Promises in the Ledger: Mapping Contingent Commitments Across a UK Group
Most UK group boards can recite the term and covenant profile of their principal banking facilities. Far fewer can say, with confidence, how many parent company guarantees their group has given, which subsidiaries stand behind them, or when the next material undertaking falls due for renewal. The gap is rarely caused by negligence. It grows from ordinary delegation: a guarantee is signed in a subsidiary, recorded in a completion bible, mentioned in a paper that has long since left the agenda, and then absorbed into the background of group life.
That is why a contingent commitment audit is less a legal exercise than an exercise in memory and sight. The aim is not to turn directors into guarantors-in-chief. It is to give them a single, dependable view of what the group has promised, to whom, on what terms and for how long — without burying the board in the fine print of every instrument.
Why commitments drift
Parent company guarantees, letters of comfort, performance bonds and cross-entity undertakings tend to arise at moments of activity: a tender, a lease, a banking arrangement, an acquisition, a customer framework or a pension discussion. Each is sensible in context. The difficulty is that the context passes. The deal team moves on, the subsidiary changes finance director, the counterparty changes name, and the instrument remains live.
The board pack may show debt, cash and covenant headroom. It seldom shows the full population of contingent exposures. Yet these commitments can matter precisely when the group is under strain: a guarantee may be called because a subsidiary has failed; a comfort letter may be relied upon in a dispute; a bond may be drawn down after a performance issue. The risk is not only financial. It is the risk of being asked a direct question in a board meeting and discovering that no one holds a complete answer.
Begin with the instruments, not the filing cabinet
A practical audit starts by agreeing the categories. Parent company guarantees are the clearest: one group company promises to meet another's obligation if it does not. Letters of comfort are less uniform and should not be treated as a single class; their effect depends on wording, context and the law governing them. Performance bonds and surety arrangements secure delivery or completion. Cross-entity undertakings may sit inside intra-group service arrangements, property documents, procurement contracts or financing packages. Indemnities and standby instruments may also create exposure that behaves like a guarantee.
The point is not to produce a legal taxonomy for its own sake. It is to ensure the sweep is broad enough. A register that captures only documents labelled 'guarantee' will miss the very commitments that tend to be overlooked.
A register built for directors
The central tool should be a register, but not a replica of the legal file. It should answer the questions a director is likely to ask. For each commitment, record the issuing entity, the beneficiary, the instrument type, the underlying obligation, the maximum amount or cap, the trigger or conditions, the expiry or review date, the governing law, the responsible owner and the board or committee approval reference.
Useful additions include whether the commitment is secured, whether it is subject to automatic renewal, and whether it is linked to a facility, contract or project that is itself under review. A simple status field — active, expired, released, disputed or under negotiation — prevents the register becoming a museum of old paper. Traffic-light ratings for value and trigger likelihood can help the board scan the population without mistaking a rating for legal advice.
Where to look
No single department will hold the full picture. Treasury and finance will know banking guarantees and facility-linked support. Legal and company secretarial teams should hold completion documents and board approvals. Procurement and commercial teams may hold bonds, performance undertakings and customer framework commitments. Property, pensions, tax and customs teams may hold instruments that never cross the group finance desk. Subsidiary managing directors should be asked directly, because local arrangements often sit outside central visibility.
It is also worth reviewing board and committee minutes, acquisition completion files, intra-group agreements, bank facility documentation and any register of delegations. The question to put to each entity is straightforward: list every instrument under which this company, or another group company, could be asked to pay or perform because of another party's obligation. Ask for the document, not just a description.
Four questions for the board
Once the register exists, the board does not need to read every entry. It needs to ask four questions. First, what is the aggregate maximum exposure, and how does it compare with the group's capacity to absorb a call? Second, which commitments are live, material or close to renewal? Third, which are subject to triggers that could arise quickly — a covenant breach, a contract termination or a subsidiary's liquidity event? Fourth, who owns each entry, and when was it last reviewed?
These questions turn a legal schedule into a governance instrument. They also expose gaps in delegated authority. If a subsidiary can issue a guarantee without central awareness, the issue is not the register; it is the approval route.
A rhythm, not a one-off
The first audit will be imperfect. Documents will be missing, descriptions will vary, and some instruments will need counsel's view. That is normal. The objective is a repeatable rhythm: a quarterly review of material entries, an annual attestation from each entity, and a gate through which new guarantees, comfort letters and bonds pass before signature. The register should be concise enough to be maintained and authoritative enough to be trusted.
For UK groups, this is a modest discipline with a clear return. It gives directors a single view of contingent commitments, reduces the chance of an unwelcome surprise, and makes the group's promises visible while they can still be managed. A guarantee is not a failure of control. An unrecorded guarantee is.