Trade Liberalization and Policy for Media Industries: A Theoretical Examination of Media Flows

Steven S. Wildman (Northwestern University)

Abstract: The recently developed microeconomic model of one-way international flows in films and television programs is shown to be a specific application of a more general model of trade in media products. Predictions of the general model are consistent with observed international flows for other media, geographic flows of media products within countries, and intertemporal flows of media products across distribution channels. The general tendency is for product to flow from large to small markets unless impeded by policy-created barriers. Earlier explanations for one-way international flows do not similarly generalize, but cultural explanations and the microeconomic model are complementary to each other. The role of market size in determining media trade flows and the effect of domestic media policies on the growth and relative sizes of national media markets should be considered in the formation of trade policies for media industries.

Résumé: Dans cet article, nous discutons du modèle micro-économique récemment développé sur la circulation internationale unidirectionnelle de films et d'émissions de télévision. Nous montrons que ce modèle est l'application spécifique d'un modèle plus général d'échanges en produits médiatiques. Ce modèle plus général explique davantage que le cinéma et la télévision. En effet, il s'applique aux circulations internationales observées pour d'autres médias, à la circulation à l'intérieur d'un pays, et aux fluctuations temporelles dans les échanges de produits médiatiques entre pays. En général, les produits tendent à circuler des grands marchés aux petits, dans les cas où il n'y a pas de politiques pour faire obstacle à de tels échanges. Des explications antérieures à la nôtre pour la circulation internationale en sens unique ne généralisent pas de la même manière. Nous croyons que les explications culturelles et le modèle micro-économique sont très complémentaires. Avant de formuler des politiques d'échange pour les industries médiatiques, il est ainsi nécessaire de tenir compte de certaines questions; il faut notamment être conscient de la grandeur du marché, et des effets que les politiques domestiques ont produits sur les marchés médiatiques nationaux.


The world trading system is liberalizing rapidly. Through worldwide agreements such as the GATT and regional liberalization efforts such as the North American Free Trade Agreement and the opening of borders within the European Community, barriers to the free flow of goods and services are being torn down. This process has elicited the expected complaints and predictions of disaster from national industries losing their protection, and--given the domestic political constituencies these industries represent--a number of exceptions to the general principles of liberalization have been made. It is doubtful, however, that liberalization of trade for any industry has been subject to as much scrutiny or aroused as much passion as it has for the cultural industries, films and television programs in particular. With the possible exception of strategic defense industries, it is also doubtful that any industry has been accorded so many exceptions to general free trade principles as have cultural industries.

The passions aroused by trade in media products, and therefore the liberalization of that trade, are understandable. Mass media both reflect and help shape cultures and national identities. As a result, the link between a country's culture and its domestic media tends to be strong and emotionally charged, and perceived threats to this link are not taken lightly. Emotional ties, however, are less-than-adequate guides to the formulation of media trade policies because they obscure important trade-offs. Import restrictions may achieve some success in keeping out unwanted cultural influences and assuring a domestic audience for domestic productions; but, given the crudeness of trade protection mechanisms, a protectionist country's citizens are almost certain to suffer from less than full access to the beneficial ideas and values of other cultures. Furthermore, retaliatory actions by other countries could reduce the access of domestic media to other markets and, perhaps more likely, harm a country's non-media exports. Therefore, it is essential that in formulating media trade policies we have a clear understanding of the economic forces underlying media trade flows.

Theory development in this field has been largely a response to the overwhelming dominance of U.S. films and television programs in the international trade in audiovisual products. To take Europe as an example, in 1992 European countries spent $3.7 billion on U.S. audiovisual products and services while managing sales to the U.S. of only $288 million (Waxman, 1993). (For consistency, these and other financial magnitudes reported in this paper are denominated in U.S. dollars.) This is not a recent phenomenon. American productions have dominated the trade in video products between Europe and the United States for decades (Guback, 1969). While the European situation is probably best documented, it is by no means unique. If anything, the relatively advanced European video industries fare quite well in comparison to the less-developed video industries of most other countries. Japan has been the largest importer of U.S. films in recent years, and, as noted by Biltereyst (1992), as a general rule, media industries in smaller countries tend to be most dependent on American films and programs.

Such a glaring and persistent imbalance would seem to demand an explanation, and media scholars have been more than obliging in their supply of theories. The microeconomic model advanced in this paper is a fairly recent addition to the list.

The microeconomic model of trade in media products

Variants of the basic model were published independently by three different sets of authors studying trade in films and television programs in the late 1980s (Hoskins & Mirus, 1988; Waterman, 1988; Wildman & Siwek, 1987, 1988). It has since been elaborated and extended to trade in other media by these and other authors (Frank, 1992; Owen & Wildman, 1992, chap. 2; Waterman, 1993; Wildman & Siwek, 1993).

At the heart of the model is the economist's distinction between private and public goods. Consumption of a private good is excludable in that one person's consumption precludes another's. A candy bar is a good example. What Tom eats, Mary can't. Public goods are non-excludable in consumption in that one person's consumption of a public good does not diminish the amount remaining to be consumed by others (or consumed again by those who wish to do so). While private goods, such as books and CDs, may be used to store and distribute the content of a media product, the content is a public good which is not used up in consumption. For example, there is no inherent limit to the number of people who may enjoy the plays of Shakespeare--or a popular film. The public-good private-good distinction is critical to the analysis of trade in media products because public-good content elements drive the demand for media products, and the production of content tends to dominate costs, especially for films and television programs.

While the economic incentives driving trade are very different for public and private goods, it is the more traditional private-good models that typically have been employed to explain media trade flows. This misapplication of private-good models is one reason some people have argued that something other than simple competitive interactions is required to explain the one-way trade in films and television programs (Pruvot, 1983).

