Film Financing in the Digital Single Market: Challenges to Territoriality

This article discusses the role of territorial licences for feature films against the background of judicial and market developments in the EU. Currently, territorial licences are deemed a cornerstone of the exploitation and financing of films in Europe. However, current models of film financing are under increasing pressure both from market developments such as the turbulent growth of global online video platforms, and from developments in EU law aimed at removing national territorial barriers to the Single Market. Examples are the rule of Union-wide exhaustion of the distribution right, the EU Portability Regulation and the country of origin rules for satellite broadcasting and online simulcasting. EU competition law sets additional limits to grants of territorial exclusivity, and prohibits clauses in broadcasting and pay television licences that prevent or restrict “passive” sales to consumers/viewers in non-licensed territories. The freedom of right holders to preserve territorial exclusivity by way of contract is likely to become increasingly vulnerable to EU competition law, as underlying territorial rights no longer support territorial grants. For the film sector where territorial exclusivity remains indispensable, the European Commission could create specific competition law rules in the form of “block exemptions”. Language exclusivity – i.e. exclusive grants of rights for distinct language versions of a film – could provide a practical and legally more robust alternative to territorial licensing.


Introduction
The digital revolution of recent decades, which has brought affordable broadband access to consumers in large parts of Europe, is having a profound impact on audiovisual markets in the European Union. While cinemas, terrestrial and satellite broadcasting and cable distribution remain important channels of film exploitation, online services and platforms have rapidly captured significant market share, with rental and sales of DVDs concomitantly decreasing. Arguably, the future of film distribution lies largely in the digital realm.
Since the internet knows no natural borders, digitisation has enabled new global players to enter European markets, and has put existing business practices and models for distributing and financing films under pressure. EU policies aimed at removing legal obstacles to achieving the Digital Single Market exacerbate this development. In the EU, the exclusivity that a copyright confers upon its owner is, in principle, limited to the Member State where the right has been granted. This allows right holders and licensees to partition markets along national borderlines, and to structure their financing model accordingly. However, since this may pose obstacles to intra-European trade and services, from the 1990s onwards a variety of European policies and regulatory instruments have been deployed to reconcile territoriality with the evolving needs of the Internal or Single Market. Such EU policies might affect current film financing practices that are considered by many stakeholders to be vital to the European audiovisual industry.
This raises the central questions of this article: What exactly is the role that territoriality plays in the complex puzzle of financing feature films in Europe today? How do judicial and legislative developments in EU law aimed at removing national barriers to the Single Market affect film financing models based on territorial exclusivity? And, if so desired, how could the model of territorial exploitation of films be reconciled with these developments?
The structure of this article is as follows: Sect. 2 briefly describes basic issues in film financing and the general role of territorial licencing, after which Sect. 3 analyses how these issues are aggravated in Europe. Section 4 examines the role of territoriality in copyright and the challenges it faces as a result of developments in EU legislation and jurisprudence in recent decades. Next, Sect. 5 introduces language exclusivity as a potentially more robust alternative territorial exclusivity. Section 6 concludes.

Film Financing and Territorial Licencing
Films are experience goods: a consumer will only know if he or she really appreciates a film after having watched it. 1 Quality signals help consumers to cope with this uncertainty. Reputations of famous actors, directors and studios are one such quality signal. Sequels can also be understood with this background in mind: they have a cast, style and storyline that is similar to a preceding film and aim to benefit from its success. Second, trailers are a way to allow consumers to sample a film before deciding whether to watch it. Third, recognition, in the form of recommendations, reviews and awards can reduce consumers' ex ante uncertainty. They can be self-reinforcing and lead to positive consumption externalities, as everyone wants to see the film that everyone is talking about. This fuels the ''nobody knows principle'': the observation that demand for creative works is intrinsically uncertain beforehand and not even easily understood afterwards. 2 It helps explain why the marketing budgets for commercial films may amount to more than half of the film's production costs. 3 But even expensive marketing campaigns are no guarantee of success.
Film production requires relatively high upfront investments, as the whole process of production, marketing and rights clearance (e.g. in case a film is based on an existing novel) must be financed during the process of creation, which can last several years. Budgets can reach tens of millions of euros and financiers usually demand that budgets are secured before production starts. High fixed ex ante costs and low marginal costs of serving additional customers imply significant economies of scale: commercially, films benefit from large markets, as the average costs per viewer drop rapidly with market size. For an international blockbuster, even production costs exceeding a hundred million euros can be easily regained.
High ex ante demand uncertainty and high upfront investments set the stage for the core challenge of financing a film: commercial investors avoid downward risk and require higher expected returns on riskier investments. The distribution rights of a film may serve as collateral for investors, but the value of these rights is as uncertain as the success of the film itself. Recommendations, reviews, awards and trailers are not available until the costs of production have been incurred. Therefore, they are useless for financing a film ex ante and investors have to rely on the reputation of so-called ''bankable names'' -a famous director, a stellar cast or other renowned parties involved. Having secured reputable names greatly facilitates film financing. Alternatively, investors may rely on the judgment of other investors and funding bodies by requiring their involvement before committing themselves.
