VoxEU Column International trade

Restrictions on trade in audiovisual services: Whom are we protecting from what?

Do cultural imports threaten domestic customs and traditions? This column explains how further liberalisation of trade in audiovisual services would indeed induce cultural change, using the example of foreign influence on names. However, these changes are generally modest, and consumers gain from the enjoyment of consuming cultural goods and a broader cultural choice set.

The presence of foreign songs, shows, and movies on domestic radios, televisions, and movie screens is one of the most salient aspects of globalisation. Trade in such cultural goods remains, however, far from free. Indeed, audiovisual services have for the most part been excluded from the progress in trade liberalisation that has occurred in other sectors. The fourth article of the GATT explicitly allows for screen quotas for minimum amounts of domestic origin cinema. Countries such as Korea, Brazil, Venezuela, Italy, and Spain, have taken advantage of this exception to institute quotas for domestic (or EU) film exhibition (Bernier, 2003).

The 1948 GATT agreement did not explicitly exempt other audiovisual services such as television and radio. They fall under the rubric of the General Agreement on Trade in Services (GATS). The GATS approach allows countries to specify a list of sectors in which they are willing to make commitments, with no commitments implied on services they omit from the list. Most countries have opted not to make commitments on audiovisual services (Bernier, 2004). The result is that domestic quotas for television and radio remain commonplace, e.g. Canada’s 35% domestic content quota on radio and France’s 60% European origin quota for broadcast television. Both Canada and the EU have exempted cultural goods from recent bilateral trade agreements.

The importance of culture

What accounts for the resistance to audiovisual service imports? One explanation would be that this sector, like agriculture, is one in which a relatively small number of domestic producers have proven themselves to be politically effective in lobbying for protection. An alternative explanation – pursued in recent applied theory papers – emphasises that audiovisual service import restrictions are directed at cultural goods. Many of these theories develop the idea that an individual’s decision to consume foreign cultural goods imposes a negative externality on other consumers. In a pioneering investigation, Francois and van Ypersele (2002) modelled the externality as being mediated by increasing returns in the production of cultural goods. The idea is that imported cultural goods might crowd out domestic alternatives, leaving some consumers (those that highly value the domestic version) worse off. More recent investigations by Janeba (2007), Rauch and Trindade (forthcoming), and Olivier et al. (2008) have built the externality between consumers directly into the utility function. The first three papers find that restrictions on trade in cultural goods can be welfare-improving. The Olivier et al. paper does not derive welfare results, but it does highlight the disutility experienced by the parent generation as their children adopt new cultures.

A 2007 survey conducted by the Pew Research Center finds some evidence of the tension in many countries over displacement of domestic customs by those imported from the US. Responses are summarised in the table below.

Response: Ratio: Canada Britain France Germany
“Our traditional way of life” Lost/Strong 3.1 4.3 3.0 3.0
“American ideas & customs are spreading here” Bad/Good 3.0 3.2 4.5 4.7
American music, movies, & TV Like/Dislike 3.8 2.3 1.9 1.8

Source: Pew Global Attitudes Project (2007), author calculations

By large ratios, participants from Canada, Britain, France, and Germany feel that their traditional way of life is being lost. By similar ratios, the survey subjects expressed concern over adverse effects of American culture (“ideas and customs”) spreading into their countries. Nevertheless, a substantial majority of these same respondents like the audiovisual services they import from the US. The empirical question raised by these survey responses is whether the foreign media that consumers enjoy have the adverse impact of spreading American customs and undermining indigenous culture. If so, then there is a kind of “tragedy of the commons,” whereby individual consumer decisions deplete a collectively valued resource. According to this way of thinking, protectionism in audiovisual services protects consumers from the externalities they impose on each other. Alternatively, it might be the case that foreign audiovisual services can be “safely” consumed without negative side effects on domestic traditions. The American ideas and customs that cause concern may be political or economic rather than cultural.

What’s in a name?

Investigations of the empirical influence of audiovisual imports on domestic culture must confront the problem of measurement. Some aspects of culture like language and religion are measured in surveys, but they change very slowly. Other aspects of culture like diet and clothing are difficult to quantify and compile. Even if they could be measured precisely, it would be very hard to identify the impact of foreign audiovisual services on the evolution of these traits. We have identified one cultural trait that is very carefully measured over an extended period of time and can be explicitly linked to foreign music, movies, and television. That trait is the frequency with which different names are given to children each year. In Disdier et al. (2009), we relate baby name shares in France to the names of characters and performers in movies, TV shows, and songs to which French parents were exposed from 1967 to 2002.