The critical difference between public-good and private-good models of trade is that domestic market size, which plays no necessary role in private-good models, matters a lot in a model of trade in which a product's public-good components (content for media products) drive demand. Profit maximization requires that expenditures on content-related inputs be increased as long as the expected response from the prospective audience (bigger ticket sales at theatres, larger cassette rentals, larger television audiences, etc.) produces enough additional revenue to more than cover the increase in cost. If we think of the effect of added production expenditures in terms of an increased likelihood that a prospective viewer will decide to watch the more expensive program or film, then clearly the expected financial return to expenditures on content-related inputs should be larger in a market with more potential viewers, because the effect on the representative viewer is multiplied by the number of viewers in the market. In other words, the larger is the market, the larger is the budget required to reach the point at which a dollar added to the production budget no longer generates at least an extra dollar in expected revenue. An increase in per-capita wealth or income should have a similar effect, since a wealthier audience can be expected to spend more on media products and to be worth more to advertisers. The result, then, is the production of more expensive films and programs for larger markets, where market size is measured in financial terms.

The effect of the number of viewers in a market on budgets for films and programs is illustrated in Figure 1, where for simplicity the media product is assumed to be a motion picture and cinema ticket sales are assumed to be the sole source of revenue. We compare a decision on whether to increase a film's production budget from an initial level in>1 to a higher level, in>2, for each of two countries, A and B. Va is the number of viewers in A and Vb is the number of viewers in B. For this illustration, Vb is three times Va. For both countries, P1 is the percentage of the population that on average will attend a film with budget in>1 and P2 is the expected percentage attendance if the budget is in>2. In other words, budgets in>1 and in>2 yield expected audiences of P1Va and P2Va and P1Vb and P2Vb in markets A and B, respectively. Attendance is higher for more expensive films because a larger production budget enables a producer to acquire more of the inputs that contribute to a film's audience appeal, and--when the quality of an input affects audience appeal--spend more for higher-quality inputs, where input quality is measured in terms of contribution to box-office success rather than the aesthetic dimensions of the final product. Examples of higher-quality inputs are bigger name stars, better directors and writers, high-tech special effects, and more intensive editing. N is the producer's net after the theatre's take on each ticket sold.

A profit-maximizing producer will increase the budget from in>1 to in>2 only if the increase in expected revenue is at least as large as in>2 in>1. For market A the increase in expected net revenue is the increase in the expected audience, Va(P2 P1), times N, which is the area of the smaller of the two long, light gray rectangles under the line at N. In market B the increase in expected net revenue is Vb(P2 P1) times N, which is the area of the larger light gray rectangle.

20P 1 Effect of Market Size on Production Budget

Clearly the expected return to this increase in the production budget is greater in the larger market. A critical implication of this relationship is that there will be input expenditures contributing to audience appeal that also contribute to profits in large markets that would reduce profits in small markets. For example, suppose that the inputs required to generate the increases in audience and net revenue shown in Figure 1 cost slightly more than the increase in expected net revenue in the smaller market. While these expenditures would be financially irresponsible in the smaller market, they would still make a healthy contribution to (expected) profits in the large market. Of course, diminishing returns means that eventually additional expenditures will be unprofitable even in the largest market, but much larger budgets are required to reach that point.

It is also easy to see how an increase in the spending power of individual consumers (due to greater wealth or income) can have an effect similar to the effect of a larger population on the size of profit maximizing production budgets within a market. Wealthier viewers spend more on media products and, as audiences, are worth more to advertisers. Both effects increase the financial return in audience share purchased with increased production budgets. This is illustrated by the dashed line at N indicating a higher expected net profit per viewer in a wealthier market. The smaller, darker rectangles represent the additional value of the same increase in market share due to the greater spending power of wealthier viewers. So, other things such as audience size held equal, more expensive films and programs should be produced for wealthier markets.

For the neoclassical media trade model, it is the relationship between market size and production budgets that leads to one-way flows. The larger worldwide market in films and programs is composed of numerous distinct submarkets, demarcated primarily by differences in language and culture that make translation necessary if the video products of one market are to find an audience in another. The fact that something of the essence of an artistic work is always lost in translation creates a natural preference among viewers for films and programs produced in their native language by domestic producers who are sensitive to the peculiarities and subtleties of their culture. For foreign-language films to find an audience, they must offer something extra to compensate for this intrinsic linguistic and cultural handicap.

The neoclassical model posits that the something extra explaining the worldwide appeal of U.S. films and programs is the major stars, expensive special effects, and generally higher production values purchased with the large production budgets that maximize profits in the enormous global English-language market. These factors counteract the linguistic and cultural handicaps English-language films encounter in other markets. On the other hand, films from smaller linguistic markets sold in the U.S. must deal with the dual handicaps of a foreign language and culture and much lower production values than what is available domestically. More generally, the neoclassical model posits that films and programs from large markets should have budget-based production value advantages over films from small markets, and that this will be reflected in a video trade balance favouring the large markets.

This is the basic neoclassical model, and it seems to do a good job of explaining the gross features of the global video trade. However, the basic model has an all-or-nothing quality that is inconsistent with actual experience. Clearly the size of the production budget is not the sole determinant of viewing choices. If this were the case, films and programs from a large market would either completely dominate viewer choices in a smaller market or would not be viewed at all, depending on whether production budgets exceeded or fell short of a critical size required to offset the linguistic/cultural handicap in the smaller market. Furthermore, viewers in the large market would never watch films and programs originating in the smaller market. Real-world comparisons typically show films and programs originating in both markets being watched in each, but with the distribution of viewers skewed toward the larger market's productions. In addition, there are markets that are more successful and markets that are less successful as exporters of video products than a simple comparison of market sizes would predict. For example, the German-language market is larger, both in population and income, than the French-language market (Siwek & Furchtgott-Roth, 1993; Wildman & Siwek, 1988), but French films are much more successful internationally.

