One must start with a fact almost no buyer has explained to it: the underwriter does not examine the target. It has no access to it, neither the time nor the mandate, and it verifies nothing itself. It reads the buyer's due diligence reports, holds one or two calls with the advisers who wrote them, and decides. What it underwrites is therefore not a company's risk, it is the quality of a piece of professional work. That inversion governs everything else, and it explains underwriting decisions that look absurd for as long as one believes the underwriter is assessing the target.
The first consequence is legal and is handled early or not at all. The reports must be capable of being relied on by the insurer, either by being addressed to it or through a reliance letter opening their benefit to it. This is an act that advisers charge for, frame carefully, and sometimes refuse, particularly specialist firms unwilling to extend their liability beyond their own client. Yet reliance is no formality for the underwriter: it is what gives it recourse against the author of the report if that report was wrong. A report it cannot rely on is worth, from its point of view, a report that does not exist, and it will treat the subject as unexamined.
The second consequence concerns the form of the reports and is counterintuitive. What creates cover is not the absence of problems, it is written proof that a subject was examined. A report stating that it reviewed all forty-three material contracts and found a change of control clause in only one creates cover on change of control: the scope is stated, the result is clean, the underwriter knows what it is insuring. A report concluding that no material issues were noted creates almost none, because nothing in it says whether the subject was looked at. Exception-based reports, very common because they are short and cheaper, are for this reason the least useful to underwriting, and nobody warns the buyer at the moment it commissions them.
The underwriting call deserves to be understood for what it is, since advisers often attend it as though it were an information meeting. The underwriter asks open questions about scope and about anything that struck it as irregular. It is looking for two things. First, what the teams know, and what knowledge of a fact entails is covered elsewhere. Second, and this is what is really at stake, the sentences it will need to draft its exclusions. An adviser who answers that it did not get into a subject has, at that precise moment, written an exclusion. The call is minuted, those minutes are underwriting information, and they will be reread at the time of a claim.
One interaction is worth knowing because it decides the product's real value. Sale agreements frequently qualify warranties by a materiality threshold, and the policy may strip those qualifiers away, so that cover applies as though the warranty were unqualified. That is a considerable enhancement, and it is genuine. But it takes effect only on the warranties the review supports: stripping materiality widens what counts as a breach, while the exclusions drawn from the review narrow the places where one can occur. The two mechanisms pull in opposite directions, and their product is computed subject by subject, never globally.
Finally comes a timetable constraint that is not an accident but the very structure of these deals. Underwriting takes place in the last ten days, in parallel with final negotiation, and the underwriter reads drafts of reports that are not final. If the final version adds a point it had not seen, its position changes or a specific exclusion appears forty-eight hours before signing, at the moment when nobody has any room left. The order in which reports are delivered relative to the close of underwriting therefore has a cash value, and that order is settled by a project manager running a calendar, not by a risk decision.
What a buyer should steer is therefore not a request for wider cover, which is not obtainable, but four steps taken early. Set the review scope against the map of exclusions rather than against the budget alone. Ask advisers for positive statements of scope, which costs a few pages rather than fees. Obtain reliance in the engagement letter, while it is still negotiable, and not nine days before signing. And on subjects where reliance is refused, treat the exclusion as settled and handle it in the deal price, rather than discovering it at signing with no choice left but to sign anyway.
An acquirer buys a software publisher at an enterprise value of 180 million euros. The policy carries an 18 million limit and a 900,000 euro retention. Legal, financial and tax reviews are run by firms granting reliance to the insurer. The technology review, covering among other things compliance of open source components embedded in the product, is entrusted to a specialist firm whose engagement letter excludes any third-party reliance; that firm offers a reliance letter for 40,000 euros. Nine days before signing, the deal team refuses that unbudgeted spend. At the underwriting call, the underwriter asks about open source components; it is told a report exists but is not available to it. The final policy excludes compliance with open source and third-party code licenses. Eleven months after closing, a copyleft obligation is found on a component at the core of the product: re-engineering and legal costs come to 6.2 million. Where was this loss decided?
It was decided nine days before signing, on a 40,000 euro line of spend, and the mechanism is the one nobody explains to buyers. The exclusion does not come from a gap in the work: the subject had indeed been examined. It comes from the work not being available to the underwriter, and an underwriter cannot underwrite a report against whose author it could not turn, because reliance is precisely what gives it that recourse. Refusing the 40,000 euros therefore saved nothing: it turned an 18 million policy into an 18 million policy with a hole located exactly where the target's principal asset sat. The tempting reading is to turn against the specialist firm, whose report had rated the component low risk while recommending a follow-up check that was never done. That is a false trail and it should be closed early: recourse against an adviser is bounded by the liability cap in its engagement letter, ordinarily a multiple of fees, that is a few hundred thousand euros at best against 6.2 million. Recourse does not replace cover, it softens it at the margin. What should have been done required no legal subtlety: on the subject carrying the target's value, reliance is not an expense, it is the condition of the cover existing, and it is negotiated in the engagement letter several weeks earlier, when it costs almost nothing.
- 01The underwriter does not examine the target, it reads the buyer's reports: what it underwrites is the quality of professional work, not a company's risk.
- 02What creates cover is written proof that a subject was examined: 'no material issues noted' creates almost none.
- 03A report the insurer cannot rely on is, to it, a report that does not exist, and refusing a reliance letter manufactures an exclusion.
- 04At the underwriting call, an adviser who says it did not get into a subject has just written an exclusion, and the call is minuted.
- 05Recourse against an adviser does not replace cover: it is bounded by the liability cap in its engagement letter.