In order to secure funding for a film, presale agreements (''presales'') and minimum guarantees are common if not indispensable. Presales are contracts between producers and distributors whereby the distributor promises to pay the producer a fixed advance on expected revenues. The advance, called a minimum guarantee, is paid upon the film's completion. In exchange, the distributor gains the right to sell or exhibit the film in a given territory and/or distribution channel. The contracting party may also be an agent, who resells the distribution rights. Presales can count for anything from a small percentage to more than half of the budget of a film and may form the collateral for a production loan from a bank. 4 2 Caves (2000). 3 E.g. https://www.reuters.com/article/us-industry/how-much-does-movie-marketing-matter-idUSTRE65 A13Q20100611. 4 FIAD (2017), p. 6. Usually, cinemas, broadcasters, pay TV and video on demand (VOD) channels demand exclusive rights for their distribution channel to reduce the risk that competitors undermine the value of their rights. For the same reason, licence agreements typically grant exclusive rights for certain timeframes, called ''windows'', to minimise cannibalisation of earlier higher-value windows by later lowervalue windows. Broadcasters and cinemas in Europe generally operate nationally and are hence predominantly interested in exclusive rights within their own territory. Their interest in rights outside their market is limited at best, and it will be more lucrative for the producer to sell such rights on a territory-by-territory basis to others.
Territorial exclusivity may also help distributors to reap the benefits of their marketing expenses. It prevents competitors from freeriding on such efforts, which would lead to underinvestment in marketing. 5 A third reason for a producer to prefer selling rights on a territorial basis is the possibility to engage in price discrimination between high-income and low-income countries. 6 Note that territorial licensing in one exploitation window tends to necessitate territorial licensing in other windows, as long as these are not synchronised across countries. For example, a license holder for transactional video on demand (TVOD) exploitation in country A, traditionally an early exploitation window, will wish to prevent competition from subscription video on demand (SVOD) exploitation or free-to-air broadcasting in country B, where a film may have been released earlier.
Through exclusive windows and territorial licences, distributors can be certain that no-one else may distribute the film in the territory, on the platform or during the timeframe agreed upon. However, such territorial exclusivity is increasingly at odds with EU law and policies, as will be discussed in Sect. 4. It is also in discord with market forces and consumers' expectations and enhances the risk that consumers will revert to online piracy to gain access to a film.
First, film consumption is shifting towards platforms that operate globally. Historically, cinema has been the most commercially attractive window, followed by the sale and rental of physical carriers such as DVDs, and broadcasting. This has rapidly changed as the DVD market collapsed to make way for TVOD, and particularly SVOD services such as Netflix and HBO. While SVOD only accounted for 7% of total pay-service revenues in Europe in 2016, it generated 60% of revenue growth. SVOD subscribership in Europe increased by 55% to 37.7 million euros in that year. 7 Unlike cinemas and DVD stores, these digital platforms operate globally and prefer global or pan-European licence agreements.
Second, territories do not perfectly match with homogenous consumer groups or cultural identities. On 1 January 2018, 22.3 million EU residents had citizenship of a non-EU country (4.4% of the EU-28 population), while 17.6 million EU citizens lived in another Member State than that of their citizenship. 8 Also, tastes differ within and between populations and cultural groups. All this implies that territorial exploitation will leave a significant share of the potential market unserved.

European Films
European films are much less successful in attracting large audiences than films produced in the United States. In 2017, European films accounted for 27.5% of EU cinema admissions; films produced in Europe with incoming US investment for another 3.7%. Between 2012 and 2017, this European share in EU cinema admissions was 27-34%, while US films accounted for 63-70% and the rest of the world for 3%. 9 On television, European films accounted for 28% of film broadcasts by 131 TV channels in 18 EU countries in the 2015-2016 season. 10 In a selection of TVOD catalogues representing 47 country catalogues, the share of European films ranged from 17 to 30%. In 37 country catalogues for nine SVOD services, this was 20% on average. 11 The modest market share of European films in Europe is not for want of European productions. In 2017, 1676 feature films were produced in the EU, against 821 in the US. 12 In other words: in their home market European films have half the market share of US films, even though twice as many European films are produced. This implies that in the EU, US films average four times as many admissions as domestic films (not taking into account that not all US films reach the EU market).
