Attacks on tech company size and power are increasing, including recent articles in The Economist and Washington Monthly. I’ve dealt with some of these complaints in earlier columns, arguing that data is not the new oil, not a digital infrastructure, and not a barrier to entry.
But something funny happened on the way to the antitrust courts – advocates realized that applying traditional antitrust remedies against tech companies would do more harm than good.
Breaking the companies into smaller parts, as some explicitly advocate, would just sacrifice the efficiencies that make tech products and services so valuable. Utility-style price and entry regulation threatens to slow innovation and seem pointless when many of the tech services are free to consumers. Also, nondiscrimination requirements would undermine a tech company’s ability to restrict harmful material on its own system.
So, advocates are turning to regulatory proposals relating to free expression, privacy, and access to information. But the urge to do something should never be stronger than the urge to do something intelligent. Bad regulatory proposals cannot be salvaged by re-describing them as antitrust remedies.
Here’s an example: in seeking ways to “tame tech,” The Economist almost off-handedly says, “Immunity to content liability must go, too.”
It is hard to see what this has to do with generating tech industry competition. Content liability just imposes regulatory burdens that make it hard or impossible for anyone to function in the market. Indeed, the burden would fall overwhelmingly on smaller players and increase barriers to entry for new players. So as opposed to “taming tech,” it would effectively lock-in the current market leaders.
How platform distributors of news and information should be regulated is a serious and profound question. As a new report from the New America Foundation and the Shorenstein Center points out, hate speech, extremist content and disinformation campaigns on social media platforms threaten the foundations of democratic discourse and call for fresh thinking to assess a range of new policy options.
But imposing on social media platforms the same content liability that print magazines face is foolish. Individuals who are completely independent of platforms use them every hour of every day to further their business and creative interests and to share and seek the information they need to be informed citizens. If the platforms had to exercise the same level of care as the editor of The New York Times before users could post anything, they would not function as effective distributors of content at all.
Policymakers have to be careful not to throw the baby out with the bathwater. As Harvard Law School professor Cass Sunstein says, on balance, social media platforms are not merely good for democracy, they are “terrific” because they make it vastly easier for citizens to get the information they need to govern themselves.
Most importantly, counterproductive new content regulations should not be disguised as antitrust remedies against big tech.
Here’s another example. To fix competitive problems in the tech industry, The Economist and others urge regulators to give people “more control over their information.”
In fact, this has been an objective of data protection policy on both sides of the Atlantic for over a generation. The EU just finished a far-reaching update of its privacy law, the General Data Protection Regulation, which is set to go into effect in May 2018. This is the most sweeping reform of data protection laws in history. To comply, technology companies and other affected businesses will be updating their systems for years to come.
What the Economist seems to mean, however, is not that we should protect user privacy, but that we should use privacy law to dry up the supply of data to tech companies. It is hard to see how this furthers competition. Instead, the lack of data would hurt everyone in the market.
The Economist suggests a further, and contradictory, data reform – user data should be turned over to anyone willing to pay a compulsory license fee. As the Deputy Assistant Attorney General for Antitrust noted recently, this forced sharing of a company’s critical asset would tend to dry up the incentive to invest in innovation. And there is no need for this step. As I pointed out in an earlier column, competition authorities have repeatedly found that data needed for competitors to challenge incumbents is in plentiful supply. Moreover, compulsory data sharing is hard to reconcile with the privacy rights of data subjects, since information that a user shares with one company would then be automatically available to the entire marketplace.
Public policy proposals can serve multiple purposes. There’s nothing wrong with a measure that by luck or intent protects privacy or free speech and also promotes competition. But these unnecessary and counterproductive proposals serve no legitimate government purposes.