The basic neoclassical model can be made to better conform to a world in which individual consumers watch films and programs from a variety of international sources by incorporating two other characteristics of film and television production. One is that the critical skills required to make these video products are craft and artistic skills that may be more highly developed in some countries or markets than in others. This could be explained, at least in part, by differences in the values different cultures place on the talents employed in making films and programs. However, the very strong correlation between domestic market size and a country's status as an exporter of video products (Waterman, 1988; Wildman & Siwek, 1988) suggests that large differences in market size overwhelm cultural factors. Rather, cultural influences seem to explain variation within a pattern determined largely by differences in market size. Thus, the worldwide French-speaking audience is somewhat smaller than the German-speaking audience, both numerically and financially (Wildman & Siwek, 1988; Siwek & Furchtgott-Roth, 1993). But film has traditionally been a more important art form in France than in Germany and French films tend to attract larger international audiences (Wildman & Siwek, 1988, Appendix A). However, both French and German films are dramatically overshadowed by the big-budget U.S. films originating in the vastly larger English-language market.

The second characteristic of the film and television industries that should be explicitly incorporated within the neoclassical framework is that, regardless of budget size and despite a variety of mechanisms employed by filmmakers to control quality, the audience appeal of individual films and programs is highly variable and difficult to predict. Viewer preferences change over time, and each film or program, by necessity, must be unique. As a consequence, a high degree of variation in audience appeal is unavoidable, even for films and programs with similar budgets. Thus, Arnold Schwarzenegger can make the high-priced flop Last Action Hero one summer, followed by the similarly high-priced hit True Lies the next. About three quarters of all prime-time U.S. network programs never see a second season (Owen & Wildman, 1992), and the overwhelming majority of feature films fail to break even. Profits from infrequent hits bankroll the process. Larger budgets just raise the mean around which the variation occurs. As a result, while the biggest international hits should come from large markets, there should also be some low-budget films from small markets that outdraw some of the big-budget flops from large markets.

The joint effects of budget size, linguistic and cultural handicaps, and uncontrollable variation in the audience appeal are represented in the four audience-appeal schedules in Figure 2. Schedule Aee ranks for a given year the appeal (strength of a viewer's desire to see it) of all large-budget English-language films to a hypothetical representative native English speaker. Films are ordered in terms of descending appeal from left to right, so the height of the schedule at its far left reflects the appeal of the top-ranked English-language film and the least appealing film is at the far right. The differences in the heights of the two ends of the curve reflect the disparity in outcomes for expensive films with similar budgets. The two Schwarzenegger films just mentioned are cases in point. True Lies would be near the left-hand peak of Aee while Last Action Hero would be close to the bottom on the far right.

Similarly, Aff represents the range of appeal of the same year's crop of French films to a representative native French speaker. Aff lies below Aee because, even though by world standards they are quite expensive, French films still tend to have production budgets nearly an order of magnitude smaller than the U.S. feature films. Films produced in other languages with even smaller domestic markets may be considerably less expensive than French films, and their schedules would fall somewhere below the French film schedule. Finally, the dotted lines Afe and Aef represent the appeal of the available French films to the representative English-language viewer and the appeal of the available English-language films to the representative French viewer, respectively. The difference in height between the representative domestic viewer's valuation of a domestic film and a representative foreign-market viewer's valuation of the same film is measured by the difference in the heights of the domestic and foreign-audience-appeal schedules for that film.

Note that both representative viewers find some foreign films to be preferable to some domestic films. Thus we would expect viewers in both markets to watch films in both languages. However, because the audience-appeal schedules for English-language films are higher to begin with, in the aggregate English-language films will attract larger audiences, and the number of English-language films watched by native French speakers will exceed the number of French-language films watched by English-language viewers. This is indicated by the positions of the two crossover points, X and Y, in Figure 2. At X a horizontal line drawn from the highest point of Aef intersects Aff. The representative French viewer finds all French films up to X more appealing than any U.S. film, but from X on some of the higher-rated U.S. films are preferred. For English-language viewers, Y is the crossover point beyond which some of the higher-rated French films are preferred to the remaining English-language films. Y lies a substantial distance to the right of X because Aee is higher than Aff and Afe is lower than Aef.

20P 2 Production Budgets and Audience Appeal in Two Countries

This completes the presentation of the microeconomic model. A final point worth emphasizing is that within the context of this model it is the relative sizes of linguistically and culturally defined markets, not nations, that matters. While the correlation between the two may be close, it is not appropriate to treat them as the same. The trade statistics of national governments measure cross-border flows, and this tends to direct attention to differences among nations. According to the microeconomic model, this focus on nations as units of analysis obscures more fundamental factors influencing trade in cultural products.

Other explanations for one-way flows

One-way flows were a prominent concern among academics for several decades preceding the development of the microeconomic model, and European policy responses to inflows of American films go back to the early part of this century (Guback, 1969). Three distinct arguments have been prominent in earlier explanations of one-way flows:

  • The imbalance of U.S. exports of films and programs relative to U.S. imports reflects deliberate and illicit policies pursued by the American political/industrial establishment.
  • American media companies have employed unfair and anti-competitive practices to weaken and eliminate rivals in world markets.
  • Unlike producers in other countries, U.S. film and program producers conscientiously select themes and other content elements that appeal to a broad international audience.

In what follows I will refer to these arguments, respectively, as the national imperialism, anti-competitive tactics, and cultural explanations for one-way flows. While these arguments are logically independent of each other, they are not mutually exclusive, and it is sometimes argued that one-way flows are the consequence of two or all three of these mechanisms working in concert.

