Category: Specialist underwriting · Reviewed by Matt Bartlett, Director · Founder · Last reviewed May 2026
A mid-sized property and construction consultancy in the South West notified a £6m claim arising from a series of measured-survey errors across a multi-phase development. The firm carried a £10m programme: a £1m primary policy and a single £9m excess layer written by a different insurer. On paper, the cover looked clean. In practice, the primary aggregate had already been partially eroded earlier in the year by an unrelated matter, the excess wording was silent on a key extension that the primary granted, and the excess insurer’s attachment language required the primary to be “exhausted by payment of loss” — not by erosion through defence costs. Six months into the claim, the firm’s finance director was discovering, line by line, that the £10m tower behaved very differently from what was assumed when the slip was bound. This article sets out, from a specialist broker’s perspective, how PI excess structures actually function and where they go wrong.
A professional indemnity tower is constructed in layers. The primary policy is the ground-up cover: it responds first, subject to its self-insured retention (the firm’s excess), and continues to pay until its limit is exhausted. Above the primary sit one or more excess layers, each attaching at the limit of the policy immediately beneath it. A £20m programme for a quantity surveying practice will frequently use five to seven excess layers, with successive insurers each writing £2m, £3m or £5m bands of cover.
“Umbrella” cover, in the strictest sense, is a different animal. A true umbrella sits above multiple underlying policies (for example PI, public liability and management liability) and drops down to fill specific gaps in any of them. In UK PI, pure umbrella structures are rare; what is more often called “umbrella” is in reality a follow-form excess policy that sits above a primary PI tower. The terminology matters because the legal behaviour of a true umbrella, including its drop-down provisions, is fundamentally different from a straightforward excess layer.
For most professional services firms — solicitors, accountants, surveyors, IFAs, consultants — the relevant architecture is a primary layer plus one or more excess layers written on a follow-form basis. The complexity sits in what “follow form” actually means.
Excess wordings are typically described as “following the form of the primary policy”. This phrase is shorthand and frequently misunderstood. In a following-form excess wording, the excess insurer agrees to indemnify on the same terms and conditions as the primary policy, subject to the excess wording’s own provisions and any specific carve-outs or amendments set out in the excess slip.
Three points need to be drawn out:
First, the excess follows the form of the primary as it stood at inception. It does not automatically follow mid-term endorsements to the primary unless the excess insurer agrees, in writing, to do so. Where the primary insurer agrees to extend cover for a new acquisition or a new service line mid-term, the excess tower may not move with it.
Second, following form does not extend to all components of the primary. Excess wordings will typically not pick up the primary’s own retention, its sublimits, its aggregate provisions (where the excess sits on a different aggregation basis) or its reinstatement mechanism. Each of these is governed by the excess wording itself. A primary with an unlimited reinstatement does not give the firm an unlimited reinstatement on the excess unless the excess wording says so.
Third, where the primary wording grants specific extensions — for example, an extended notification period for circumstances or a mitigation costs extension — the excess policy will only pick those up if it is genuinely silent on the point and the following-form clause is broad. If the excess wording deals with the same subject matter on different terms, the excess wording prevails.
For brokers, the practical question is always: where does the following-form language stop and the excess wording’s own provisions begin? Excess slips need to be read against the primary clause-by-clause, not skimmed.
Drop-down is the provision under which an excess layer responds in place of the primary where the primary has not paid out for some reason. There are three common triggers in theory: where the primary insurer is insolvent; where the primary policy has been exhausted by other claims; and, more controversially, where a claim falls within a gap in the primary policy that the excess layer would otherwise have covered.
The market position in UK PI is largely consistent on this point. Most excess wordings explicitly disclaim drop-down in cases of primary insurer insolvency or gap-filling. The excess will respond when, and only when, the primary limit is exhausted by payment of covered loss. Where the primary insurer is insolvent and has not paid, the excess will not usually drop down — the firm carries the gap, subject to any rights against the FSCS where the primary insurer was an authorised UK insurer eligible for FSCS protection.
This is the area where the broker’s analysis at placement matters most. Three excess wordings might appear similar at headline level and behave very differently in an insolvency scenario. Some excess wordings provide that the excess attaches when the primary is “exhausted by payment of loss otherwise than by the bankruptcy, insolvency, liquidation or refusal to pay of the primary insurer”. Others use language to the effect that the excess attaches on payment by the primary insurer or by anyone on its behalf. The difference is significant for any firm with a substantial tower whose primary insurer’s financial strength rating shifts adversely mid-year.
