Virtual proximity and audiovisual services trade

Abstract

Audiovisual services such as music and movies in digital formats have gained substantial importance over the last decade. This paper analyses audiovisual services in a gravity model framework. In particular, we explore the role of virtual proximity – a new proxy for cultural proximity based on bilateral hyperlinks and bilateral website visits between countries – and find that ‘virtually-proximate’ countries trade significantly larger amounts of audiovisual services. Our results show that virtual proximity also has a larger impact on trade in audiovisual services than on total services trade. Moreover, in line with Hanson and Xiang (2011), our analysis indicates that in the audiovisual services sector, global fixed export costs dominate bilateral fixed export costs for most countries in our sample.

Introduction

Audiovisual and related services are a component of personal, cultural and recreational services in international services trade statistics.¹ These pertain to movies, music, radio and television as well as services relating to performing arts. They can be in various electronic formats (e.g. mp3 files) and include licenses for broadcasting purposes (e.g. to show the latest James Bond movie in a local cinema) and live performances by artists (e.g. Beyonce Knowles performing in London is an audiovisual services import to the United Kingdom from the United States).²

Cultural products have largely been traded in the form of goods, but due to technological advancements over the last decade, they increasingly appear in the services sector. This is reflected by decreasing sales of compact discs and other physical products in lieu of electronic formats such as mp3 files. In the United States, digital album sales accounted for 37% of all album purchases in 2012 compared to 5.5% in 2006 (according to the Nielsen Company & Billboard׳s 2012 Music Industry Report). Over time, services have continuously replaced goods, with sale volumes of services increasing and customers gaining ever more control and choice over their consumption patterns, for example by means of iTunes, Netflix or pay-per-view channels, making the audiovisual services sector one with the most dynamic growth in recent years.

This paper analyses audiovisual services trade in a gravity model framework. Our particular interest is in the role of cultural proximity, for which we propose new measures that reflect the virtual proximity of two trading partners. One of these novel proximity measures captures bilateral hyperlinks on webpages as an indicator of countries׳ virtual linkages with each other. The idea is to reflect, for instance, how often British or French internet users set links to websites from the United States (say the homepage of the New York Times). In addition, we construct an alternative indicator of virtual proximity based on bilateral website visits using data from GoogleAd Planner. In case these indicators are, ceteris paribus, higher for the United Kingdom than for France, we interpret this as British citizens being culturally closer to the United States than the French. The underlying assumption is that cultural proximity of two countries increases with the interest shown in each others׳ web content. In terms of the empirical specification, the virtual proximity measure is either used in lagged form in an otherwise standard gravity model, or we employ an instrumental variable technique to account for potential endogeneity issues and reverse causality.

Virtual proximity is a bilateral, bidirectional and potentially asymmetric indicator of revealed preferences. A key advantage of using ‘virtual proximity’ as a measure of integration is that internet activities are relatively costless (in particular as they have zero variable trade costs), i.e. with few usage barriers in light of high and rising global internet penetration rates. Given the growing importance of communication and information technology, and the sharp increase of its influence on almost all transactions – be it in a business or private setting – we propose this measure as an up-to-date alternative to traditionally used measures of cultural proximity. As such, virtual proximity might just be an updated reflection of traditional cultural and societal features, but it could also reveal additional information, for example induced by the almost unrestricted access and anonymity provided by the world wide web. In related research, Freund and Weinhold (2002) find that internet development in its partner countries has resulted in increased exports of services to the United States. Choi (2010) reports that a doubling of internet usage in a country leads to a 2–4% increase in services trade.

Despite the fact that audiovisual services only accounted for about 1% of global services trade in 2009, it is an important example of the increasing role of the services industry in the economy. Moreover, audiovisual services are contentiously discussed in international trade negotiations, since music, movies and television shows play a key role in transmitting culture and values across the globe. We consider the audiovisual sector to be a suitable candidate for testing our new virtual proximity indicator as the sector׳s very nature is closely linked to a country׳s cultural identity. Thus far, the most commonly used measures of cultural proximity in gravity trade models have been common language and common religion. Melitz (2008) and Melitz and Toubal (2014) compile and analyse an extensive dataset on the importance of language in trade which greatly expands on previously existing measures. They find that countries of the same linguistic roots tend to be closer trading partners. A proxy that bears similarity to our virtual proximity measure, but is only available for European countries, was introduced by Felbermayr and Toubal (2010) in the form of bilateral scores in the Eurovision song contest; this is also an annual, bilateral and bidirectional measure of revealed preferences.

Our theoretical framework comprises two models, based on Felbermayr and Toubal (2010) and Hanson and Xiang (2011). The former model yields our baseline gravity model equation, while the latter allows for gauging the relative importance of global fixed costs and bilateral fixed costs in audiovisual services trade.

Research interest in services trade has generally grown over past years, concomitant with its rising importance in the global economy. Services span a wide range of economic activities, are very heterogeneous and due to their intangible characteristics inherently more difficult to define and measure than goods. Kimura and Lee (2006) find that services trade is better predicted by gravity equations than goods trade. Moreover, recent papers by Francois and Manchin (2011) and Johnson and Noguera (2012) report that the role of the services sector in international trade is much larger in value added terms than suggested by gross trade flow data.

Our paper also ties in with the literature on trade in cultural goods and services. Marvasti and Canterbery (2005) find in their analysis of US motion picture exports that common language, religion and a higher education level exert a positive impact on bilateral exports from the US. Hanson and Xiang (2009) confirm the importance of market size and language for bilateral US movie exports and report that countries with more trade barriers import fewer US movies. Hanson and Xiang (2011) propose an adaptation of the Melitz (2003) model allowing fixed trade costs to be broken down into global and bilateral fixed export costs, and test this model empirically by analysing imports of US movies relative to domestic production sales. Ferreira et al. (2012) develop a model of the global movie industry and are able to quantify the welfare consequences of foreign trade in movies. For example, they find that EU subsidies mostly benefit EU producers, indicating that these function as a trade, rather than cultural policy. Disdier et al. (2010) estimate determinants of trade in cultural goods and use bilateral trade in cultural goods as a measure of cultural proximity. Ferreira and Waldfogel (2013) analyse global music trade and find an increasing bias toward consumption of domestic music which is not negatively affected by new communication channels – such as local MTV stations and higher internet penetration rates. Trade policies, like radio airplay quotas, are found to be effective in that they increase local music consumption at the expense of foreign music. On the impact of trade in cultural services and national culture, Disdier et al. (2010) find a significant link between non-national baby names and the prominence of foreign television shows.

The remainder of the paper proceeds as follows: in Section 2, we present the empirical gravity model and its theoretical foundation. Section 3 introduces the data, while the empirical results are shown in Section 4. Section 5 concludes.