“Apple undermines apps, products, and services that would otherwise make users less reliant on the iPhone, promote interoperability, and lower costs for consumers and developers.” That is the Justice Department’s own statement of its case, filed in New Jersey on Thursday with more than a dozen states. The attorney general put the same theory more sharply at the podium: that the company consolidated its position not by making its own products better, but by making other products worse.
The suit alleges illegal monopolization of the smartphone market, built on a list of specific practices: blocking super apps that would make switching easy, blocking cloud game streaming, degrading cross-platform messaging, and designing features to raise the cost of leaving.
Seventeen days earlier and four thousand miles away, a different government did something about very similar conduct, and the comparison is more instructive than either action is alone.
The two instruments
On 4 March the European Commission fined Apple about 1.84 billion euros over anti-steering rules: the App Store terms that stopped music streaming apps telling their own users they could subscribe more cheaply on the web. The Commission ordered the terms removed, and Apple had already removed them as the investigation closed.
Both actions concern the same underlying complaint, which is that the owner of a platform sets the terms on which everything else runs on it. What differs is the instrument, and the instruments are not variations on a theme. They are different machines.
Washington filed a lawsuit. It has to prove a market, prove monopoly power in it, prove anticompetitive conduct, and then persuade a court to order a remedy. That is a matter of years.
Brussels ran an administrative proceeding under an existing prohibition, reached a decision, imposed a penalty and ordered specific conduct to stop. That is a matter of a finished investigation.
What each one has to prove
The American case turns on a question the European one did not have to reach: what is the market.
Monopolization requires monopoly power, and power is always power over something. Define the market as smartphones in the United States and a share can be argued. Define it as smartphones worldwide and it looks different. Define it as all devices people carry, or all ways of running software, and it looks different again.
Apple’s defense will be substantially about that definition, because it has to be, and because the government’s own list of practices cuts both ways: an integrated product that is harder to leave is also, as a description, a product that works well because its parts were designed together.
The European proceeding did not need a court to agree on a market before anything could happen. Dominance was established, the conduct was found abusive, the fine was calculated, and the order issued.
What each one can achieve
This is where the difference stops being procedural.
A fine is a cost. 1.84 billion euros is a large number and it is not large relative to the revenue of the business it concerns, which means the deterrent effect depends on the expectation of it recurring rather than on the sum. The accompanying order to stop specific conduct is the part with teeth, and its scope is exactly as wide as the conduct investigated: anti-steering in music streaming, and nothing else.
A monopolization case can reach much further. The government has said structural relief is on the table, which is the term of art for breaking a company up, and even short of that a court’s remedy can govern conduct across a business rather than one practice in one category.
So the European instrument is fast, narrow and certain. The American one is slow, broad and speculative. Neither of those is better; they are different bets about how to intervene in a market that keeps moving while you intervene in it.
What the difference teaches
The lesson is about timing more than about power.
A narrow order that lands now changes what developers can do this year. Music streaming apps can tell their users about their own websites, which is a small change with an immediate effect on somebody’s business.
A broad case that lands in 2029 changes something about a market that will not be the market it was filed against. The Microsoft case of the late nineties is the standing example: the remedy arrived, and by the time it did the thing that actually changed the industry was the web browser becoming irrelevant to the next platform.
Against which: narrow orders are a treadmill. Each one requires its own investigation, each addresses one practice, and a company facing them can comply precisely and change nothing structural. Europe’s answer to that is a separate regime of standing obligations for large platforms, which is a third instrument again and an admission that case-by-case enforcement was not keeping up.
There is a second asymmetry worth naming, which is who bears the risk of being wrong. A fine that is later annulled on appeal is refunded, with interest, and the market has in the meantime operated under the corrected rule. A monopolization case that is lost after five years leaves no remedy, no refund and five years of conduct that is now, by judicial finding, lawful.
What neither one does
Neither instrument answers the question that makes this hard, which is whether the integration users like and the integration developers object to are the same thing.
A phone where the messaging, the payments, the watch and the store are designed as one system is easier to use and harder to leave, and those are not two properties. They are one property described by two parties with different interests.
Both processes will have to make that judgment implicitly, and neither is equipped to make it explicitly, which is why the reasoning in both will read as being about something narrower than what is at stake. A court can rule on whether particular conduct crossed a line. A regulator can order a practice stopped. Neither has a method for deciding how much convenience is worth how much lock-in, and both will end up deciding it anyway, in the course of deciding something narrower.
Where this sits
So: two governments, one company, one underlying complaint, and two instruments with very nearly opposite profiles. One has already produced a fine and an order and is finished. The other has produced a complaint and will not produce anything else for years.
The useful thing to hold is that the fast one changed a specific rule about a specific category of app, and the slow one is the only one of the two that could change what the phone is. Which of those matters more depends entirely on whether you are shipping something this year.