Mis-stacking is the broker’s term for situations in which the layers of a tower do not align cleanly. The classic example: a primary policy written on an aggregate basis of £1m, with an excess layer of £5m written on an each-and-every-claim basis above it. On a £4m loss, the primary pays £1m, exhausts, and the excess attaches. So far, so straightforward. But two further claims arrive in the same period. On the second claim, the primary aggregate is already exhausted and there is no primary cover left; the excess wording requires the primary to respond first, but there is no primary response available, so the firm’s retention applies again before the excess attaches on each subsequent claim. The firm finds itself paying its primary retention multiple times where it had assumed the aggregate exhaustion of the primary would simply leave the excess sitting at ground level on the next loss.
A second pattern: a primary policy with one reinstatement of the limit, and an excess layer with no reinstatement. After the primary reinstates, there is more primary cover available but no excess sitting above the reinstated limit. The reinstated primary effectively becomes the top of the tower for any subsequent loss.
A third pattern: aggregation. The primary aggregates a series of related matters as a single claim under broad “originating cause” language. The excess wording uses narrower “single act or omission” aggregation. The primary insurer treats fifteen related matters as one claim eroding its limit once; the excess insurer treats them as fifteen separate claims, each of which must independently reach the excess attachment point. The result: the firm believes the excess is engaged, the excess insurer disagrees, and the position turns on a clause that was not negotiated at placement.
A primary policy can be eroded in two ways: by payment of damages or settlement amounts on covered claims, and by payment of defence costs. Whether defence costs erode the limit depends on whether the policy is written on a costs-inclusive or costs-in-addition basis. UK PI policies for most professional services firms are written costs-inclusive — defence costs erode the limit pound for pound.
For excess attachment, this matters because the excess wording needs to recognise erosion by defence costs. Modern excess wordings typically do — they will attach when the primary limit is “exhausted by payment of loss” with “loss” defined to include defence costs. Older excess wordings, or those drafted by reinsurance-style departments unfamiliar with primary PI mechanics, sometimes use narrower language. The distinction between “exhausted” and “settled down to” is also material: some excess wordings provide that the excess attaches when the primary insurer admits the claim and settles down to its limit, others require actual payment.
A specific scenario the broker market sees repeatedly: a claim is reserved at £900k, defence costs of £150k are incurred, and the primary insurer offers to settle at policy limit on a “settle down to limit” basis. The excess insurer, on a strict “payment of loss” attachment, may not engage until the primary has actually paid out. In practice, primary and excess insurers will normally cooperate on cash flow, but the wording sets the firm’s legal position if cooperation fails.
Market standard in the UK PI primary market is one reinstatement of the limit per policy period, often subject to insurer agreement and sometimes subject to an additional premium. Where the primary reinstates, the excess wording does not automatically reinstate with it. An excess layer with no reinstatement provision, written above a primary with one reinstatement, gives the firm two tranches of primary cover but only one tranche of excess cover.
Where the firm wants matching reinstatement across the tower, this needs to be negotiated layer by layer. Some excess insurers will agree to follow the primary’s reinstatement mechanism; others will not. For firms with a higher claims frequency profile — IFA practices with significant complaint volumes, for example — the reinstatement architecture across the tower may matter more than the headline limit.
Beyond the structural issues set out above, excess wordings frequently contain provisions that diverge from the primary in ways the firm has not appreciated. A short list:
The excess wording is silent on extensions granted by the primary. Mitigation costs, civil fines and penalties cover, court attendance costs, public relations costs — all of which the primary may grant — may not be picked up by the excess unless the following-form language is broad enough.
The excess wording uses a narrower definition of “insured”. The primary may include former partners and consultants; the excess may not, or may do so only for acts committed during specified periods.
The excess wording uses different aggregation language. As above, this is the most material divergence and the one that produces the largest litigation downstream.
The excess wording has its own notification provisions that operate independently of the primary’s. Most excess wordings require separate notification once the claim is likely to penetrate the excess layer. Late notification to the excess insurer, even where the primary was notified promptly, may give the excess insurer remedies.
The excess wording has its own dispute resolution and jurisdiction clauses. A primary may be subject to English law and English court jurisdiction; an excess layer placed in a different market may attract different law.
A quota share arrangement is, structurally, very different from a tower of excess layers. In quota share, two or more insurers each take an agreed proportion of the same ground-up risk. A 50/50 quota share on a £5m primary means each insurer is liable for 50% of every loss from the ground up, up to £2.5m each.
In a tower, the primary insurer pays first and the excess insurers do not engage until the primary is exhausted. Each excess insurer is liable for losses falling within its specific layer only.
