Whoever controls the counterfactual controls the decision

Posted on: 30 October 2026

This morning the Financial Times reported that Virgin Media O2 is close to a deal to sell broadband over Openreach, BT's network, in the areas its own network does not reach. The same report carries a detail that matters more: the company is said to be pausing the fibre upgrade of its cable network as part of a cost-cutting programme. Neither side has confirmed anything and no terms are known, so what follows is a reading rather than a certainty.

The timing, however, does not strike me as accidental. In February Nexfibre, the joint venture between Liberty Global, Telefónica and InfraVia, agreed to buy Netomnia for £2 billion. In October the CMA provisionally found that the deal would substantially lessen competition, and the owners chose not to offer remedies but to challenge the regulator's reasoning outright. The final decision is due by 15 December.

Part of that reasoning rests on a hypothetical future. If VMO2 upgraded its entire cable network to fibre, its overlap with Netomnia would rise from around 26% to around 82%, which is why the regulator concluded that VMO2's own upgrade programme would erode most of the benefit the deal claims to deliver. Six weeks before the ruling, that programme is put on ice.

Every competition authority works by comparing the world with the deal against the world without it. The second world does not exist, though. It has to be reconstructed from plans and intentions that belong to the very company under scrutiny. When that company can switch off the alternative scenario with a board decision, the counterfactual stops being a fact and becomes a bargaining chip. The Openreach deal supplies the missing piece, because it makes credible the idea that without Netomnia, BT's main infrastructure challenger would retreat into reselling on its rival's network.

There is a duller reading too. VMO2 may simply lack the money to do both, in which case it is confessing weakness rather than applying pressure. The structural lesson is the same either way.

And it travels well beyond telecoms. A board that approves an acquisition on the strength of management's business plan makes the regulator's mistake, because it is judging a scenario the other party can cancel cheaply. Before asking whether an assumption is plausible, it is worth asking who has the power to kill it and what it would cost them. If the answer is "not much", you are not assessing an assumption. You are negotiating without knowing it.