Faculty of law blogs / UNIVERSITY OF OXFORD

The Economics of Online Copyright: A New Framework for Platform Regulation

Posted:

Time to read:

2 Minutes

Author(s):

Alessandro De Chiara
Associate Professor at the Department of Economics of the Universitat de Barcelona
Ester Manna
Associate Professor of Economics at the Department of Economics of the Universitat de Barcelona
Antoni Rubí-Puig
Associate Professor in Civil Law at the Universitat Pompeu Fabra, Barcelona.
Adrián Segura Moreiras
Part-time Lecturer at the Department of Economics and Business of the Universitat Pompeu Fabra

In the paper Efficient Copyright Filters for Online Hosting Platforms, forthcoming in Information Economics and Policy, we develop a formal economic model to explore how online platforms can be effectively regulated in their role as intermediaries that host user-generated content potentially infringing copyright. The regulation of these platforms—such as YouTube, Facebook, or TikTok—has become a central concern in both the US and the EU, particularly given their reliance on automated filtering systems and their use of notice-and-takedown procedures to handle copyright claims. While these mechanisms are designed to reduce litigation and enhance copyright enforcement, they often lead to over-removal of legitimate content and under-removal of infringing material.

Our analysis starts from the observation that both pre-publication filters and post-publication notice-and-takedown systems are imperfect. Filters frequently make ‘Type I errors’ by blocking non-infringing content, namely, material that qualifies as fair use. Similarly, automated systems used by right-holders to send takedown notices can be inaccurate. Platforms, lacking the incentive to verify the merit of the claims, often comply without challenge.

We study a model in which a platform decides how much to invest in its filter technology and subsequently a right-holder decides how much effort to put into evaluating whether the content posted online infringes its copyright. We show that the prevailing legal standard—based on safe harbour rules that shield platforms from liability if they promptly comply with takedown requests—induces suboptimal filtering. Specifically, platforms may over-rely on takedown notices and under-invest in effective filters. To restore efficiency, we suggest a policy shift: platforms should face a calibrated expected loss each time an infringement is detected, irrespective of whether a takedown notice is received and acted upon. This expected loss can take the form of a payment to the state and should be set to account not only for the copyright holder’s harm but also for the value that users and content creators attach to the content. We also propose introducing a penalty whenever a court upholds the platform’s decision not to remove the content. This penalty should be linked to the damages the right-holder would have received if the content had been infringing, thereby encouraging careful evaluation before issuing notices. This will improve accuracy and reduce unnecessary over-removal.

The dual regulatory system we advocate thus includes both an ex-ante mechanism, which induces efficient filtering through strict liability, and ex-post penalties that discipline the behaviour of right-holders. This structure achieves a socially optimal balance between the protection of copyright and the preservation of freedom of expression.

Beyond the core model, we explore several important extensions. We show that if the platform outsources the development of the filtering technology to a third party (as many do, using tools like Audible Magic), the policy must account for new incentive distortions—specifically, moral hazard from the external developer. We also consider scenarios where copyright infringement causes only minor harm, and the right-holder may prefer monetization over removal. Here, we show how monetization can be socially beneficial and discuss how platforms’ filtering incentives should change when infringement is not necessarily a reason for removal. Additionally, we examine the intrinsic value that users attach to original content and show that platforms may already have incentives to invest in filtering if originality boosts user demand.

Finally, our paper connects with ongoing policy debates, including the implementation of Article 17 of the EU Copyright Directive and proposals to revise the US DMCA. In both jurisdictions, lawmakers face the challenge of balancing the enforcement of copyright with users’ rights to access and share lawful content. Our model provides a framework to assess existing liability regimes and design reforms that promote efficiency, minimize litigation, and reduce wrongful censorship.

The full paper is available here.

Alessandro De Chiara and Ester Manna are Associate Professors at the Department of Economics of the Universitat de Barcelona.

Antoni Rubí-Puig is Associate Professor in Civil Law and Adrián Segura-Moreiras is a part-time Lecturer at the Department of Economics and Business at the Universitat Pompeu Fabra, Barcelona.