherlands	Beer, Calvinism, Dutch, king, speed skating, painting, Renaissance
	New Zealand	Anglican, ANZAC, drum and bass, English, Maori, Pacific, travel
	Norway	Cod, Jante Law, mountain plateaus, salmon, ski, Viking, winter sports
	South Africa	Afrikaans, boxing, Dutch, English, postapartheid, rugby, wine, Zulu
	Sweden	Cider, death metal, fish, Lutheran, midsummer, Rinkeby, Viking
	Switzerland	Alps, cheese, French, German, mountain biking, tennis, yodeling
	United Kingdom	BBC, Beatles, cricket, English, Industrial Revolution, Irish, royalty Italy-Japan
Italy-Japan	Hungary	Barack, ceramics, multicultural, rhapsodies, water polo, wine
	Italy	Architecture, art, cappuccino, espresso, fashion, olive oil, pizza, Roman
	Japan	Baseball, Godzilla, hiragana, judo, kabuki, karaoke, manga, samurai
	Slovakia	Ice hockey, lakes, Roman Catholic, Slavic, Slovac, volcanoes
Spain-Brazil	Argentina	Argentine rock, Catholic, simple pottery, soccer, tango, tennis
	Belgium	Baroque, beer, Brussels, chocolate, Dutch, Flanders, French-speaking
	Brazil	Amazon, Bossa Nova, carnival, coastal, coffee, Portuguese, Rio
	Bulgaria	Balkan, Black Sea, communism, Danube, Gypsies, Turkey, yogurt
	Chile	Andes, beaches, Catholic, Chilean peso, colonized, volcanic, wine
	Colombia	Andes, Catholic, coffee, colonization, El Dorado, Hispanic, Pacific
	Czech	Art Nouveau, beer festivals, Bohemia, cathedrals, Central Europe
	Ecuador	Alpaca textiles, banana, carnival, Ecuadorian real, Galapagos
	Egypt	Alexandria, Arabic, Cairo, camels, earliest civilization, Nile, pyramid
	France	Alps, architecture, arts, café, croissant, cuisine, Louvre Museum
	Greece	Democracy, Greek, Orthodox Church, gyro, Homer, Olympics
	Mexico	Aztec, burrito, Cancun, Catholic, chili peppers, soccer, Spanish
	Peru	Andean music, Andes, avocado, Inca, Indians, Lima, poncho, Spanish
	Poland	Chopin, European Union, kompot, Nazi invasions, Roman Catholic
	Portugal	Atlantic, beaches, Cristiano Ronaldo, Lisbon, Portuguese, soccer
	Romania	1989 Revolution, Black Sea, Bucharest, Danube, dictator, gymnastics
	Russia	Ballet, Bolshevik, Crimea, czars, gymnastics, hammer and sickle
	South Korea	Acupuncture, baduk, Buddhism, division of Korea, hangul, Samsung
	Spain	Barcelona, Basque, bull fighting, Catalan, Columbus, monarchy
	Taiwan	Ang Lee, baseball, Buddhism, rice, Taiwanese, Taoist, typhoons
	Thailand	Agriculture, Bangkok, boxing, Buddha, emerging economy, monarchy
Spain-Brazil	Turkey	Basketball, Byzantine Empire, Istanbul, Kurds, secular, Turkish coffee
	United Arab Emirates	Abu Dhabi, Arabian Peninsula, Dubai, fishing, mosques, skyscraper
	Venezuela	Baseball, beautiful women, Caracas, Chavez, petroleum, poverty
India-Indonesia	Hong Kong	British, Buddhism, Cantonese, Jackie Chan, kung fu, water sports
	India	Bollywood, caste, English, Hindi, Hinduism, Punjabi, Taj Mahal
	Indonesia	Badminton, Bali, bamboo, Islam, Javanese, Kroncong, Muslim, rice
	Malaysia	Chinese culture, dragon dancing, Malays, Muslim, water villages
	Philippines	Basketball, Catholic, cock fighting, coconut, Filipino, seafood, yo-yo
	Singapore	Bilingual, British, cosmopolitan, English, government control, seafood

References

Archak

Nikolay

, Ghose

Anindya

, and Ipeirotis

Panagiotis G.