Herbert Schiller is the most prominent of the scholars associated with the national imperialism argument, which, in its most extreme form posits a plot by the "American military industrial complex to subject the world to military control, electronic surveillance and homogenized American commercial culture" (Schiller, 1969, p. 81). According to this argument, the U.S. government applies pressure on foreign governments and corporations to ensure a wider dissemination of U.S. films and programs than would be achieved if commercial forces alone were operating. The objective is to promote through their representations in the cinema and on TV political values that favour U.S. geopolitical interests and promote demand for the products of U.S. manufacturers.

Guback's (1969) seminal study of the film trade between the United States and Europe is probably the most influential statement of the anti-competitive argument. According to this argument, the dominance of U.S. works in other markets is due to the U.S. majors' use of tactics designed to create and preserve market dominance and monopoly profits--tactics that would violate the competition (anti-trust) laws of the United States and many other nations. Control of distribution channels by the major U.S. producers, who then refuse to distribute films from other countries, plays a prominent role in Guback's analysis; but other authors have pointed to other practices, such as block booking (Pruvot, 1983) and dumping (Schiller, 1969), which are also alleged to be anti-competitive and to further American dominance (Pruvot, 1983).

Cultural analyses of one-way video trade flows focus on cultural forces influencing content decisions in different countries and the extent to which national culture is reflected in the content of video products. Of particular importance is the choice producers make in deciding whether to explore themes and issues of particular interest to people who share their culture or, alternatively, to choose themes and shape content in ways that appeal to viewers in a variety of nations and cultures. It is often argued that American producers go for the international audience, while filmmakers from other countries develop films that appeal to more limited national or cultural audiences (Biltereyst, 1992). Differences in choices among content options may reflect particular artistic and thematic concerns, differences in commercial motivation, or differing skill sets that have developed over time in the film communities of different nations.

An assessment of the evidence and the arguments

In assessing the evidence for the four explanations for one-way flows just reviewed, it is important to recognize that in principle the different activities central to each of these explanations could be occurring simultaneously. That is, the U.S. government could be pressuring other governments and corporations in other countries, U.S. producers could be employing anti-competitive tactics, and U.S. (or English-language) producers could be producing more expensive films and programs and selecting content with an eye to the international market--all at the same time. As a result, there is no simple test that can simultaneously support one of these explanations and disprove the others. Instead, it is necessary to assess the merits of each independently.

A critical starting point is to check each explanation for consistency with facts known to characterize one-way film and program flows. In addition, for those explanations that imply secondary predictions, the accuracy of these predictions should also be checked. It might also be asked of each explanation whether it contributes to our understanding of why the U.S. has consistently realized so much greater export success with video products than with the outputs of other U.S. industries. In other words, does the explanation fit well within a larger pattern of which films and programs are only a small part?

Finally, the mass media are characterized by other prominent, though less noted, one-way flows that are similar in many respects to the international film and program flows. This raises the interesting possibility that one-way flows are reflections of a certain underlying logic of industrial organization that is operative across mass-media industries. Therefore, the extent to which models devised to explain international trade flows can be generalized to explain apparently similar media flows that arise in other situations might be taken as further evidence of the veracity of the basic assumptions underlying these explanations. While it is possible that one-way flows might arise for different reasons in different industries, specialized explanations become less credible as their numbers multiply.

Consistency with basic characteristics of international film and program flows

The largely one-way flows from the U.S. and English-language countries in general to other countries are the most salient feature of the international trade in films and programs. The two most noteworthy characteristics of these flows are that they are virtually global--the primary exceptions being non-market countries with controlled economies--and that they have persisted for a very long time. A successful explanation of one-way flows must be consistent with and account for the globality and the long-term persistence of these flows.

The microeconomic and cultural explanations are both consistent with the globality of one-way flows. The microeconomic model has video products flowing from larger markets to smaller markets. The English-language market, which includes all English-speaking countries and English speakers in other countries, dwarfs all other linguistic markets in raw spending power, with a collective annual income three to four times that of the analogously defined Japanese and German linguistic markets, which rank second and third on this measure, respectively. The microeconomic model would thus predict that U.S. products would dominate just about all bilateral trading relationships with countries in other linguistic markets. Furthermore, the United States' position as the largest country within the English-language block, combined with comparatively few restrictions on the commercial development of U.S. media relative to other English-language countries, is consistent with the pre-eminence of U.S. films and programs within the larger English-language market. Similarly, the uniqueness of the international outlook ascribed to U.S. producers by the cultural explanation is consistent with one-way flows being a global phenomenon.

The microeconomic and cultural explanations are also both consistent with the long-term persistence of one-way flows from the United States and other English-language countries. Cultural traits may be perpetuated within a population for generations, and, in financial terms, the English-language market was the largest of the world's linguistic markets for a long time before one-way flows became a subject of academic interest.

Anti-competitive tactics does not hold up well as an explanation for why one-way flows should be such a global phenomenon. Excepting the dumping, or below-cost pricing, argument, the various tactics postulated to affect trade depend on control of within-country distribution systems; but the major U.S. studios have been barred from distribution activities in many countries. Yet, U.S. films dominate the box office both in countries in which U.S. majors are allowed to distribute their films and in countries in which they are not (Wildman & Siwek, 1988). So tactics dependent on control of distribution channels cannot explain the global success of U.S. films. Anti-competitive tactics also fails as an explanation for U.S. dominance of international television program sales. Until the recent wave of privatization, television was a publicly provided service under either direct or indirect government control in most countries, which means government control of distribution facilities.