One explanation for this relatively weak position of European films are the language barriers and cultural differences within Europe. The EU has 24 official languages, almost as many as it has Member States, and EU citizens predominantly watch films spoken in their native language or in English. In 2016, European films generated 60% of their admissions in their domestic markets, while many do not even circulate outside their domestic market. They generated half of the remaining admissions in the rest of the EU and only 19% outside Europe. About 5200 European films were on release in at least one European market in 2016, while only 650 were screened outside Europe: a ratio of eight-to-one. Co-productions between European countries perform much better in this respect. They generate three times as many admissions as purely national European films and circulate almost twice as widely. 13 As a result of their modest position in terms of audience size, most European films do not benefit from economies of scale as much as American blockbusters do: fixed costs can be spread over fewer admissions, raising the average costs per admission. This makes it harder for such films to recover their production costs and further enhances the financial risks for investors. 14 To recoup investments from a smaller customer base, production budgets need to be smaller for a film to remain commercially sound. In many genres, smaller production budgets make it harder to compete with high-budget films. As a result, ''European films'' are often synonymous with ''arthouse films'', having an intrinsically smaller audience and relying more heavily on public funding.
Contracting a stellar cast and other bankable names is much more difficult for films with a smaller budget. As these ex ante quality signals for consumers and investors are out of reach, financing such films becomes even more difficult. More than blockbusters, the typical European film has to rely on reviews, recommendations and awards, which are not available until after their release. Even larger European productions often lack the marketing power and budget to start a pan-European marketing campaign. They have to gain a reputation outside their producing countries after winning awards or nominations at festivals or through success in their home markets. Acquiring a reputation in this way may take over a year from release, and is often a prerequisite for films to start circulating abroad. Unlike large blockbusters, for which theatrical releases have become ever more synchronised across territories, 15 smaller yet successful European productions must expect a significant time span during which territorial exploitation windows are outof-sync. Not being able to grant territorial licences during this period could increase risks of cannibalisation and erode distributors' investments in licences and the opportunities to recoup such investments. Consequently, European productions are typically more dependent on territorial licences for international success than American blockbusters.
A free-market outcome that would result from a lack of scale for the average European film would be fewer productions. However, this may be undesirable for a variety of reasons. From a cultural and political perspective, films are seen as carriers of the cultural identity of Europe and its Member States or regions, which deserve protection and promotion. 16 From an economic perspective, a healthy film industry is considered important for the jobs, incoming investment and tax revenues it generates. National film industries have stressed this repeatedly. 17 In addition, successful films may attract expenditure by tourists -as Lord of the Rings did for New Zealand 18 and Harry Potter for London. These cultural, political and economic arguments have resulted in a wide variety of film support policies in Europe, which have evolved into significant film financing sources and have contributed to the fact that the EU produces twice as many feature films as the US. 14 This is illustrated by the observation that only 6.6% of UK independent films produced between 2003 and 2011 were profitable, Oxera and O&O (2016), p. 3. 15 To illustrate this: the average time between the US and UK cinema release for the top 100 films dropped from around a 100 days up to the year 2000, to just 10 days in 2016. See https://stephenfollows. com/changing-movie-release-patterns. 16  In the end, financing films in Europe is a highly complex business. Most European films, in particular larger European co-productions, typically obtain their production budget from 20 or more financing sources. Usually, this includes grants and/or loans in each of the co-producing countries, incentive schemes in the coproducing countries, support from European funds such as Eurimages and, where applicable, international co-production funds. Presales and minimum guarantees from exclusive deals, or co-production agreements with public or private broadcasters and other distributors supplement these receipts from support schemes. Also, investments are often made by the producers themselves and by other private investors. 19 Territorial exclusivity allows distributors and public and private broadcasters (often acting as co-producers) to invest in and advertise theatrical releases or broadcasts specifically geared towards local audience markets, without having to fear competition from concurrent offers of the same film. Larger film productions are often the product of wide-ranging trans-European co-production agreements involving multiple distributors and/or broadcasters, whereby distribution or broadcasting rights are ''split up'' following territorial boundaries. Most publicly financed film funds also operate at the Member States level, and sometimes demand that national broadcasters or distributors secure rights in subsidised films. Private investors likewise often require transfers of territorially defined rights as collateral for film production loans. In sum, territorially defined rights are deeply entrenched in European film exploitation and financing.

Territoriality in Copyright and the Challenges It Faces in the EU
Section 3 described how territorial exclusivity is a key factor in European film financing. Since territorial exclusivity relies on copyright law, the following section first describes the role and meaning of territoriality in copyright law, and the way EU law has in the past tried to reconcile territoriality with the needs of the Internal Market (Sect. 4.1). Thereafter, Sect. 4.2 discusses recent and future regulatory challenges to territoriality in copyright, after which limits to territorial grants of rights set by EU competition law are examined in Sect. 4.3. Section 4.4 queries to what extent film licensing contracts may still include territorial clauses despite regulatory intervention.

Territoriality and the Internal Market
Copyright creates exclusive rights in works of literature, science and art. Despite the existence of the Berne Convention and other international treaties, and nearly 30 years of harmonisation in the European Union, copyright has remained essentially national law, with each of the Union's Member States having its own national law on copyright and neighbouring (i.e. related) rights. As a consequence, copyright owners in the EU are protected not by a single European copyright law, but by a ''bundle'' of parallel national copyrights. A film will, for example, be protected in Germany under the rules of the German Urheberrechtsgesetz, whereas the same film will be protected in France under the Code de la propriété intellectuelle. Since copyright's exclusive rights are generally transferable or licensable, either in whole or in part, a direct consequence of territoriality is that rights in a film can be split into multiple territorially defined national rights, which may be owned or exercised for each national territory by a different entity. This explains why different film distributors may concurrently own exclusive rights to the same film for different Member States, for example Germany and France.
