When we think about technology and technological change, information technology is capital, so the owners of that capital have done better and better. This whole literature on the decreasing labour share and the increasing capital share of the economy is partly related to the ability to scale through information technology.
The impact of these particular data-driven technologies on market power is more subtle. The reason is that, in a Stats 101 sense—and I don't know if and when you took statistics—there are decreasing returns to scale and data in a formal technical sense. It's like this: If you have 10 people and you get an eleventh, you learn a lot; if you have a million people and you get one more, you don't learn that much. In an explicit technical sense, there are no economies of scale in data.
On data-driven and machine-learning technologies, there are reasons not to expect monopolization. There are, importantly, other forces going in the other direction, and those are the things that we should keep an eye on and regulate. The forces going in the other direction are things like.... The ability to use data requires certain other technologies or can benefit from certain other markets where there is dominance. The ability to use data, and use it well, requires computing, so if the cloud services market is monopolized or has strong market power, that's going to impede innovation and be a real competition worry. Also, if media, in some ways, are monopolized, and therefore the ability to understand users as they interact with media becomes monopolized, that's another way that could [Technical difficulty—Editor] in terms of competition.
In my view, those are related, but only tangentially related, to most of the content of Bill C-27.