Dumping, or below-cost pricing, does not require control of distribution. However, Hoskins, Mirus, & Rozeboom's (1989) econometric study of prices for U.S. programs in foreign markets found that, if anything, cartel activities raised prices above competitive levels, which should encourage rather than discourage purchases of programs from other countries. Obviously, if anti-competitive tactics cannot explain the global character of one-way flows, they cannot explain persistence either.

Globality is not a problem for the national imperialism explanation because U.S. pressure on policymakers and industry officials in other countries is hypothesized to be a global phenomenon. However, the fact that this explanation does not depend on specific mechanisms for exerting influence makes it essentially non-falsifiable on grounds of globality.

Persistence over time is another matter. The fact that U.S. video productions have dominated trade flows since early in the century means the phenomenon has persisted through eras when the U.S. was isolationist, such as prior to the First World War and during much of the interwar period, as well as when its foreign policy was interventionist. The immediate postwar era, when the U.S. was the clearly pre-eminent world power, both economically and militarily, should have been the high-water mark of U.S. leverage in international matters. Yet international sales of U.S. films and programs rose even as the Soviet Union grew to be a formidable military and geopolitical rival and while the U.S. share of the world economy declined as countries devastated by the war recovered and grew to challenge, and frequently best, the U.S. in industry after industry.

Secondary predictions

While the flow of programs and films from the U.S. to other countries is the primary phenomenon of interest, the internal logic of an explanation for these flows may imply other empirical regularities that may also be used to test its validity. Secondary predictions are probably clearest for the microeconomic model, and consistency of observation with secondary predictions has been one of the strongest arguments offered by the advocates of this explanation for one-way flows. Two secondary predictions have received particular attention:

  • Production budgets, as well as export success, will be positively correlated with domestic (or linguistic) market size; and
  • The relationship between export success and relative sizes of home markets will hold for any pair of countries, not just the U.S. and its trading partners.

The first secondary prediction is supported by data on film budgets reported by Wildman & Siwek (1988) and data on production budgets for television programs reported by Waterman (1988). Budgets for U.S. feature films were five to six times larger than budgets for films from France, Italy, and Japan, which produced the most expensive films in languages other than English. While Waterman found smaller multiples for programs, the differences in budgets were still substantial.

The second secondary prediction is borne out in UNESCO data on film exports analyzed in Wildman & Siwek (1988) and data on television program exports reported by Varis & Nordenstreng (1974). In each case, while the U.S. clearly stands alone at the top, the populations of the second tier of exporting countries speak the languages of major linguistic markets such as French, German, Japanese, and Hindi/Urdu.

Secondary predictions have not been emphasized in the literature associated with the national imperialism explanation of one-way flows, but several seem fairly obvious. One is that the export success of U.S. films and programs should fluctuate with variation in U.S. military and economic might relative to the rest of the world. As discussed above, there is no evidence that this has happened. Another is that the countries most open to video imports from the U.S. should be countries most dependent on the U.S., either militarily or economically, since greater dependence should make these countries more susceptible to pressures exerted by U.S. policymakers. Here again, the evidence is not supportive of the national imperialism explanation. No country is more dependent on U.S. aid than South Korea, where many thousands of U.S. troops are stationed as a trip-wire along the demilitarized zone separating it from North Korea. Yet South Korea has historically maintained some of the world's most severe restrictions on imports of U.S. films and programs.

Secondary predictions are less obvious for the anti-competitive tactics and cultural explanations of one-way flows. However, the empirically supported secondary prediction of the microeconomic model poses challenges to both of these explanations. If success in using anti-competitive tactics explains imbalances in video trade flows, then it must be the case that film and program producers from all successful exporting countries employ them and, for reasons unexplained, success in using these tactics increases with domestic market size. Similarly, if generality of content explains international success, then proponents of the cultural explanation have the burden of showing why generality should be positively correlated with both budget size and domestic market size. Of course it is possible that content decisions regarding breadth of appeal may be responses to economic decisions--a position argued for below--but in this case cultural factors are no longer primary, but rather are secondary manifestations of economic forces.

Plausibility within a larger context

U.S. successes in international film and program sales are not typical of the performance of U.S. goods and services overall. Trade balances for most U.S. industries are negative and trade deficits have been a political issue within the United States for well over a decade. Furthermore, with the possible exception of agriculture, it is doubtful that trade surpluses in any industry have persisted for as long as they have in films, or even in the newer medium of television. Given that U.S. films and programs are conspicuous among U.S. exports for their success in foreign markets, it seems reasonable to ask whether the explanations offered for one-way film and program flows have at least facial validity in this larger context.

Nothing inherent in the logic of either the national imperialism explanation or the anti-competitive practices explanation appears to provide a serious response to the query. Domination of film and program flows may have appeared to be part of a larger pattern of American economic domination in the immediate postwar era before the war-ravaged economies of Europe and Japan had time to recover, but American domination is a thing of the past in most markets today. To maintain the imperialism hypothesis it is necessary to identify something that makes video products uniquely vulnerable to the postulated political pressures. Yet, if one looks at the outcomes of trade negotiations, precisely the opposite would seem to be the case. The U.S. has been largely successful in seeking the removal of various nations' barriers to trade in U.S. goods, and, in the most recent Uruguay round of the GATT, to a wide range of services. The U.S. has failed conspicuously, however, in removing barriers to trade in cultural products, including films and programs. The mass media are granted exceptions to the general rules for liberalization in the GATT and in regional trade accords like NAFTA. Even ignoring the United States' very visible failures to get its way in trade accords, it is hard to believe that countries such as France, that have vigorously opposed major U.S. foreign initiatives such as the Vietnam War, would knuckle under to U.S. pressure on media trade policy.