Obviously, the grant or exercise of territorially defined rights creates a potential conflict with the norms and ambitions of the Internal Market or Single Market. The inherent tension between territorially limited grants of rights and the principles of the Internal Market are particularly apparent in the practice of ''geo-blocking'' access to audiovisual services from Member States where rights have not been cleared. Since the exercise of territorial rights poses obstacles to intra-European trade and services, from the 1990s onwards a variety of European policies and regulatory instruments have been deployed to reconcile territoriality with the evolving needs of the Internal Market.
For example, the rule of exhaustion of the distribution right, also known as ''first sale doctrine'', which was initially developed by the Court of Justice and later codified in the Information Society Directive, 20 has removed undue market fragmentation by allowing tangible goods incorporating copyright works to be resold without permission of the copyright owner after their initial authorised sale in a Member State. As a consequence, DVDs that have been sold in a Member State under licence from a local right holder may be parallel imported into another Member State without additional permission of the right holders.
Another noteworthy measure, particularly designed for the audiovisual industry, is the Satellite and Cable Directive 21 that determines that satellite broadcasting is a relevant act for copyright purposes only in the country of origin of the broadcast signal. 22 A satellite-broadcasting organisation will therefore need to acquire licences only from right holders in the Member State of origin of the signal. However, the Directive does not rule out licence fees and other contractual conditions that consider the size of the footprint (i.e. number of countries reached) of the satellite broadcast. On the contrary, Recital 17 instructs the parties concerned to ''take account of all aspects of the broadcast, such as the actual audience, the potential audience and the language version''.

Recent Regulatory Challenges to Territoriality
In the online environment, the European Commission has identified ''unjustified'' geo-blocking and other forms of geographical discrimination as an obstacle to attaining the Digital Single Market in multiple policy documents. 23 The European Parliament has also expressed its concern about these practices repeatedly. 24 A European Parliament study published in 2013 25 distinguishes two types of geographical discrimination: ''geo-blocking'' -i.e. refusal to sell or provide access 26 -and ''geo-relocation'' or ''geo-filtering'' -i.e. conditioning of sales or rerouting of services. Both practices are based on the geographical location of the consumer. In the area of audiovisual services, both types of geographical restriction occur. 27 With the Portability Regulation 28 that was adopted in 2017, the EU has effectively put an end to the practice of geo-relocation in the audiovisual realm. The Portability Regulation ensures that European consumers travelling across the EU will have continued access to the online content services to which they have subscribed in their home countries. The Regulation obligates operators of subscription-based online audiovisual services such as Netflix to provide migrant subscribers access to the content the catalogues of their country of residence, whenever travelling in Europe. Since the making available to a travelling subscriber of movies, for which the rights have not been licensed in the country where the subscriber is temporarily located, would normally amount to copyright infringement, the Regulation provides that the provision of the service ''shall be deemed to occur solely in the subscriber's Member State of residence''. The Regulation thus allows Netflix and other audiovisual content providers to offer content to their subscribers, based solely on licences secured in their country of residence.
In 2018, EU Regulation 2018/302 was adopted, which prohibits ''unjustified geoblocking''. 29 However, ''audiovisual services, including services the principle purpose of which is the provision of access to broadcasts of sports events and which are provided on the basis of exclusive territorial licenses'', are excluded from the scope of this Regulation, so the Regulation is of limited consequence for the film industry. 23 European Commission (2014) Gomez and Martens (2015). The recently adopted Online Broadcasting Directive 30 poses another challenge to territorial exclusivity in the audiovisual realm. The Directive extends the Satellite and Cable Directive's country of origin rule to radio and television broadcasts offered by broadcasting entities over the internet simultaneously with their terrestrial broadcasts, and to other ''ancillary online services'' such as catch-up platforms. Under current law, broadcasters wishing to simulcast their broadcasts over the internet require licences from all relevant right holders in all countries where the broadcasts are made available. The Impact Assessment study preceding the Commission's proposal for the Online Broadcasting Directive identified the ''territoriality of copyright'' as a major obstacle to rights clearance, 31 as a result of which the number of rights to be cleared is multiplied.
Under the rules of the new Directive, broadcasters wishing to simulcast their broadcasts over the internet across the EU require licences only in their country of establishment. The Directive has been criticised, notably by right holders, as unduly affecting exclusive territorial grants of rights and thereby undermining customary film financing practices. In response to this criticism, the scope of the Directive's country of origin rule was narrowed down in the course of the discussions in the European Parliament. As finally adopted, the new rule applies only in respect of radio programmes, televised news and current affairs programmes, and ''fully financed own productions of the broadcasting organisation''.
