Europe's Draft 6% Global-Turnover Penalty Exceeds Amazon's Entire German Business — and the Real Fight Is Over Territoriality

The Commission's leaked Digital Fairness Act draft does one thing that matters more than its headline number: it makes the design of a sales funnel a matter of EU law with a global price tag. Articles 10 to 17 of the text prohibit manipulative interface designs by description, extend profiling-based advertising rules to minors and vulnerable consumers, require features that encourage compulsive use to be switched off, and push disclosure duties onto the brands and agencies behind influencer campaigns. The draft would also repeal and replace the Unfair Commercial Practices Directive with a directly applicable regulation, which is what opens the door to Commission-level enforcement rather than enforcement by national watchdogs alone.

Investors should not treat the 6% as news. The Digital Services Act has carried the same ceiling since it applied in 2024, and the Commission restated it in February 2026. What the draft changes is who can be caught by it: the DSA reaches very large online platforms and search engines, while the leaked text reaches retailers, app makers, game studios, streaming services and marketplaces, with penalties running through a revised consumer-protection cooperation framework rather than a platform-size threshold. A compliance assessment published in October 2026 argues that most of what the act targets is already unlawful — which is precisely why the new part is the enforcement architecture, not the prohibition list.

Apply the draft's formula to the latest filed revenue and the ceiling stops being abstract. For Amazon it is larger than the company's entire German business; for Criteo it is one and a half times a full year of profit; for PDD Holdings it is close to a third of trailing net income. Against that, Europe is a rounding error in group revenue for every one of them, which is why a rational board does not litigate the perimeter — it redesigns the interface and ships the change worldwide. That is the investable asymmetry: rational boards redesign the interface rather than litigate the perimeter.

Here is the causal chain that decides who pays. Because liability scales with worldwide turnover while the offending conduct is one button, one countdown or one pricing rule, the cheapest response is not to build a compliant EU version of the product — it is to replace the interface everywhere. That converts a European consumer rule into a global product standard, and converts a legal cost into a fixed engineering cost spread across every market a company serves. The cost per customer is therefore near-zero for a platform with hundreds of millions of users and prohibitive for a marketplace running on thin per-order margin, even though both are bound by the same 6% formula.

That last point is what turns a consumer rule into an industrial policy with an external target. Cross-border, low-price, high-turnover commerce models depend on interface mechanics — countdown timers, tiered discounts, defaults, urgency framing — that Article 10 would prohibit by description rather than case by case. Domestic incumbents with scale and legal depth can absorb the same rewrite at a lower cost per customer. The practical effect of a global-turnover penalty base is to price access to European consumers, and the party least able to pay that price is the seller whose European economics were built on conversion mechanics that now need redesigning or withdrawal.

Brussels has been collecting, and the tempo is what matters. The DSA's 6% ceiling was theoretical until July 2026; by then the Commission had already fined Temu, and it added Google and AliExpress in the same month — three cash decisions inside ten weeks turned an abstract ceiling into precedent. The July 23 decisions against Google are the clearest signal of appetite: they priced behaviour, not paperwork, and ordered the non-compliance brought to an end. Booking's third quarter is the mirror image from the other side of the Atlantic, where a US regulator took $900 million of revenue and $100 million of profit out of a European platform's US flows in a single quarter.

The transatlantic frame is already set. In March 2025 the US trade representative named the Digital Markets Act and the Digital Services Act as digital trade barriers in its National Trade Estimate. In August 2025 the United States and the EU signed a framework agreement committing to address \"unjustified digital trade barriers\", and the President threatened additional tariffs against countries with \"digital taxes, legislation or regulations\". The Commission has responded that the DMA and DSA are not part of trade negotiations. The Digital Fairness Act lands squarely in that fault line: a consumer rule whose penalty base is global turnover gives Washington a specific, nameable grievance and gives Brussels a priced threat.

The law does not sit on one industry; it runs through the whole advertising-to-checkout chain. Upstream sit the firms that sell the mechanic itself — the ad-tech and measurement layer whose European revenue is built on personalisation. Midstream sit the platforms that own the interface and the brands and agencies that specify it. Downstream sit the cross-border sellers whose unit economics depend on urgency mechanics, and the European consumers who are nominally protected by all of it.

The winners are less obvious than the losers. Scaled platforms that can standardise an interface globally should, in theory, gain relative share from cross-border sellers whose economics cannot absorb the rewrite — a moat that widens with compliance rather than narrows. That is the uncomfortable read for investors in the incumbents: the same law that raises their own compliance billhardens their position against the sellers it is aimed at. The bill lands on the platforms, the moat lands on the platforms, and the bill for market access is paid by whoever was selling into Europe from outside it.

In the next two quarters nothing is fined and no product changes. The proposal goes to the College of Commissioners on November 11, 2026 alongside a wider consumer package that also reviews the rules on cross-border consumer-protection cooperation; from there it moves to the Council and Parliament. The near-term trade is entirely in the text, because the first live question is not the percentage but who may enforce it and against which revenue.

Over one to three years the structural effect is standard-setting, and it is larger than any fine. A Brussels-written interface standard becomes the global default because the cheapest way to avoid a global-turnover penalty is to comply globally; every seller that wants to keep operating in Europe then inherits the same design rules for every market. The main risk to that thesis is territoriality. If the final law measures the penalty base on EU turnover or EU-market sales, the global arithmetic disappears, the instrument deflates, and Washington's case for treating this as a trade barrier weakens with it.

Europe is not raising its digital fine ceiling; it is widening who can be fined, for what, and against whose revenue. The 6% figure was already in force for the largest platforms, and the draft's real move is to attach it to every trader that sells to EU consumers, to make manipulative design unlawful by description, and to let Brussels enforce it directly. The consequences follow from the base rather than the percentage: because the base is global turnover, compliance becomes a global product decision, which raises the fixed cost of selling into Europe, which prices access for thin-margin sellers while hardening the position of scaled platforms.

What to watch is therefore narrow and datable. The Commission's November 11 text, the articles that decide direct enforcement, the Council and Parliament amendments that decide territoriality, and the first issuer disclosures that attach a number to the exposure. On the current evidence the fine itself is close to irrelevant — the largest EU digital penalty of 2026 was €890 million against a company with $471 billion of revenue — while the commercial consequence runs through redesign cost, monetisation limits in Europe, and the transatlantic tariff threat that a global base makes credible. Brussels and Washington will end up trading the penalty base, not the percentage, and that is where the value in these names sits for the next two years.