For similar reasons, it is hard to take the anti-competitive tactics seriously as an explanation for U.S. domination of video trade flows. If anti-competitive tactics work for media products, why do they not work for the vast majority of industries for which the U.S. runs a deficit? If anti-competitive practices explain U.S. surpluses in films and programs, do they also explain U.S. trade surpluses in agricultural products? Are U.S. deficits the result of anti-competitive tactics employed by other countries in other products? Few people would answer these questions in the affirmative because to do so would be to deny the generally accepted principle of comparative advantage as a force motivating international trade. Viewed from this larger vantage point, it is hard to see the anti-competitive practices explanation as anything other than ad hoc.

This is not to say that American majors have not employed anti-competitive practices. And it is conceivable that their dominant position provides the opportunity to do so, although whether this is so can only be determined by examining the particulars of specific cases. But at best this would be a consequence, not a cause, of domination.

In contrast, the larger picture of U.S. trade surpluses in films and programs being more than balanced by deficits in other industries does raise troubling questions about the plausibility of the microeconomic and the cultural explanations of media trade flows, because central to both is the assumption that films and programs differ in important respects from most other goods and services. For the microeconomic explanation it is the public-good nature of these products; for the cultural explanation it is the fact that films and programs are predominantly cultural products.


One-way flows are not unique to the international trade in films and television programs. To the contrary, they seem to be endemic to mass-media industries. The U.S. is a large net exporter of printed matter, and for recorded music the pattern of international trade is very similar to that in films (Wildman & Siwek, 1993). There are also largely one-way flows that occur entirely within national boundaries and on even smaller geographic scales. While they are seldom equated with international flows, the parallels are striking. Rosse's well-known umbrella model of newspaper competition in major U.S. metropolitan areas is a good example (Rosse, 1975; Rosse & Dertouzos, 1978). Rosse's model describes a situation in which suburban residents read the major dailies originating in the large cities at the hearts of major metropolitan areas, while centre-city residents show no interest in papers published in the suburbs. In other words, the flow of newspapers is one way from centre cities to suburbs. Furthermore, centre-city residents exhibit an indifference to suburban papers analogous to the U.S. cinema audience's disinterest in films from non-English countries.

There is also a largely one-way flow of local broadcast television signals from large cities to smaller urban areas and rural areas within the United States. Popular independent television stations in the largest cities are picked up and retransmitted by satellite as superstations to cable television subscribers around the country. The four most popular are WTBS (Atlanta), WGN (Chicago), WWOR (New York), and WPIX (New York). Nation-wide distribution adds millions of viewers to the audiences of these stations. Because there is little viewer demand for signals imported from small local television markets, there are no small market superstations.

When the same phenomenon seems to occur in a variety of situations, it is natural to look for a common explanation. Wildman (1994) has argued that the microeconomic model of trade in films and programs is a special case of a more general model of trade in media products that successfully explains these flows and others. Briefly stated, the more general model posits that imbalanced geographic flows of media products are likely to be observed when the following six conditions are satisfied:

  • The demand for a media product is determined largely by its content, which is a public good.
  • Content is costly to produce.
  • Increased spending on content increases a media product's value to consumers, but the rate of increase in appeal slows as content expenditures get larger.10
  • Geographically separate "local" media markets that differ in size.
  • Other things being equal, local producers are better able to satisfy the preferences of local viewers.
  • The ability to transport media products from one market to another.

The logic of the more general model is essentially the same as the logic of the microeconomic model of film and program flows set out above. Larger potential audiences in larger markets give producers of media products an incentive to spend more on content to turn potential audiences into actual audiences. Location within a market is an advantage to home-market producers because they are better able to gauge the tastes and preferences of the local population. However, greater expenditure on the production and procurement of content is a source of appeal to consumers that exists independent of, and, in the case of imported media products, may substitute for, attention to local tastes. Thus readers, viewers, and listeners in small markets often find media products from large markets appealing enough to compete with local product, but there is little demand in large markets for media products produced in small markets.

It is important to note that the distinctness of local markets may arise for a variety of reasons: linguistic differences for popular music, as with films and programs; separate local political jurisdictions, which create a demand for coverage of hometown politics, in the case of newspapers in major metropolitan areas in the U.S.; and the limited reach of terrestrial broadcast signals in the case of superstations.

The secondary predictions of the model seem to be borne out in each of these cases as well. Centre-city papers are larger and spend more on content than do suburban papers (e.g., more writers, larger production staffs, and more and higher-priced syndicated columnists). Superstations carry more expensive programs,11 including, in the case of active superstations, acquiring exclusive national rights to some programs, and employ larger programming staffs than do independent stations in smaller markets. While systematic data on this issue have not been compiled for popular music, it also appears that the major record companies spend much more finding, developing, funding, and promoting English-language talent than is the case for recordings in other languages. In the case of newspapers, it has been observed that the largest satellite cities in major metropolitan areas cast smaller circulation umbrellas of their own over neighbouring, but smaller, suburbs of the centre city (Owen, 1975; Rosse, 1975; Rosse & Dertouzos, 1978). The pattern is similar to that of second-tier countries in the film and program trade.

While the general version of the microeconomic model plausibly accounts for these parallels across media industries, the same cannot be said for the other three explanations, the anti-competitive practices and national imperialism explanations in particular. For example, a generalization of the national imperialism explanation would have city government officials and newspapers in virtually every major metropolitan area in the United States somehow forcing suburban readers to subscribe to centre-city papers in greater numbers than they otherwise would.12 In the case of superstations, they would have to somehow force thousands of cable operators in distant markets to take the satellite feeds for major market television stations over smaller market stations they would have preferred. In both cases, it is hard to imagine the mechanism by which the necessary leverage could be exercised on such a vast scale by so many independent players.