The ultimate challenge to territoriality in copyright lies in the more distant future. Whereas the EU has employed a variety of regulatory instruments, such as the exhaustion rule and the country of origin principle, to overcome some of the obstacles to the Internal Market that territorially exercised rights entail, the final step towards a truly unified European copyright law has yet to be taken. One day, such unification might be shaped as a general EU Copyright Regulation that would provide for unitary copyright protection across the Union, replacing existing national copyright laws. 32 In recent years, the idea of unification is gradually gaining support, both in scholarly and in political circles. For example, the European Copyright Society, a society of professors, has sent a public letter to the European Commission urging it to start a unification project. 33 A year later, the European Commission committed itself to the idea of unitary copyright in an official Communication to the Council and the European Parliament. 34 In doing so, the Commission recognised that the road to a European Copyright Regulation is still a long and complex one. According to the Commission, the introduction of unitary copyright should be accompanied by the designation of an exclusively competent Union court ''to avoid inconsistent 30 Directive (EU) 2019/789 of the European Parliament and of the Council of 17 April 2019 laying down rules on the exercise of copyright and related rights applicable to certain online transmissions of broadcasting organisations and retransmissions of television and radio programmes, and amending Council Directive 93/83/EEC. 31 European Commission (2016), p. 14. 32 Article 118 TFEU has created an express EU competence for such unification. 33 European Copyright Society (2014). 34 European Commission (2015a). jurisprudence leading to more fragmentation''. But ''these difficulties should not lead to the abandonment of this vision as a long-term objective''. If adopted, a truly unified EU copyright law would do away with nationally defined territorial rights, since national copyright laws would cease to exist. Instead, a unified EU copyright law would apply homogenously in the entire territory of the Union.

Limits to Territorial Grants of Rights by EU Competition Law
Apart from specific regulatory intervention, territorial grants of rights might also run afoul of the general rules of EU competition law, notably Arts. 101 (anti-trust) and 102 (abuse of dominance) of the Treaty on the Functioning of the European Union (TFEU) -formerly Arts. 81 and 82 of the EC Treaty. Over the years, the European Commission, which is tasked with competition oversight, and the European Court of Justice have produced extensive case law on the issue. In a string of cases concerning the interface between intellectual property law, competition law and the EC Treaty's Internal Market freedoms, the Court has developed the doctrine that the application of competition law and the economic freedoms should not affect the ''specific subject matter'' of the intellectual property right concerned. 35 This doctrine has led the Court in the past to accept certain contractual restrictions to competition and to the freedom of circulation of goods and services.
In the landmark case of Coditel II, which was decided by the Court in 1982, the Court held that an exclusive territorial grant of copyright in respect of a film does not per se amount to infringement of anti-trust law, but that such a contract may well violate competition law ''if it has as its object or effect the restriction of film distribution or the distortion of competition on the cinematographic market''. 36 In its decision, the Court expressly took account of the specific economic characteristics of the film industry in Europe: The characteristics of the cinematographic industry and of its markets in the Community, especially those relating to dubbing and subtitling for the benefit of different language groups, to the possibilities of television broadcasts, and to the system of financing cinematographic production in Europe serve to show that an exclusive exhibition licence is not, in itself, such as to prevent, restrict or distort competition. (para. 16) Whether or not a territorial grant of rights amounts to an infringement of antitrust law must therefore be assessed in light of the ''specific characteristics'' of the cinematographic market.
In its more recent Premier League decision of 2011, the European Court of Justice held that an exclusive pay television licence agreement that obliges the pay television broadcaster not to supply decoding devices to users outside the territory 35 Triaille (2013) covered by that licence agreement infringes Art. 101 TFEU. 37 This decision is in line with general EU competition law, 38 which distinguishes between ''active'' and ''passive'' sales to consumers in Member States not covered by a territorial licence. Whereas competition law does allow a licensor to oblige a licensee not to actively seek customers outside the licensed territory, a licensee may not be prevented from ''passively'' selling to such consumers. In other words, a territorial broadcasting licence will never be absolute.
Following the Premier League decision, commentators have questioned whether the Court's reasoning would also apply to the film sector. 39 Whereas the economics of broadcasting sports and films are clearly not identical, the Court's decision in Premier League has inspired the European Commission to start investigations into possibly anti-competitive licensing practices involving films licensed to pay television platforms. The investigation focuses on licensing contracts involving a number of major film studios and Sky UK Limited (and subsidiaries). So far, this has led to commitments by Paramount and Disney to remove or no longer enforce restrictions in their licensing contracts that prevent pay television providers from responding to ''passive'' requests from consumers outside the licensed territory. 40 If similar restrictions were to be imposed on film producers with regard to online audiovisual streaming services, the implication could be that they could no longer oblige their licensees to ''geo-block'' consumers residing outside the licensed territory.