(2011), “Deriving the Pricing Power of Product Features by Mining Consumer Reviews,” Management Science, 57(8), 1485–509.

Arndt

Sven W.

(1999), “Globalization and Economic Development,” Journal of International Trade & Economic Development: An International and Comparative Review, 8(3), 309–18.

Applbaum

Kalman

, and Jordt

Ingrid

(1996), “Notes Toward an Application of McCracken's ‘Cultural Categories’ for Cross-Cultural Consumer Research,” Journal of Consumer Research, 23(3), 204–18.

Arnould

Eric J.

, and Thompson

Craig J.

(2005), “Consumer Culture Theory (CCT): Twenty Years of Research, “Journal of Consumer Research, 31(4), 868–82.

Ashraf

Abdul R.

, Thongpapanl

Narongsak (Tek)

, and Auh

Seigyoung

(2014), “The Application of the Technology Acceptance Model under Different Cultural Contexts: The Case of Online Shopping Adoption,” Journal of International Marketing, 22(3), 68–93.

Bakker

Gerben

(2005), “The Decline and Fall of the European Film Industry: Sunk Costs, Market Size, and Market Structure, 1890–1927,” Economic History Review, 58(2), 310–51.

Boskem

Maarten A.S.

, and Smidts

Ale

(2015), “Brain Responses to Movie Trailers Predict Individual Preferences for Movies and Their Population-Wide Commercial Success,” Journal of Marketing Research, 52(August), 482–92.

Carvalho

Sergio W.

, Block

Lauren G.

, Sivaramakrishnan

Subramanian

, Manchanda

Rajesh V.

, and Mitakakis

Chrissy

(2008), “Risk Perception and Risk Avoidance: The Role of Cultural Identity and Personal Relevance,” International Journal of Research in Marketing, 25(4), 319–26.

Chandrasekaran

Deepa

, and Tellis

Gerard J.

(2008), “Global Takeoff of New Products; Culture, Wealth, or Vanishing Differences?” Marketing Science, 27(5), 844–60.

10.

Chevalier

Judith A.

, and Mayzlin

Dina

(2006), “The Effect of Word of Mouth on Sales: Online Book Reviews,” Journal of Marketing Research, 43(August), 345–54.

11.

Cui

Geng

, and Liu

Qiming

(2001), “Emerging Market Segments in a Transitional Economy: A Study of Urban Consumers in China,” Journal of International Marketing, 9(1), 84–106.

12.

Craig

C. Samuel

, Greene

William H.

, and Douglas

Susan P.

(2005), “Culture Matters: Consumer Acceptance of U.S. Films in Foreign Markets,” Journal of International Marketing, 13(4), 80–103.

13.

Dawar

Niraj

, and Parker

Philip

(1994), “Marketing Universals: Consumers’ Use of Brand Name, Price, Physical Appearance, and Retailer Reputation as Signals of Product Quality,” Journal of Marketing, 58(April), 81–95.

14.

Decker

, and Trusov

(2010), “Estimating Aggregate Consumer Preferences from Online Product Reviews,” International Journal of Research in Marketing, 27(4), 293–307.

15.

Dhar

Ravi

, and Wertenbroch

Klaus

(2000), “Consumer Choice Between Hedonic and Utilitarian Goods,” Journal of Marketing Research, 37(February), 60–71.

16.

Dimofte

Claudio V.

, Johansson

Johny K.

, and Ronkainen

Ilkka A.

(2008), “Cognitive and Affective Reactions of U.S. Consumers to Global Brands,” Journal of International Marketing, 16(4), 113–35.

17.

