Deutsche Telekom's proposed acquisition of Fiberhost and INEA is easy to describe as a €1 billion bet on convergence. The harder question is what must remain separate after the deal. Fiberhost sells access to infrastructure; INEA sells broadband and television to households; T-Mobile Polska sells mobile and a growing fixed offer. Joining those layers can improve economics, but only if a network owned by one retailer remains credible to other retailers.
That makes wholesale neutrality part of the asset rather than a regulatory afterthought. The transaction can add customers, reduce dependence on rented lines and improve bundle retention. It can also make rival internet providers question whether the owner of the pipe will treat them on equal terms. The investment case turns on solving both sides at once.
The deal joins a pipe to a subscriber book
The official announcement says Deutsche Telekom agreed to acquire 100% of Fiberhost and INEA from Macquarie European Infrastructure Fund 5 and minority shareholders at an approximate €1 billion valuation. Fiberhost's fibre-to-the-home network reaches more than 1.4 million households, while INEA contributes more than 300,000 broadband and television customers. Completion requires approval from Poland's competition authority, UOKiK.
Those are two different assets. The network earns value by carrying traffic for multiple service providers across a largely fixed cost base. The retail business earns value from monthly customer relationships, pricing, service quality and churn. Owning both can remove contracting friction, but a valuation cannot be tested properly until investors know how much belongs to infrastructure cash flow, how much to the customer book, and what debt and capital commitments accompany them.
The customer addition is material. T-Mobile Polska reported 771,000 fixed-broadband customers at the end of the second quarter, up 14.2% year over year. INEA's stated base is equivalent to roughly 39% of that number before allowing for differences in definitions or any overlap. This is an inference about scale, not a forecast that every INEA account simply becomes an incremental T-Mobile account.
The same quarterly release put T-Mobile Polska's revenue at PLN 1.9 billion and EBITDA after leases at PLN 584 million. Those figures show an operating business with positive momentum, but they do not establish the acquired return. The missing bridge is the acquired companies' revenue, EBITDA, capital expenditure, wholesale mix and financing structure.
Wholesale tenants do not disappear at closing
T-Mobile says Fiberhost will continue to offer transparent, equal and non-discriminatory access to all current and potential internet providers in Poland. That promise preserves an important source of utilisation. A fibre line has high construction cost but low incremental cost when another provider lights a household already passed. More wholesale tenants can therefore spread fixed costs across more revenue.
The market has many potential tenants. Poland's electronic-communications regulator reported more than 2,600 telecommunications operators, mostly local or regional, and said micro, small and medium-sized providers deliver more than half of fixed-internet services in rural areas. For Fiberhost, those firms are not peripheral: they can convert coverage in smaller communities into paying connections.
Ownership nevertheless changes incentives. T-Mobile could benefit when a household buys its retail bundle instead of a rival's service over the same fibre. A rival may worry about wholesale prices, installation queues, fault repair or access to new network locations even if formal terms remain available. Discrimination need not take the form of refusing access; small differences in service quality can affect customer acquisition and churn.
There is a strong counterargument. In January 2026, UKE found effective competition in both local and central wholesale fixed-access markets. T-Mobile was already a Fiberhost wholesale customer, so common ownership could improve network planning and investment coordination without displacing other providers. The competitive baseline is therefore stronger than a simple monopoly story suggests.
Convergence must improve yield without shrinking choice
Convergence creates value through specific mechanisms, not through the label itself. A mobile operator can offer one bill, coordinate installation, market to an existing subscriber base and use a broader relationship to reduce churn. A network owner can also direct expansion toward areas where retail demand is visible. If those effects raise take-up on existing fibre, the additional revenue can carry attractive incremental margins.
But ownership does not erase network costs. Fibre still needs maintenance, electronics, customer installations and expansion capital. INEA's customers still need support, content and competitive prices. Integration can also duplicate systems before it removes them. The deal would destroy value if the buyer paid for growth that already depended on wholesale partners, then lost those partners while carrying the same physical network.
The most useful comparison is therefore not T-Mobile's mobile margin against INEA's retail margin. It is the combined yield per home passed: retail and wholesale revenue generated by each reachable household, less operating and sustaining capital costs. Neither the announcement nor the discovery report supplies that measure, so any claim of immediate accretion would be premature.
Approval and churn will expose the economics
UOKiK's concentration framework allows it to approve a transaction, impose conditions or prohibit it if competition would be significantly restricted. The review can test more than market share. Because this transaction combines infrastructure and retail activity, access terms, service quality and the ability of third parties to compete are economically relevant.
The thesis would strengthen if approval preserves workable wholesale access, third-party volumes remain stable, fibre take-up rises, fixed-mobile churn falls and capital expenditure produces more connected homes rather than only more homes passed. Separate reporting of wholesale and retail performance would make that progress much easier to judge.
It would weaken if remedies materially limit integration, wholesale customers reduce usage, service complaints rise, or the acquired network requires more capital than its utilisation can support. The approximate €1 billion valuation is only an opening number. The decisive record will be written in access agreements, connection rates, churn and cash flow after control changes hands.