Quota share is uncommon in UK PI except on the hardest-to-place risks — large law firms with complex international exposures, certain financial advice firms with material historical claims activity, and run-off accounts where no single market is willing to write the full primary line. Where quota share appears, the firm’s claims experience is more complex to administer: every loss involves multiple insurers from the ground up, claims-handling agreements need to be drafted to determine which insurer leads, and excess attachment above a quota-shared primary requires careful wording.
Placing a multi-layered PI tower is a coordination exercise. The primary insurer’s wording sets the framework. Each excess insurer then offers a quote on a following-form basis, subject to its own subjectivities. Subjectivities at one layer — for example, a requirement to see updated revenue figures or a specific underwriting question — need to be aligned with the others so that the whole tower can incept at the same date.
Side-letter agreements between insurers in the tower are sometimes used to clarify aggregation, drop-down and reinstatement positions where the standard wordings are ambiguous. These need to be reviewed by the broker, not signed insurer-to-insurer behind the firm’s back.
Following-broker structures appear at higher layers. A leading broker negotiates terms with the lead insurer in each layer, and following brokers place the remaining proportion of the layer with subscribing insurers on the same terms. Where the firm is dealing with a single placing broker, this happens behind the scenes; where multiple brokers are involved, coordination is more complex.
The broker’s role in tower placement is not only to obtain terms but to test the wordings against likely loss scenarios before binding. A wording that looks acceptable on a single-claim review may produce mis-stacking under multi-claim scenarios.
An excess layer sits above a specific underlying policy and attaches when that underlying policy’s limit is exhausted. An umbrella policy, in its purest form, sits above multiple underlying covers and can drop down to fill gaps in any of them. UK PI towers almost always use excess layers rather than true umbrellas; the term “umbrella” is sometimes used loosely to describe high-attachment excess cover, which can cause confusion.
In most UK PI excess wordings, no. The excess typically requires the primary to be exhausted by payment of covered loss, and explicitly disclaims drop-down where the primary fails to pay because of insolvency. FSCS protection may apply if the primary insurer was UK-authorised. Firms should review excess wordings carefully on this point and consider the financial strength of the primary insurer as part of placement.
The excess agrees to indemnify on the same terms and conditions as the primary as at inception, subject to the excess wording’s own provisions. It generally does not pick up the primary’s retention, sublimits, reinstatements or any extensions granted in different language to that used in the excess wording. Mid-term endorsements to the primary do not automatically extend to the excess.
Most UK PI primary policies are written costs-inclusive, so defence costs erode the primary limit. Excess wordings will typically pick up “loss” as including defence costs for attachment purposes, but the precise language matters: some require the primary to be “exhausted by payment of loss” (cash out), others permit attachment when the primary insurer “settles down to” its limit.
Mis-stacking is the situation where the layers of a tower do not align cleanly — different aggregation bases, different reinstatement provisions, narrower definitions of insured at higher layers. It can be reduced by ensuring excess wordings are reviewed clause-by-clause against the primary, by negotiating matching aggregation language, and by stress-testing the tower against multi-claim and high-quantum scenarios at placement.
No. Reinstatement provisions are governed by each layer’s own wording. Where the primary reinstates after a loss, the excess does not reinstate unless the excess wording says so. Firms wanting matching reinstatement across the tower need to negotiate that with each excess insurer at placement.
Quota share is unusual in UK PI and tends to appear only on the hardest-to-place risks — large or complex firms, accounts with significant historical claims, run-off arrangements. It is structurally different from a tower: two or more insurers share the ground-up risk proportionally rather than one sitting above the other.
Most do. Even where the primary has been notified promptly, excess insurers typically expect separate notification once a claim is reasonably likely to penetrate the excess layer. Failure to notify the excess in accordance with its own provisions can produce coverage arguments at the excess level, even where the primary position is secure.
Apex Insurance Brokers Ltd is an independent UK insurance broker based in Bristol, advising professional services firms on professional indemnity insurance and related covers. The firm is authorised and regulated by the Financial Conduct Authority (firm reference 724952) and registered at Companies House (company number 07014570).
This commentary reflects market conditions as at May 2026 and is provided for general information. Insurance market conditions, policy wordings and regulatory positions change frequently; firms should obtain advice specific to their circumstances rather than rely on general commentary.
Apex Insurance Brokers serves UK professional services firms and commercial businesses. Call 0117 325 0027, email hello@apexinsurancebrokers.co.uk, or request a quotation.
Get a quote