Copyright, re-use and digital business models

This post also appeared as part of a series of evidence summaries for the 21 for 2021 project, a CREATe project within the AHRC Creative Industries Policy and Evidence Centre (PEC). The 21 for 2021 project offers a synthesis of empirical evidence catalogued on the Copyright Evidence Portal, answering 21 topical copyright questions for the 21st century. With Margaritha Windisch, I identify empirical research relating to copyright, re-use and digital business models.

Through digital technologies, works have become easily reproducible, and new content is produced and published every day due to the simplicity and low costs of production, distribution, and promotion. Music, film, books, as key industries protected by copyright, are most affected by digitization. In addition to the many opportunities, digitization also poses challenges, especially to copyright protection. Digital distribution technologies, in particular, have significantly changed the way copyrighted works are reproduced, distributed, and used. But the screening of digital content for copyright infringement also transformed. Automated notification systems and takedowns, such as YouTube’s Content ID, greatly facilitate the process of licensing; but at the same time, these private institutions also offer some risks.

Our blog presents the evolution of copyright in the digital age in terms of business models and (private) institutions, focusing particularly on the challenges we face in this regard today or in the future. The Copyright Evidence Wiki provides a broad category on business models and licensing among more than 800 papers, discussing, e.g., strategies and licensing solutions for exploiting copyright-protected material and how legal markets attempt to match production to consumption. We will focus on a few selected pieces to give a brief yet incomplete overview of the research area.

Recent evolutions, existing evidence, and research agendas

Digitization and the emergence of digital distribution technologies have facilitated unpaid consumption, whereby commercial providers have had difficulty generating the same revenues with traditional product and market strategies. In particular, in media industries, such as music and film, the digitization process favoured piracy, i.e., unlicensed copying and sharing of copyrighted content, threatening further investments in products (Waldfogel, 2017). Creators need the rents from exclusive intellectual property rights to recoup their investments. The disruptions and reductions in revenue streams from digital distribution technologies have weakened copyright protection. However, the importance of intellectual property rights and their weakening varies depending on the type of cultural asset. New business models have been developed to protect creators’ rights through licensing and ensure that producers’ incentives to create products and thus maintain cultural participation are sufficiently large. At the same time, it should become more attractive for consumers to listen to legally copyrighted music recordings, for example, than to pirated substitute products. In the recording industry, Apple took the first step and developed an attractive legal digital distribution method with the iTunes Music Store, which functions as a transaction-based model. Subsequently, advertising-supported business models, such as YouTube, gained importance. Besides, subscription-based models, such as Spotify (with usage-based license fees) and Netflix (with a fixed upfront payment) use, were developed.

The development of new technologies has led to reduced production, distribution, and advertising costs, particularly in music, films, books, and television, making it much easier to launch new products (Waldfogel, 2017). The low cost of bringing new products to market also means that the level of sales required to recoup investments is lower, which leads to a countervailing effect on weaker intellectual property protection. While technology does reduce copyright protection in copyright-driven industries, the lower cost provides an incentive to continue to generate new products and even leads to increased creative output; without lowering quality. Consequently, consumers benefit from a broader range of media products (Waldfogel, 2012). AI and machine-generated works are also playing an increasingly important role in the context of cultural production. Current AI systems can already quickly create customized royalty-free background music and jungles, as well as entire albums. Handke et al. (2015) show that significantly more musical works and films have been published in recent years than before digitization, while the average quality of creations has remained stable according to users’ assessments. However, they note that countries with relatively strict copyright protection are less likely to encourage cultural participation and consequently do not have a higher supply of new creative works.

The digital environment enables disseminating and communicating artistic and cultural expressions across borders, whether in a professional or user-generated context. The growing prevalence and consumption of user-generated content (UGC) demonstrate that these types of works are of considerable value and complement professional content (Handke et al., 2015). In particular, advancing technology has facilitated the modification and reuse of digital content, such as sampling, remixes, covers, and parodies, which play an increasingly important role. As a result, a work may have multiple copyright owners. In such a scenario, it is essential to define the relationship between copyright holders. Arai and Kinukawa (2014) show that derivative works can benefit copyright holders and the general welfare, so more and more artists are encouraging the supply of derivative works. Regardless of their economic value, they contribute to individual fulfillment and participation in a democratic society.