Eisingerich

Andreas B.

, and Rubera

Gaia

(2010), “Drivers of Brand Commitment: A Cross-National Investigation,” Journal of International Marketing, 18(2), 64–79.

18.

Elliott

Gregory R.

, and Cameron

Ross C.

(1994), “Consumer Perception of Product Quality and the Country-of-Origin Effect,” Journal of International Marketing, 2(2), 49–62.

19.

Feldman

Ronen

, and Sanger

James

(2007), The Text Mining Handbook: Advanced Approaches in Analyzing Unstructured Data. Cambridge, UK: Cambridge University Press.

20.

W. Wayne

, and Lee

Tracy K.

(2008), “Economic and Cultural Influences on the Theoretical Consumption of Foreign Films in Singapore,” Journal of Media Economics, 21(1), 1–27.

21.

Ghose

Anindya

, Ipeirotis

Panagiotis G.

, and Li

Beibei

(2012), “Designing Ranking Systems for Hotels on Travel Search Engines by Mining User-Generated and Crowd-Sourced Content,” Marketing Science, 31(3), 493–520.

22.

Greene

William H.

(2003), Econometric Analysis, 3rd ed. Upper Saddle River, NJ: Prentice Hall.

23.

Griffith

David A.

, Cavusgil

Salih Tamer

, and Xu

Schichun

(2008), “Emerging Themes in International Business Research,” Journal of International Business Studies, 39(7), 1220–35.

24.

Griffith

David A.

, Cavusgil

Salih Tamer

, and Xu

Schichun

, Yalcinkaya

Goksel

, and Rubera

Gaia

(2014), “Country-Level Performance of New Experience Products in a Global Rollout: The Moderating Effects of Economic Wealth and National Culture,” Journal of International Marketing, 22(4), 1–20.

25.

Guo

Xiaoling

(2013), “Living in a Global World: Influence of Consumer Global Orientation on Attitudes Toward Global Brands from Developed Versus Emerging Countries,” Journal of International Marketing, 21(1), 1–22.

26.

Hennig-Thurau

Thorsten

, Houston

Mark B.

, and Heitjans

Torsten

(2009), “Conceptualizing and Measuring the Monetary Value of Brand Extensions: The Case of Motion Pictures,” Journal of Marketing, 73(November), 167–83.

27.

Hennig-Thurau

Thorsten

, Houston

Mark B.

, and Heitjans

Torsten

, and Walsh

Gianfranco

(2006), “The Differing Roles of Success Drivers Across Sequential Channels: An Application to the Motion Picture Industry,” Journal of the Academy of Marketing Science, 34(4), 559–75.

28.

Hofstede

Geert

, Hofstede

Gert Jan

, and Minkov

Michael

(2010), Cultures and Organizations: Software of the Mind, 3rd ed. New York: McGraw-Hill.

29.

Holbrook

Morris B.

, and Schindler

Robert M.

(1994), “Age, Sex, and Attitude Toward the Past as Predictors of Consumers’ Aesthetic Tastes for Cultural Products,” Journal of Marketing Research, 31(August), 412–22.

30.

Hoskins

, and Mirus

(1988), “Reasons for U.S. Dominance of the International Trade in Television Programmes,” Media Culture & Society, 10(4), 499–515.

31.

Hsieh

Ming H.

(2002), “Identifying Brand Image Dimensionality and Measuring the Degree of Brand Globalization: A Cross-National Study,” Journal of International Marketing, 10(2), 46–67.

32.

Hultman

Magnus

, Robson

Matthew J.

, and Katsikeas

Constantine S.

(2009), “Export Product Strategy Fit: An Empirical Investigation,” Journal of International Marketing, 17(4), 1–23.

33.

Jayakar

K.P.

, and Waterman

(2000), “The Economics of American Theatrical Movie Exports: An Empirical Analysis,” Journal of Media Economics, 13(3), 153–69.

34.

