The problem with ex-ante regulation is that you can kill really good ideas before they take off. Government regulations should target known problems rather than

The problem with ex-ante regulation is that you can kill really good ideas before they take off. Government regulations should target known problems rather than problems that people imagine will happen. That does mean we need to let some problems happen, but that's a tradeoff worth making in a society that cares about making innovation possible. By the way, when Facebook was new nobody would have guessed that it might be involved in a Russian plot to influence a presidential election more than a decade in the future, so regulation would not have been proposed anyway. And back in the 90s, the Justice Department claimed Microsoft was a monopoly that could only be stopped by government action. And then over the next few decades, Microsoft was knocked off its pedestal by companies and market trends that nobody anticipated. If it really had been a monopoly, that wouldn't have been possible. Governments have a pretty bad track record of predicting the future, especially regarding the interaction between technology and culture. > You can't know what their self-regulation means. You also can't know what internal steps a company like Facebook has taken to deal with government regulations like GDPR. But you can measure outcomes. Given that regulations are intended to produce outcomes, that's what you ought to be measuring anyway.