SO MANY false starts would have soured other romances. Resistance from antitrust authorities halted a union between T-Mobile and Sprint, America’s third- and fourth-largest wireless carriers, in 2014. A row over merger terms scuppered talks last year. But the attraction never dimmed.

This week the pair announced an all-stock deal that would create a company with a heft similar to that of AT&T and Verizon. The happy couple promises lower prices for customers, higher profits for shareholders and a sharpening of America’s technological edge (see article). Regulators should be sceptical. The tie-up is bad for consumers; and there are better ways to build whizzy new networks.

Consumer welfare first. The international evidence suggests that cutting the number of big operators would be bad for customers. Research by British regulators into 25 countries shows that average prices were up to one-fifth lower in markets with four network operators than in those with three. (Ignore the claims of T-Mobile and Sprint that the American market is contested by as many as eight firms: in its latest report on the industry, the Federal Communications Commission found that the four carriers accounted for over 98% of connections.)

T-Mobile itself is testament to the benefits of a more crowded market. Trustbusters not only zapped its discussions in 2014 with Sprint but also blocked an earlier attempt by AT&T to buy it in 2011. The firm has thrived on its own. It has added almost 40m customers in the past five years by cutting prices and adding features such as free video-streaming. Subscribers everywhere have felt the benefits. Between 2013 and 2016 overall consumer prices in America rose by 4.5%; prices for wireless telephone services decreased by 8%. Consolidation threatens a different outcome. The combined firm projects relatively slow growth in revenue, a jump in profit margins and rapid deleveraging. That does not sound like the plan for a price war.

If regulators have opposed such tie-ups before, why do T-Mobile and Sprint expect a different answer this time? One explanation is the risk that Sprint, which is heavily indebted and has been struggling for a while, might go bust if it remains a stand-alone entity. But that ought not to sway the trustbusters. Sprint could shed its debts in a Chapter 11 bankruptcy process and re-emerge in better shape, or it might get swallowed up by a different firm entirely.

Trumpeterian destruction

The second explanation is that the two firms think that they can win a public-interest argument about technological leadership. The bosses of T-Mobile and Sprint argue that by bringing together their bands of spectrum, they would be able to build America’s first national 5G network. Their merger presentation, featuring slides with headlines such as “US must lead innovation again” and “Global economic leadership is at stake!”, was aimed as much at economic nationalists in the White House as analysts with spreadsheets.

It is true that 5G networks are expensive to build: they require more antennae, base stations and fibre-optic cables than their predecessors. It is also true that 5G’s speed provides a platform upon which all sorts of data-hungry new services, from self-driving cars to industrial robots, can develop. But that does not mean operators have to build their own, separate networks. Mobile providers in South Korea have agreed to share the costs and use of 5G infrastructure. Mexico is building a wholesale mobile network; its capacity can be leased out to different firms. Better this approach than muted competition and price-gouged consumers. The union of T-Mobile and Sprint is one that regulators should not bless.