Jeong

Soon-Ok

, and Park

Suk-Hee

(1997), “A Cross-Cultural Application of the Novelty Scale,” Annals of Tourism Research, 24(1), 238–40.

35.

Joachims

Thorsten

(1998), “Text Categorization with Support Vector Machines: Learning with Many Relevant Features,” Machine Learning, ECML-98, 137–42.

36.

Kale

Sudhir H.

(1986), “Dealer Perceptions of Manufacturer Power and Influence Strategies in a Developing Country,” Journal of Marketing Research, 23(November), 387–93.

37.

Kalish

Shlomo

, Mahajan

Vijay

, and Muller

Eitan

(1995), “Waterfall and Sprinkler New-Product Strategies in Competitive Global Markets,” International Journal of Research in Marketing, 12(2), 105–19.

38.

Katsikeas

Constantine S.

, Samiee

Saeed

, and Theodosiou

Marios

(2006), “Strategy Fit and Performance Consequences of International Marketing Standardization,” Strategic Management Journal, 27(9), 867–90.

39.

Kjeldgaard

Dannie

, and Askegaard

Søren

(2006), “The Glocalization of Youth Culture: The Global Youth Segment as Structures of Common Difference,” Journal of Consumer Research, 33(2), 231–47.

40.

Kogut

Bruce

, and Singh

Harbir

(1988), “The Effect of National Culture on the Choice of Entry Mode,” Journal of International Business Studies, 19(3), 411–32.

41.

Kraidy

Marwin M.

(2002), “Hybridity in Cultural Globalization,” Communication Theory, 12(3), 316–39.

42.

Lee

Francis L.F.

(2006), “Cultural Discount and Cross-Culture Predictability: Examining the Box Office Performance of American Movies in Hong Kong,” Journal of Media Economics, 19(4), 259–78.

43.

Lee

Thomas Y.

, and Bradlow

Eric T.

(2011), “Automated Marketing Research Using Online Customer Reviews,” Journal of Marketing Research, 48(October), 881–94.

44.

Levitt

Theodore

(1983), “The Globalization of Markets,” Harvard Business Review, 61(3), 92–102.

45.

Lizardo

Omar

(2008), “Cultural Consumption in the Fine and Popular Arts Realms,” Sociology Compass, 2(2), 485–502.

46.

Markus

H.R.

, and Kitayama

(1991), “Culture and the Self: Implications for Cognition, Emotion, and Motivation,” Psychological Review, 98(2), 224–53.

47.

McCracken

Grant

(1986), “Culture and Consumption: A Theoretical Account of the Structure and Movement of the Cultural Meaning of Consumer Goods,” Journal of Consumer Research, 13(June), 71–84.

48.

Moon

Sangkil

, Bergey

Paul K.

, and Iacobucci

Dawn

(2010), “Dynamic Effects Among Movie Ratings, Movie Revenues, and Viewer Satisfaction,” Journal of Marketing, 74(January), 108–21.

49.

Moon

Sangkil

, Bergey

Paul K.

, and Iacobucci

Dawn

, Park

Yoonseo

, and Kim

Yong Seog

(2014), “The Impact of Text Product Reviews on Sales,” European Journal of Marketing, 48(11/12), 2176–97.

50.

Moon

Sangkil

, Bergey

Paul K.

, and Iacobucci

Dawn

, Park

Yoonseo

, and Kim

Yong Seog

, and Song

Reo

(2015), “The Roles of Cultural Elements in International Retailing of Cultural Products: An Application to the Motion Picture Industry,” Journal of Retailing, 91(1), 154–70.

51.

Newman

Karen L.

, and Nollen

Stanley D.

(1996), “Culture and Congruence: The Fit Between Management Practices and National Culture,” Journal of International Business Studies, 27(4), 753–79.

52.

North

Douglass C.

(1995), “The New Institutional Economics and Third World Development,” in The New Institutional Economics and Third World Development, Harriss

, Hunter

, and Lewis

C.M.