Freedom of Contract and Territorial Exclusivity
The gradual abolition of state-based territoriality in EU copyright law does not necessarily spell the end of exclusive grants of rights with territorial effect. Even if territoriality can no longer be fully based in national copyright law, freedom of contract, i.e. the freedom of parties to govern their own actions by contract, persists. Based on freedom of contract, a film producer and a distributor might still agree on a territorially restricted exclusive licence. For example, the introduction of the country of origin rule in the Satellite and Cable Directive has not put an end to the practice of territorially restricted licensing in the field of satellite broadcasting. The Directive does not prevent the owners of copyright in a film from contractually obliging a satellite broadcaster to apply encryption or other technical means to avoid reception in countries for which the broadcast is not intended. Thus, territorial exclusivity is still achieved, notwithstanding the clear aim of the Directive to create an internal market for trans-frontier satellite broadcasting. 41 This issue has re-emerged during the discussions surrounding the Online Broadcasting Directive. For this reason, Art. 3(3) of the Directive provides that ''the country of origin principle set out in paragraph 1 shall be without prejudice to the contractual freedom of the rightholders and broadcasting organisations to agree, in compliance with Union law, to limit the exploitation of such rights […]''. And Recital 10 of the Directive states: ''The country of origin principle should not affect the freedom of rightholders and broadcasting organisations to agree, in compliance with Union law, on limitations, including territorial limitations, to the exploitation of their rights.'' While these texts clearly express the EU legislator's desire to preserve the practice of territorial licensing in the audiovisual realm 42 notwithstanding the introduction of a limited country of origin rule, both Art. 3 and the accompanying recital are conditioned on the assumption that contractual grants of territorial exclusivity ''are in compliance with Union and national law''. What is meant here is that such contractual clauses are not immune to the application of EU competition law. Territorial exclusivity in licensing agreements must still comply with the general rules of Arts. 101 and 102 TFEU, as applied by the Commission and interpreted by the CJEU.
The question therefore remains whether exclusive territorial grants of rights will stand the test of competition law in cases where the EU legislator has diminished or even abolished the underlying territorial copyright. Since relevant case law concerns situations where territorial grants are supported by territorial rights -see the Coditel II and Premier League cases discussed above -it is hard to predict how the European Court would assess a case where territorial exclusivity would be solely based on contract.
In its Coditel II decision, the European Court deemed a territorial grant of copyright potentially justified by reference to Art. 36 of the EC Treaty (now Art. 36 TFEU). Under this provision, restrictions to trade may be justified on grounds of ''the protection of industrial and commercial property'', a term which includes copyright. Whether the Court would accept purely contractual territorial exclusivity in a situation where the EU legislator has purposefully removed -in whole or in part -the underlying territorial right seems questionable, particularly where such exclusivity would keep intact the Internal Market fragmentation that the EU legislator wanted to remove.
It follows that a more reliable way of preserving territorial grants of rights in the audiovisual field would be to task the European Commission with codifying filmspecific rules on exclusive territorial grants of rights in the form of an amended Commission Regulation, somewhat similar to the ''block exemptions'' (i.e. generic exemptions to the EU competition rules) that allow exclusive territorial allocation of markets in technology licence agreements in well-defined situations. 43 One could, for example, imagine an exemption that would allow distributors territorial exclusivity in respect of films produced in the EU, for a limited duration of, say, two or three years following a film's release. This would be in line with current practices in the exploitation of the more successful European films, as explained in Sect. 3.
The economic justification for such a generic exemption to the EU rules on antitrust, as required by Art. 101(3) TFEU, 44 could be found in the ''special economic characteristics'' of the cinematographic industry that were recognised by the European Court in Coditel II. If it can be established that temporary territorial exclusivity serves as an indispensable incentive to national film production without unduly distorting competition in the European marketplace, arguably such an exemption might stand the test of Art. 101(3) TFEU.

An Alternative to Territorial Exclusivity: Language Exclusivity
As the previous section demonstrates, recent and future developments in EU copyright and competition law have questioned the sustainability of film financing models based on territorial exclusivity. What is often overlooked, however, is that curbing territoriality in copyright does not necessarily spell the end of exclusive grants of rights in respect of language markets. The country of origin rule enshrined in the Satellite and Cable Directive and the Online Broadcasting Directive applies only to the versions of the audiovisual work that are initially broadcast. For example, if a broadcaster based in Germany acquires an exclusive licence to broadcast by satellite a James Bond movie in the dubbed German-language version, only this version may be broadcast to audiences across the EU. Licences for other language versions (e.g. French, Spanish, Italian or Dutch) may still be granted to other broadcasters on an exclusive basis. Therefore, the recent extension of the country of origin rule to forms of online broadcasting does not affect market partitioning and price discrimination based on language versions. Even territoriality in copyright were, in the distant future, to be completely abolished by way of an EU Copyright Regulation that replaces national copyright laws, exclusive grants of rights in different language versions would likely remain completely legal. An exclusive grant to distribute, broadcast or make available a film in a specific language version could therefore co-exist with other exclusive grants for other language versions, regardless of the territories concerned.