A common concern expressed directly by creators is whether something they are doing might infringe on someone else’s copyright. When researching to avoid copyright conflicts, creators often encounter and are confused by the law or unable to find satisfactory answers (Fiesler et al., 2015). They consider five main barriers: (i) avoiding trouble, (ii) dealing with consequences, (iii) fear of infringement, (iv) dealing with infringement, and (v) incomplete information. It can, furthermore, be a time-consuming process to get in contact and negotiate with rights holders. While the vast majority get great permission, the biggest obstacle to getting the license is non-response. The data suggest that commercial copyright owners are more likely to deny permission than other types of copyright owners (Akmon, 2010). Problems are also caused by works for which the copyright owner cannot be located; so-called orphan works. Photographs and audiovisual materials are particularly affected. Data show that obtaining licenses to digitize works is often time-consuming and more costly the older the work and the less economic value it has (Vuopala, 2010).

The introduction of automated licensing systems has made it possible to license thousands of derivative works at negligible cost. In this regard, automatic notification systems, such as YouTube’s Content ID, aim to enforce intellectual property and check all uploaded videos against a database of copyrighted material before releasing them. Based on a content review and their risk tolerance, these services decide whether to approve or remove material. The systems included special takedown privileges for trusted senders, advanced profiling of users, restrictive terms of use, modifications to search functions, and other approaches that went beyond the requirements of the US Digital Millennium Copyright Act (Karaganis and Urban, 2015).

This privatization of decision-making in blocking, filtering, or removing digital content poses some risks due to the lack of transparency in the design, implementation, and use of such services. Besides, it can harm fair use, free speech, and creativity. Often it is not clear what unauthorized use constitutes “infringement.” It also happens that the algorithm flags and removes the content of critical works or satirical posts, even though they are perfectly legal. Rightsholders have considerable leverage to prevent follow-on innovations in this regard. Therefore, protection or stronger enforcement of intellectual property can be contrary to the promotion of cultural diversity (Jacques et al., 2017). Moreover, the filter is constantly changing, so videos that initially pass a test must be re-edited later. Using, modifying, and reusing digital content has become risky. Through its automatic copyright filter, YouTube replaces the legal use of copyrighted material with its own rules. To counter abusive takedowns and ensure that takedown systems remain fast, efficient, and error-free, Seng (2015) suggests tightening the certification requirements for notices, requiring reporters and validating all submitted takedown requests.

Concerning algorithmic licensing, Owen and O’Dair (2020) provide insights into the potential applications of blockchain technologies to optimize music industry business models. They find blockchain to be a more efficient and effective way to license and pay artists than a centralized database of music rights. Copyright information is stored in decentralized nodes in the blockchain, enabling unique attribution of licensor and licensee and protects against further tampering (Jing et al., 2021). Moreover, it provides the technological infrastructure to automatically enforce license payments using so-called smart contracts (Catalini and Gans, 2016). However, the legal understanding of various automated systems, including smart contracts, has yet to be developed. The question, therefore, arises as to how we shape (platform) markets in the future and what importance we attach to intellectual property rights to stimulate digital cultural participation optimally.

Future directions for research

The evidence base primarily focuses on online piracy, i.e., the impact of weakened or absent intellectual property protection, including first-order interactions between digital technology and cultural product revenues. In particular, with respect to the current or future role of intellectual property rights, business models, and customization, there are several issues worth exploring (see also Peukert, 2019 for more detail).

For example, few studies to date have examined the relationship between intellectual property rights and supply behavior. Concerning the implications for individual cultural participation, concrete questions for a research agenda include:

  • What are the trade-offs that individual creators face when deciding about digital cultural participation? What is the role of intellectual property rights when individual creators decide about digital cultural participation?

  • How can we think about the distributional effects from producers to platforms that either charge for access to cultural content or sell advertisements? Are there circumstances where a market-based organization of digital cultural participation fails and how should governments (re-)allocate intellectual property institutions and cultural subsidies?

Also, algorithmic licensing, the modeling of algorithmic licensing, and its welfare effects are topics that researchers may want to consider. In doing so, it seems useful to think about the additional value that derivative works can create for consumers, with questions such as:

  • What are the trade-offs involved in algorithmic licensing of cultural products, for individual cultural participation, established firms, and the industrial organization of cultural markets?

  • Can we measure and compare the welfare effects of derivative works, under traditional licensing and under algorithmic licensing?

Thus, we can see that there is still a need for research in copyright due to the advancing digitization, emerging technologies, and new business models.