The evidence does support a role for the cultural explanation, although not the primary role envisioned by its proponents. It is certainly the case that the centre-city dailies that dominate metropolitan newspaper markets in the United States appeal to a broader range of reader interests than do the suburban papers they compete against. The same might also be argued for the programming of superstations compared with independent stations in smaller television markets. Some of this is just the added appeal that comes from larger expenditures on content that allows a centre-city paper, for example, to offer many more pages of international news coverage than can be found in a suburban paper. But it is also clear in the case of centre-city newspapers that there is a deliberate attempt to broaden their appeal to suburban readers by covering suburban news and, in many cases, by publishing regionally zoned editions with special sections and inserts targeted to the interests of readers in specific suburbs. Similarly, the most popular superstations acquire some programs that will not be duplicated in other television markets, and local news is replaced with other programming on the satellite signal of TBS, the most popular U.S. superstation.

This systematic association of greater breadth of appeal with larger home markets and higher expenditures on content suggests that breadth of appeal is at least in part a response to market opportunities. The more expensive media products created for large-market audiences naturally attract media consumers in other markets; but given that customer base it makes sense financially to modify content in ways that will increase it further. For the same reasons, the focus on more restricted audiences that is characteristic of media products created in smaller markets must also reflect in some degree the more limited commercial opportunities presented by those markets.

Implications for liberalization and media trade policies


What should the microeconomic explanation of one-way international flows for media products lead us to expect from trade liberalization? In answering this question it is important to distinguish between trade liberalization for media products and the liberalization of trade in goods and services generally. Cultural exceptions in the GATT and NAFTA make it clear that the two do not necessarily go together, although a strong worldwide movement toward general liberalization makes it increasingly difficult to resist pressures to relax restrictions on cultural imports.

If media trade flows reflect differences in market size and freer trade spurs economic growth as is widely anticipated, the exports of media industries in those countries (or cultural/linguistic markets) that grow most rapidly should benefit the most. With more of their economic potential yet to be realized, less-advanced economies might be expected to realize the largest gains in income. However, in many countries the extent to which this potential is realized will be determined in large part by the extent to which trade liberalization is complemented by domestic economic reforms favourable to growth. In the microeconomic model, domestic media industries and their export successes are linked to the rest of the domestic economy by media firms' responses to commercial opportunities within their home markets. The strength of these responses is determined by both the size of the domestic market and laws and regulations governing the commercial activities of media firms. All countries limit media commercialization to some degree, but historically English-language countries have permitted considerably more commercialization than have other countries, with U.S. media probably being the most commercialized. Because the various forms of public support for the mass media in other countries have not come close to matching the resources available to media firms through the marketplace in the United States and English-language countries generally, the effect has been equivalent to an artificial shrinking of other linguistic markets relative to the English-language market (Wildman & Siwek, 1988).

In the long run, the growing movement to privatize electronic media and promote their commercialization that began in Europe in the late 1980s should reduce the domestic market advantage of the United States and strengthen the media production sectors of non-English-language countries, although in the shorter term the demand for U.S. product may surge because distribution capacity is likely to expand more rapidly than video production capacity. The example of Italy, which privatized its television industry before the rest of the continent, shows that in the long run domestic producers can thrive and be more than competitive with American imports, at least in television. An initial upsurge in U.S. imports was followed by growth in the quantity and quality of domestic television programs, which eventually displaced American product during prime viewing periods (Wildman & Siwek, 1993). Similar domestic successes have been observed in Latin America (Antola & Rogers, 1984).

Implication for media trade policies

No policy prescriptions are implicit in the microeconomic model of media trade. Its value lies in contextualizing these flows and providing a framework for assessing unavoidable policy trade-offs. It makes clear that the forces producing one-way flows will be with us for the foreseeable future. It also makes it clear that these forces are impersonal, and that the tendency too often indulged in the past of attributing U.S. successes to bad acts and bad actors is a counterproductive diversion of attention from the true causes of these flows. It is important to know that limits on the commercialization of domestic media have the long-run effect of limiting international demand for their products and creating a stronger internal demand for imports. While restrictions on commercialization may serve a variety of public policy objectives, the costs of these restrictions must also be recognized.

Expanding domestic opportunities for indigenous media by relaxing restrictions on commercialization is probably the best way to ensure their vitality and further their international appeal, although import restrictions and direct financial support may also contribute to the financial well-being of domestic producers. The way in which direct support is managed is likely to matter a great deal and, again, should be handled with awareness of critical trade-offs. While financial support contributes to viability, to the extent that it supports the development of narrowly targeted domestic themes, it is likely to stifle international sales. That is, the more obviously domestic is the content, the poorer are its international prospects.


