This (short) piece – that can be found here – makes a rather straightforward argument: as competition law confronts new business models enabled by new technologies, it is important to advance theories of harm that are scientific – in the Popperian sense that they must be drawn in such a way as to allow one to demonstrate that the theory might be wrong (or, as the technical jargon puts it, falsified).
The paper begins with a description of network effects, before describing how quality is at the heart of a number of cases concerning digital platforms. network effects mean that increased concentration – and reduced competition – may improve the perceived quality of the product. This leads to a heretical question: ‘should we relax antitrust enforcement when network effects are strong? (…) as big data and network effects become more common in the economy, competition enforcement should become less aggressive so as to allow the benefits of network effects to be enjoyed.’
This hypothesis is quickly rejected, however. In doing so, the author adopts two arguments:
- On the empirical front, while research is at an early stage, there are already some conclusions on the relation between network effects, competition and consumer welfare. The problem is that those conclusions are often at odds with each other.
- On the theoretical front, there are also different views regarding the costs and benefits of limited competition in industries that exhibit significant network effects.
Given this, we are currently at a stage where we still don’t really understand the impact of digital platforms with network effects on consumer welfare. As such, we should try to come up with theories that can be tested in order to build up our knowledge over time.