Since film markets in Europe are intrinsically separated by language, 45 all these versions might be concurrently, and exclusively, licensed and exploited. Following Brexit, the European Union will have 27 Member States and 24 official languages. Moreover, it is home to more than 60 indigenous regional or minority languages such as Basque, Catalan, Frisian, Saami, Welsh and Yiddish, spoken by about 40 million people. 46 Despite the imperfect match between territories and cultural groups in the EU mentioned in Sect. 2, an overwhelming majority of EU citizens resides in a country where their mother tongue is the official state language or one of the official languages. 47 This implies that -assuming EU citizens want to consume films subtitled or dubbed in their mother tongue -exclusive territorial licencing and exclusive licencing based on language versions would have very similar effects for a large majority of the population. 48 There are, however, a few caveats to this argument. First, the assumption that consumers prefer to watch films dubbed or subtitled in their mother tongue may not be entirely true. In 2012, 44% of EU citizens self-reported to be able to follow news on radio or television in at least one foreign language and this percentage is likely to have increased since then. Nevertheless, being able to follow a news item in a foreign language is generally easier than watching and understanding an entire film, and preferring the foreign version is different altogether. In 2012, 52% of EU citizens still preferred dubbing to subtitling. 49 Those viewers are unlikely to revert to foreign language versions of films that do not even have subtitles.
In 2012, English was the most widely spoken foreign language in Europe at 38%, followed at some distance by French (12%) and German (11%). In 19 out of 25 Member States where English is not an official language, it was the most widely spoken foreign tongue. 50 Thus, a non-negligible part of the population might prefer a version offered in a foreign language, if the offer comes at a substantially lower price or is available sooner than the version in their mother tongue. In particular, these consumers might opt for the English spoken version if that is the original or for a version with English subtitles if a film is produced in any other language. It seems unlikely for viewers to be interested in a version dubbed in a language that is not their mother tongue. 45 Cabrera Blázquez et al. (2015), p. 59. 46 See https://europa.eu/european-union/about-eu/eu-languages_en. 47 European Commission (2012), pp. 4, 11. 48 Taking minority languages into account, the room for market segmentation based on language versions -in this case most likely for subtitles -is in fact even larger. 49 European Commission (2012), p. 118. 50 European Commission (2012 It is worthwhile to analyse how this plays out for different films, depending on their original language and genre: -First, consider a typical non-English European film that takes a significant amount of time to reach other EU markets outside its country of origin (see Sect. 3). Think for instance of a Polish film that is successful in its domestic market, after which it wins the Oscar for the best foreign film. This prize generates interest with a wider audience and enables it to travel to cinemas throughout the EU. For this film, the distortive effect of licencing based on language versions in comparison to territorial licensing will be minimal. Even if the Polish original is effectively available in other EU markets at an earlier stage, this will hardly be of interest to the vast majority of the EU market that requires dubbing or subtitles. -For a film originally produced in English, this may be different. Perhaps 10-20% of the EU population that are not native English speakers might prefer the English original without subtitles or with English subtitles, if it is available at an earlier stage or comes at a substantially lower price. This would increase the demand for, and hence the exploitation value of, the English version, and would accordingly decrease the value of other language versions of the same film. Via online distribution platforms, the English original would service some of the demand in Member States where English is not an official language. If film licencing were to be based on language versions, producers would still be able to grant territorial licences for the UK market for this English film following Brexit. This would avoid most of the market distortion that languagebased licencing might give rise to. However, English-spoken originals offered online in the Irish or Maltese markets would become available to Englishspeaking citizens elsewhere in Europe. For English-spoken films produced in these two countries, the effects of language-based licensing might therefore be more significant, but not necessarily detrimental. -For global blockbusters (such as Avengers: Endgame) that are released on the same day all across Europe, doing away with territoriality in favour of language exclusivity could have a more profound impact. For such films, territorial licencing primarily serves the purpose of price discrimination between highincome and low-income Member States and these practices would indeed be distorted if English-speaking citizens massively revert to, say, the Latvian subtitled version and are able to switch off these subtitles. This could lead to the adverse effect of driving up the license price and consumer price for versions aimed at language areas with lower average incomes. An alternative strategy by right holders could be to make sure subtitles for these language versions cannot be switched off, in order to make them less attractive for other markets.
A second caveat is that replacing territorial exclusivity by language exclusivity would affect co-production agreements between producers or broadcasters operating in Member States sharing the same language, such as Germany and Austria. Assuming that territorial partitioning of rights in this scenario would be ruled out, the co-producers in this situation would become joint right holders for the entire German-language area. While the co-producers could contractually agree to allocate first broadcasting rights to (public) broadcasters operating in each country, grants of downstream rights in the German-language version to distributors or online platforms would require the consent of both co-producers. Rights in other language versions (e.g. English or French) could, however, be allocated, granted and exercised separately.