, eds. London: Routledge, 17–26.

53.

Özsomer

Ayşegul

(2012), “The Interplay Between Global and Local Brands: A Closer Look at Perceived Brand Globalness and Local Iconness,” Journal of International Marketing, 20(1), 71–94.

54.

Ryans

John K.

Jr. , Griffith

David A.

, and White

Steven

(2003), “Standardized/Adaptation of International Marketing Strategy: Necessary Conditions for the Advancement of Knowledge,” International Marketing Review, 20(6), 588–603.

55.

Schilke

Oliver

, Reimann

Martin

, and Thomas

Jacquelyn S.

(2009), “When Does International Marketing Standardization Matter to Firm Performance?” Journal of International Marketing, 17(4), 24–46.

56.

Schuker

Lauren A.E.

(2010), “Plot Change: Foreign Forces Transform Hollywood Films,” The Wall Street Journal, (August 2), (accessed June 27, 2016), [available http://www.wsj.com/articles/SB10001424052748704913304575371394036766312].

57.

Scott

Allen J.

(2004), “Cultural-Products Industries and Urban Economic Development: Prospects for Growth and Marketing Contestation in Global Context,” Urban Affairs Review, 39(4), 461–90.

58.

Sheth

Jagdish N.

(2011), “Impact of Emerging Markets on Marketing: Rethinking Existing Perspectives and Practices,” Journal of Marketing, 75(July), 166–82.

59.

Soares

Ana Maria

, Farhangmehr

Minoo

, and Shoham

Aviv

(2007), “Hofstede's Dimensions of Culture in International Marketing Studies,” Journal of Business Research, 60(3), 277–84.

60.

Strizhakova

Yuliya

, Coulter

Robin A.

, and Price

Linda L.

(2008), “Branded Products as a Passport to Global Citizenship: Perspectives from Developed and Developing Countries,” Journal of International Marketing, 16(4), 57–85.

61.

Swati

Sharma

, and Divya

Vyas

(2011), “A New Horizon in Agriculture Marketing: Contract Farming,” International Journal of Marketing and Management Research, 2(5), 196–206.

62.

Tellis

Gerard J.

, Stremersch

Stefan

, and Yin

Eden

(2003), “The International Takeoff of New Products: The Role of Economics, Culture, and Country Innovativeness,” Marketing Science, 22(2), 188–208.

63.

Thompson

Frauke Mattison

, and Chmura

Thorsten

(2015), “Loyalty Programs in Emerging and Developed Markets: The Impact of Cultural Values on Loyalty Program Choice,” Journal of International Marketing, 23(3), 87–103.

64.

Tihanyi

Laszlo

, Griffith

David A.

, and Russell

Craig J.

(2005), “The Effect of Cultural Distance on Entry Mode Choice, International Diversification, and MNE Performance: A Meta-Analysis,” Journal of International Business Studies, 36(3), 270–83.

65.

Tse

Peter

, and Crotts

John C.

(2005), “Antecedents of Novelty Seeking: International Visitors Propensity to Experiment Across Hong Kong's Culinary Traditions,” Tourism Management, 26, 965–68.

66.

Van Everdingen

Yvonne

, Fok

Dennis

, and Stremersch

Stefan

(2009), “Modeling Global Spillover of New Product Takeoff,” Journal of Marketing Research, 46(October), 637–52.

67.

Wang

Cheng-Lu

, Chen

Zhen-Xiong

, Chan

Allan K.K.

, and Zheng

Zong-Cheng

(2000), “The Influence of Hedonic Values on Consumer Behaviors,” Journal of Global Marketing, 14(1/2), 169–86.

68.

Zou

Shaoming

, and Tamer Cavusgil

(2002), “The GMS: A Broad Conceptualization of Global Marketing Strategy and Its Effect on Firm Performance,” Journal of Marketing, 66(October), 40–56.