DPI for Competition

The draft Digital Competition Bill, 2024 is so heavily influenced by European regulation that it feels like yet another example of premature imitation. Which is strange considering that we can so easily leverage our DPI to deliver much more nuanced outcomes.

This is a link-enhanced version of an article that first appeared in the Mint. You can read the original here.

Countries around the world are looking for ways in which to address the challenge of competition in the digital world.

In America, there are a number of bills before the US Congress seeking to provide legislative avenues through which this problem can be tackled. Both the US Department of Justice and the Federal Trade Commission have initiated a number of high profile antitrust lawsuits against Big Tech companies. Europe has already enacted two significant legislations—the Digital Markets Act and Digital Services Act—and has, if anything, further ramped up its enforcement actions against Big Tech.

While India is undoubtedly late to the party, it has recently initiated a consultation on a new draft legislation that aims to remedy this. In its recitals, the draft Digital Competition Bill states that it will identify systemically significant digital enterprises (SSDEs) and regulate the core digital services that they provide. And that it is being enacted to foster innovation, promote competition and protect the interests of users in India. Once in force, it will offer regulators new tools with which to tackle digital competition.

In actual fact, this is a legislation based on the misbegotten notion that ‘Big is always bad.’ By making SSDEs the sole target of new regulation, it will likely put at risk much of the progress that India has achieved in the digital sphere through decades of permission-less digital innovation because of its underlying presumption that all companies in the digital sphere above a given size must be anticompetitive. In a recent paper, Shreyas Narla and Shruti Rajagopalan have critically analysed the proposed law and pointed out its many shortcomings. 

They argue that instead of building the internal capacity to carry out a comprehensive assessment of the behaviour of tech companies, the Competition Commission of India seems to be taking the lazy approach of applying ex ante regulations to a subset of tech companies based solely on the magnitude of their turnover and the size of their user base. Not only does this “blunt instrument" approach pander to the low level of state capacity in our regulatory system when it comes to properly assessing competition in digital markets, it, in the process, disregards important principles of competition law, such as ‘relevant market’ and a proper assessment of market dominance, that must necessarily be applied in all instances.

One of the examples they use to make their point is the prohibition under the proposed law of self-preferencing that would automatically apply to all SSDEs. This, they argue, could have the effect of preventing Google from listing Google Pay on its Play Store, although integrating this UPI-based payment system greatly improves the user experience and can significantly leverage the security features of Google’s operating system. Given that India’s digital payments market is one of the most widely interoperable fast-payment systems in the world, this should hardly be the focus of competition regulation. What’s more, given that this sector is already being actively regulated by the Reserve Bank of India, additional legislative intervention of this sort seems unnecessary.

It is impossible to ignore the similarities between the proposed draft law and the legislation that has already been enacted into law in the EU. But India is a very different market in comparison with the EU.  Simply lifting a legislative approach that might have worked in Europe without paying due heed to our domestic context is “premature imitation," a term that Rajagopalan and Tabarrok use to describe the practice of blindly copying legal measures without properly assessing the consequences of it.

As I have often said in this column, India’s digital public infrastructure makes it possible for regulatory principles to be directly embedded in the software code of the underlying digital infrastructure. It allows us to put in place mechanisms that will let data generated in digital ecosystems be better used by all participants in those ecosystems, and in ways that not only respect the intellectual property and unique business models of companies that process data, but also the privacy of those to whom it pertains.

That being the case, instead of imposing differentiated standards on technology companies because of their size, as the proposed digital competition law does, we should be looking to leverage the power of digital public infrastructure to achieve these outcomes. We could, for instance, keep data federated while at the same time making it more accessible. In this way, we will be able to allow entities that control the processing of that data to continue to benefit from it, while still allowing others to use it with the consent of those to whom it pertains.

If the software protocols through which these measures can be undertaken remain under the control of the regulator, we can achieve our regulatory objectives without having to change the law. What’s more, we will be able to incentivise appropriate behaviour by simply adjusting the relevant parameters of the protocol. Regulators that know how to use techno-legal tools to achieve their objectives will be far more effective in controlling harmful behaviours than would otherwise have been possible.

After all, we will be able to achieve governance goals in much less time and with far greater effect by controlling the relevant protocols than by enacting laws to deal with rapidly evolving technologies.