Third, a potential challenge to language exclusivity might come from automatic captioning, subtitling or even dubbing tools. To illustrate this, YouTube currently offers automatic captioning in English, Dutch, French, German, Italian, Japanese, Korean, Portuguese, Russian, and Spanish. 51 With the rise of artificial intelligence has come increasingly powerful translation software that might one day be built into smart televisions or other viewing devices. Although robotic translation could probably never compete with human actor dubbing, this technology might still negatively affect exclusive language rights. Nonetheless, when assessing this alternative model it is important to realise that market separation by way of territorial grants of rights is also vulnerable to circumvention, as is demonstrated by consumer practices applying virtual private networks and other technical measures to circumvent ''geo-blocks'' aimed at ensuring territorial exclusivity.
In sum, language exclusivity in audiovisual content distribution contracts may provide for a ''natural'' alternative for market segmentation and price discrimination along strictly defined national territorial borderlines. Most likely, since the principles of cultural and language diversity are deeply rooted in the EU, and language regions are distinct from national territories, no conflict with freedom of competition or any of the other Internal Market freedoms would arise. A shift from territorial to language exclusivity would, however, not be without practical and technical challenges and contractual consequences. Nevertheless, moving away from territorial grants of rights to language-based grants would not be a radical departure from current licensing practices in the film industry. Indeed, the fact that territorial modes of licensing in this field have remained prevalent despite the gradual shift of Europe's economy towards a Single Market can be largely explained by existing linguistic differences that largely, albeit not entirely, overlap with state borderlines. 52 Arguably, the practical disadvantages of a shift towards language exclusivity would be outweighed by the legal robustness of this alternative model. While grants of territorially exclusive rights will become increasingly vulnerable, under any scenario, to the combined challenges of further unification and EU competition law, this model would be legally far more robust.

Conclusions
This article discussed the complexities of film financing in Europe, and the role of territorial licences therein. European films are on average far less successful in attracting large audiences than American films. The 2017 market share of European 51 See https://support.google.com/youtube/answer/6373554?hl=en (Accessed 10 September 2019). 52 Cabrera Blázquez et al. (2015), pp. 58-59. films in Europe was typically 20 to 30% of admissions, TV broadcasts, and international SVOD and TVOD catalogues, despite the fact that Europe produces twice as many feature films per year as the US. This relatively weak position causes a typical European film to benefit much less from the economies of scale that films potentially have, making it more dependent on territorial licencing and on public support policies.
Territorial licencing is more important for films with a smaller audience and hence a smaller budget, as it is harder for such films to contract a well-known director or famous actors -bankable names -that may serve as a quality signal to the public and to financiers. Such films need to build their reputation via festivals and awards. Even larger European productions often lack the marketing budget to start a pan-European campaign and do not gain a reputation outside their producing or co-producing countries until they win awards at festivals or are successful in their home markets -only then can they start gaining revenues in other EU Member States. This process may take more than a year from a film's first release. For smaller productions, not being able to grant market-specific licences during this period could erode distributors' willingness to invest in licences and the opportunities to recoup part of the investments. This is different from the trend for large blockbusters, for which theatrical releases have become increasingly synchronised globally.
Therefore, drastic and sudden changes to the possibility for producers to licence on a territory-by-territory basis could have significant effects on the European film industry. At the same time, however, the turbulent growth of internationally operating VOD platforms and the far from perfect match between national borderlines and cultural groups call for flexibility and a willingness to reconsider traditional practices in film financing.
Moreover, EU law aimed at removing national barriers to the Single Market has gradually diminished the role that territoriality can play in copyright. EU competition law also sets strict limits on grants of territorial exclusivity, and prohibits clauses in broadcasting and pay television licences that prevent or restrict ''passive'' sales to consumers/viewers in non-licensed territories. The freedom of right holders to preserve territorial exclusivity by way of contract is likely to become increasingly vulnerable to EU competition law, as territorial grants are no longer supported by underlying territorial rights.
If territorial exclusivity were considered to remain indispensable for the financing and exploitation of EU films, the European Commission could be tasked with laying down film-specific rules on exclusive territorial grants of rights. This could take the form of an amended Commission Regulation, somewhat similar to the ''block exemptions'' that presently allow exclusive territorial allocation of markets in technology licence agreements in well-defined situations.
An alternative to territorial licensing which the film industry can deploy without the need for any regulatory intervention is found in language exclusivity, i.e. exclusive grants of rights for distinct language versions of a film. Despite some practical, contractual and technological challenges that were identified in this article, language exclusivity in audiovisual content distribution contracts provides a more natural and legally more robust alternative for market segmentation along national borderlines. In particular for EU films produced in any other language than English, the distortive effect of licencing based on language exclusivity rather than territorial exclusivity is expected to be minimal. For European films produced in English, the effects may be more significant, but they are not necessarily negative and largely resolved by Brexit. The most profound effect of doing away with territorial exclusivity might be on US blockbusters for which territorial licencing serves primarily as a means of price discrimination.