Media exposure is of course not the only determinant of the choice of name for one’s baby. We model this choice as following a process of “selective imitation.” Because of the simple desire to conform to choices of peers, or because those choices transmit information about unknown names, parents tend to prefer names that they hear often, either through social interactions (which increase with the stock of people using that name) or media exposure. More exposure is not always better. Name choice is subject to fashion, which means that parents do not want to give names that sound outdated. This suggests the impact of social and media exposure could differ between new names and those that have been in use for longer periods. Our results suggest that media exposure has an effect that is very sensitive to fashion motives. Media appearances for names that have not been used before in France deliver a 62% boost to the popularity of that name. The effect falls to 10% for the average-aged name in the population (15 years) and is nil when the name is associated with people older than 30.

Foreign media’s influence on names

Our naming model permits counterfactual simulations related to the public debate surrounding free trade in audiovisual services. We simulate the removal of foreign media exposure, a policy experiment with interesting dynamics. The removal of media exposure harms exposed names and helps the other set of names, which tend to be chosen more often. There is more that happens in the long run, since media removal reduces the stock of babies born with an exposed name, which reduces social exposure and increases the age associated with a name, complementing the static effect.

Figure 1. The influence of foreign media on two girls’ names

Our simulation reveals that at most 5% of the babies born in France over the 1967–2002 period would have had a different name if no foreign media had entered the country. This modest overall effect hides some large and complex impacts for particular names. Figures 1 and 2 take two example names for each sex, Jennifer and Margot for girls and Jason and Sébastien for boys. The figures represent media exposures using red circles with the number below stating the count of appearances in the foreign media each year. Neither Margot nor Sébastien appeared in our foreign media exposure data, whereas Jennifer was heavily exposed starting in 1971 and Jason experienced two very distinct exposure periods. The lines represent the ratio of real to simulated (under the counterfactual of no foreign media exposure) births for each name over time, which hence gives the impact of media exposure.

Figure 2. The influence of foreign media on two boys’ names

In both graphs, we can see the strong and lasting effects of media exposure on certain names. The Jason case is particularly interesting. The name was actually not present in the official statistics before 1969, a year when it also appeared in foreign media (actor Jason Robards in Once Upon a Time in the West). During those early years, the exposure increased the share of Jasons born in the French population by a large amount (more than 50%). There is then a decade of non-exposure of that name, but the ratio of true to simulated births does not fall to one immediately, due to the persistent influence of previous exposures via increased stocks and reduced age. In the nineties, when Jason reappeared in foreign media (most prominently as the lead actor in Beverly Hills 90210), the media impact ratio goes up again quite strongly.

Jennifer is also interesting since it illustrates how much stronger the impact of media is for young names. In the early seventies, the ratio increases quickly with small amounts of exposure, while it peaks and even falls back in the nineties despite much larger exposure.

By the 1990s, Jennifer was not a name associated with children, which leads to a muted impact for media exposure. Margot and Sébastien have similar declining profiles.
Would free cultural trade destroy traditions?

Our results suggest that further liberalisation of trade in audiovisual services would indeed induce cultural change. The substantial stimulus that we estimate for names like Jennifer and Jason helps us make sense of the concerns expressed in surveys over foreign cultural invasion. The simulations suggest that such cases are exceptional and that the changes from audiovisual trade liberalisation would be modest for most names and would almost certainly not involve wholesale abandonment of domestic traditions. Furthermore, increased audiovisual imports benefit consumers not only through the direct enjoyment of the cultural good (song, movie, or show), but also (in our model) via the enhancement of the cultural choice set.


Bala, Venkatesh and Ngo Van Long, 2005, “International trade and cultural diversity with preference selection,” European Journal of Political Economy 21, 143–162.

Bernier, Ivan, 2003, “Local Content Requirements For Film, Radio, And Television As A Means Of Protecting Cultural Diversity: Theory And Reality

Bernier, Ivan, 2004, “The Recent Free Trade Agreements of the United States as Illustration of Their New Strategy Regarding the Audiovisual Sector

Disdier, A.C., K. Head, and T. Mayer, 2009, “Exposure to foreign media and changes in cultural traits: Evidence from naming patterns in France,” revised version of CEPR DP 5674.

Francois, Patrick and Tanguy van Ypersele, 2002, “On the protection of cultural goods,” Journal of International Economics 56, 359369.

Janeba, Eckhard, 2007, “International Trade and Consumption Network ExternalitiesEuropean Economic Review 51(4), 781–803.

Olivier, J., M. Thoenig, T. Verdier, 2008, “Globalization and the dynamics of cultural identity,” Journal of International Economics, 76(2):356–370.

Pew Global Attitudes Project, 2007, “World Publics Welcome Global Trade – But Not Immigration.”

Rauch, James E. and Vitor Trindade, forthcoming, “Neckties in The Tropics: A Model of International Trade and Cultural DiversityCanadian Journal of Economics.

United States Trade Representative, 2008, “National Trade Estimate Report on Foreign Trade Barriers—European Union”

42,204 Reads