In addition to Guback (1969), see Varis (1984), Varis & Nordenstreng (1974), Waterman (1988, 1993), and Wildman & Siwek (1987, 1988, 1993) for more extensive documentation of the prominence of American productions in the international trade in films and television programs.
The insightful article by Schement, Gonzalez, Lewis, & Valencia (1984) sets out a framework for trade in television programs that has many of the elements of microeconomic models developed subsequently. The most important difference is the analysis of the public-good elements of media products developed below.
Mathematical formulations of the basic neoclassical model (Wildman & Siwek, 1988, Appendix B; Frank, 1992) do not have the all-or-nothing quality of the verbal formulations. However, beyond the claim that on average increases in budget contribute to viewer appeal and increased audience shares, the assumptions regarding the nature of video products and consumer responses to them that underlie these mathematical formulations are not well specified.
See also Guback (1969), especially chap. 7 where financial issues on which the Motion Picture Export Association of America (MPEAA) and U.S. occupation authorities in postwar Germany were not in agreement are also discussed.
See also Guback (1977).
The first estimate is reported in Wildman & Siwek (1988) and the second in Siwek & Furchtgott-Roth (1993). Variation in this estimate is due in part to differences in relative growth among different linguistic populations, but is primarily a reflection of fluctuations in the exchange rates that are used to convert purchasing power in one currency into another.
Government control is indirect when it is controlled by an independent agency created by the government.
It is important to note that advances in economic theory have shown that many practices, such as block booking, which used to be taken as prima facie evidence of anti-competitive intent, instead have strong efficiency rationales. See, for example, Kenney & Klein (1983).
With some modification, the more general model can also be extended to explain flows of media products among channels of distribution over time (Wildman, 1994). These intertemporal interchannel flows are also largely unidirectional. For example, films made for the cinema later show up on television and on video cassettes; but made-for-television or made-for-video movies do not move from these venues to the cinema. While the fact that consumption of a media product distributed in one channel may dissuade a consumer from taking it from a different channel leads to some reordering, for the most part these flows are from distribution channels reaching large audiences with substantial revenues to channels with smaller audiences and revenues.
This is just the principle of diminishing returns applied to media content expenditures.
Active superstations are distinguished from passive superstations by the fact that they promote themselves as such to cable system operators.
There have been a few anti-trust suits by surburban papers against centre-city dailies, which is consistent with the anti-competitive tactics explanation, but these have been isolated instances and the merits of these cases are far from obvious.


Antola, L., & Rogers, Everett M. (1984). Trade flows in Latin America. Communication Research, 11(2), 183-202.

Biltereyst, Daniël. (1992). Language and culture as ultimate barriers? An analysis of the circulation, consumption and popularity of fiction in small European countries. European Journal of Communication, 7, 517-540.

Frank, Björn. (1992, Spring). A note on the international dominance of the U.S. in the trade in movies and television fiction. Journal of Media Economics, 5(1), 31-38.

Guback, Thomas H. (1969). The international film industry. Bloomington, IN: Indiana University Press.

Guback, Thomas H. (1977). The international film industry. In George Gerbner (Ed.), Mass media policies in changing cultures (pp. 21-40). New York: John Wiley & Sons.

Hoskins, Colin, & Mirus, Rolf. (1988). Reasons for U.S. dominance of the international trade in television programmes. Media, Culture and Society, 10, 499-515.

Hoskins, Colin, Mirus, Rolf, & Rozeboom, William. (1989, Spring). U.S. programs in the international market: Unfair pricing? Journal of Communication, 39(2), 55-75.

Kenney, Roy W., & Klein, Benjamin. (1983, October). The economics of block booking. Journal of Law and Economics, 26(3), 491-540.

Owen, Bruce M. (1975). Economics and freedom of expression: Media structure and the first amendment. Cambridge, MA: Ballinger Publishing.

Owen, Bruce M., & Wildman, Steven S. (1992). Video economics. Cambridge, MA: Harvard University Press.

Pruvot, Deputy M.-J. (1983, July 15). On the promotion of film making in the community (Report on behalf of the Committee on Youth, Culture, Education, Information and Sport to the European Parliament, Working Document 1-504/83). Brussels: Commission of the European Communities.

Rosse, James N. (1975). Economic limits of press responsibility (Discussion paper, Studies in Industry Economics No. 56). Stanford, CA: Stanford University.

Rosse, James N., & Dertouzos, James N. (1978). Economic issues in mass communication industries. In Proceedings of the Symposium on Media Concentration (Vol. 1, pp. 40-192). Washington, DC: Federal Trade Commission.

Schement, Jorge, Gonzalez, Ibarra, Lewis, Patricia, & Valencia, Rosita. (1984, April). The international flow of television programs. Communication Research, 2, 163-182.

Schiller, Herbert I. (1969). Mass communication and American empire. New York: Kelley.

Siwek, Stephen E., & Furchtgott-Roth, Harold W. (1993). International trade in computer software. Westport, CT: Quorum Books.

Varis, Tapio, & Nordenstreng, Kaarle. (1974). Television traffic: A one-way street? Paris: UNESCO.

Varis, Tapio. (1984, Winter). The international flow of television programs. Journal of Communication, 34(1), 143-152.

Waterman, David. (1988, June). World television trade: The economic effects of privatization and new technology. Telecommunications Policy, pp. 141-151.

Waterman, David. (1993). World television trade: The economic effects of privatization and new technology. In Eli M. Noam & Joel C. Millonzi (Eds.), The international market in film and television programs (pp. 59-82). Norwood, NJ: Ablex.

Waxman, Sharon. (1993, October 3). France fears U.S. will smother its cinema. Chicago Tribune, sec. 7, p. 2.

Wildman, Steven S. (1994). One-way flows and the economics of audiencemaking. In James S. Ettema & D. Charles Whitney (Eds.), Audiencemaking: How the media create the audience (pp. 115-141). Thousand Oaks, CA: Sage.

Wildman, Steven S., & Siwek, Stephen E. (1987). The privatization of European television: Effects on international markets for programs. Columbia Journal of World Business, 22(3), 71-76.

Wildman, Steven S., & Siwek, Stephen E. (1988). International trade in films and television programs. Cambridge, MA: Ballinger.

Wildman, Steven S., & Siwek, Stephen E. (1993). The economics of trade in recorded media products in a multilingual world: Implications for national media policies. In Eli M. Noam & Joel C. Millonzi (Eds.), The international market in film and television programs (pp. 13-40). Norwood, NJ: Ablex.

  •  Announcements
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Current Issue
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Thesis Abstracts
    Atom logo
    RSS2 logo
    RSS1 logo

We wish to acknowledge the financial support of the Social Sciences and Humanities Research Council for their financial support through theAid to Scholarly Journals Program.