Hot Take: Europe’s Digital Services Act

Earlier this week, the European Commission released two wide-ranging proposals to govern digital platforms. The Digital Services Act (DSA) and Digital Markets Act (DMA) are based on the understanding that platforms have become quasi-public spaces for communications and trade. Like with Europe’s privacy law GDPR, the implementation will likely fall much short of promise, even if these laws see light of day. However, the signalling power of these measures cannot be over-stated: just like GDPR reverberated across the globe through multiple data protection laws, the DSA and DMA will likely provide moral sanction for other countries to take more radical measures. With that in mind, here’s what I think these two laws mean for us-

I. Rebooting of anti-trust for the digital age has begun- The DMA forbids platforms from treating their own services and products more favourably in ranking than similar services or products offered by third parties on the gatekeeper’s platform. This recognises the ‘public utility’ nature of online platforms and therefore brings in ‘platform neutrality’, i.e. preventing discrimination between participants on the platform. We’ve already seen countries applying that principle to specific sectors, including India’s e-commerce directives. The DMA will likely set the stage for broader application of platform neutrality to all large online intermediaries.

II. Data underpinnings of anti-trust are enshrined in law- The DMA recognises that data is the source of market power, and therefore mandates that platforms must share with their business users the data that the latter generate on the platform. This addresses a long-standing complaint against Amazon and others, i.e. that they use data about sellers on their platforms to out-compete them. However, much like data portability in the GDPR, it stands to question whether this will truly be helpful. As IT for Change argues in its recent paper, real economic value lies in the ‘intelligence layer’ rather than the ‘data layer’. Unless we find a way of sharing that intelligence, data sharing will remain a toothless tiger. The DSA also includes a requirement to explain rankings and hierarchies that platforms generate, while not revealing the algorithm itself. Platforms will also need to open key data for researchers, so that they can look into the inner functioning of these platforms. All of these are useful moves that will be a significant net positive even if they are only partially successful.

III. Interoperability is now a mainstream conversation- The DMA requires gatekeepers to allow third-parties to inter-operate with their services in certain situations. When I first wrote about interoperability in April 2019, it seemed like a pipe dream. However, I also felt that it was the only sustainable, markets-based way to reduce the power of Big Tech firms. It’s been great to see the topic receive more mainstream attention, including Senator Warner’s Interoperability Bill in the US. Interoperability is easier said than done, but it is now clearly a mainstream proposal on the table. Europe’s moves may also provide inspiration to countries in the global south, who are often able to move faster in regulating American tech giants.

IV. Fight against tech giants received more teeth- The GDPR, which allowed for fines for up to 4 percent of gross annual turnover, has not lived up to its deterrence promise. The largest fine on a Big Tech company is the recent $120M fine (and the earlier $60M fine) on Google by the French CNIL. That’s petty change for Google, though- less than 0.08% of its global turnover. The DMA provides fines of up to 10 percent of global turnover. More importantly, it provides more potent tools for regulators to hurt big tech firms where it matters- by separating out different parts of the business. It will be long before we see whether such enforcement mechanisms make it to the final law, and whether they are enforced, but their very inclusion shows a recognition that financial penalties alone will be insufficient, and we need to address the underlying structural and behavioral barriers.

V. The most complex trade-offs remain unresolved- Individuals’ ability to post freely online, and platforms’ obligations to remove such content remains a vexed issue online. We need to strike a fine balance between freedom of expression and creation of safe spaces. The Acts don’t offer radically new ideas. They retain safe harbours from user content unless the platforms detect illegal content, and reiterate that good faith removal of illegal content by platforms will not create liability. When they remove content, platforms will need to explain the cause and provide avenues for the user to contest the decision. The proposals also create a new category of “trusted flaggers”, who platforms must cooperate with in removing illegal content. But determination of illegality of content required nuanced judicial thinking, and platforms (or even trusted flaggers) may not be set up for the task. While these proposals are the best we have today, we need more radical ideas on how to keep the internet safer while protecting fundamental rights.