CCORTESIn calls to protect consumers, children, rights-holders a lot is being asked of the various online intermediaries that we use every day. Carlos Cortés of the Center for Freedom of Expression at Palermo University in Argentina warns that sounder clearer policy is needed on the liability of these online intermediaries.

In any given day in front of our computer we use all sorts of services: first, we access the global network through an Internet Service Provider (ISP). And once online, we update our Facebook status, post a blog entry in WordPress, tweet an interesting article on Twitter, search a topic in Google, listen to music through Spotify…

These companies, known generally as online intermediaries, are in an ideal position to control users’ behaviour –something of great interest to governments. If a state wants to prevent or deter any given online activity (child pornography, copyright infringement), there is one straightforward strategy: use these intermediaries as a proxy for enforcement. This is what is commonly known as the gatekeeper strategy.

The best way to encourage online intermediaries to act as gatekeepers is by means of liability schemes. In other words, by establishing conditions under which companies like Google or Twitter are responsible for the alleged misdeeds of their users. There are no one-size-fits-all solutions. Liability schemes range from total immunity (the Communications Decency Act in the US, for example) to conditioned immunity (the US Digital Millennium Copyright Act or the European E-Commerce Directive).

Setting aside the specific solutions, we are missing or taking for granted an important point. Any gatekeeper strategy, such as intermediary liability in Internet, should be assessed against its costs, reasonableness and efficiency. Otherwise we would be, or certainly are, imposing constraints on intermediaries and users with no apparent reason. Almost thirty years ago, Reiner Kraakman drafted the elements of a sound gatekeeping strategy: i) There is a clear undesired conduct that cannot be deterred through regular enforcement; ii) intermediaries have no private incentives to control it; iii) such intermediaries are in a good position to control the conduct regardless of some alternative options; and, iv) the costs to encourage intermediaries to cooperate are reasonable.

Much of the online conduct that governments or private stakeholders want to control would not meet Kraakman’s framework. Piracy is perhaps the best example. While the gatekeeping strategy is supposedly meant to tackle a specific conduct –copyright infringement–, its actual application affects all sorts of activities: Many lawful and socially desirable activities –creation, speech, education– end up being curtailed along with the ‘undesired’ conduct.

On the other hand, although intermediaries would not have an incentive to control piracy if laws like the Digital Millennium Copyright Act did not exist –in principle, pirated videos do not affect the business of YouTube or Google – it is arguable that such schemes are deterring the general conduct. The first users fleeing YouTube are those in the piracy industry, who will certainly find other channels for their infringing activity.

Finally, online intermediaries have to assume the costs of a gatekeeping regime (setting up a legal unit, adjusting their applications to comply with their gatekeeping duties, etc.). This is might be reasonable under effective schemes, but becomes onerous and pointless in the face of ineffective ones, especially for newcomers. Let’s suppose Google and Facebook already did the math and will remain in business despite some extra spending. What about start-ups? What about intermediaries in developing countries?

These are some of the issues we examine in a recent paper with the Center for Freedom of Expression at Palermo University, in Buenos Aires, Argentina. In the Latin American context, where the online intermediary liability debate is still in its initial stages, it is necessary to reassess previous experiences and redraw existing policy responses. Re-examining intermediary liability schemes under comprehensive policy frameworks is a global need. Not only are online services cross-border in nature, but also regulation is being adopted similarly in many countries following trade agreements and multi-lateral instruments.

Speaking from the ‘pre-Internet’ era, Kraakman is warning us about the risk of policy drafted solely with the undesired conduct in mind, disregarding possible negative consequences of the policy, the interests of stakeholders –users, among them – and balance with equally important values and conduct.  We should take this advice seriously.

This article gives the views of the author, and does not represent the position of the LSE Media Policy Project blog, nor of the London School of Economics.