Restructuring a Layered PI Programme at Renewal

Very few substantial PI programmes were drawn on a clean sheet of paper. The usual history runs like this: the firm bought a primary policy years ago, a client contract demanded a higher limit, an excess layer was bolted on, then another, and perhaps a third after a merger. Each addition was rational at the time. The result, five or ten years on, is a tower that nobody has looked at as a whole — layers priced in different market conditions, placed with whichever insurers were available that week, on wordings that were never read side by side.

Renewal is the natural moment to correct that, because it is the one point in the year when every layer is repriced and every insurer recommits. Below is what a proper restructure actually examines. None of it requires the firm to move a single policy; sometimes the right answer after the analysis is to change nothing. But the analysis itself is rarely wasted.

Layer pricing relativities: does the tower price make sense?

In a well-priced tower, each layer costs less per pound of limit than the one beneath it, because it is further from the claims. The rate falls as you climb. When a programme has grown by accretion, that curve is often distorted: an excess layer bought in a hard market may still be priced as though capacity were scarce, while the primary beneath it has softened. A layer that costs nearly as much as the one below is a signal — either the market has moved and the incumbent has not, or the layer is with an insurer whose appetite for the risk has cooled and whose price is a polite way of saying so.

The exercise is straightforward arithmetic once the schedule is assembled in one place: premium per million of limit, layer by layer, compared against where the market currently prices comparable attachment points. It cannot be done from the invoices alone, because the invoices arrive at different times from different brokers and nobody totals the column. Assembling that single view is often the most useful single page a review produces.

Attachment points against real claims experience

Attachment points — where each layer begins to pay — were usually set by what the market offered on the day, not by anything about the firm. Ten years of claims history changes that. If every notification the firm has ever made resolved comfortably within the primary limit, the question is whether the primary is carrying the right share of the programme, and whether the excess layers are structured efficiently above it. If, on the other hand, one or two matters have pushed into the first excess layer, the pricing of that layer — and the willingness of its insurer to stay — deserves close attention before renewal terms arrive.

Claims experience also informs the self-insured excess at the very bottom. A firm with strong risk management and a clean record may be paying for ground-level cover it has never used, while a firm whose deductible was set when it was half its current size may be carrying a retention that no longer reflects its balance sheet either way.

Follow-form consistency: the silent gaps in the tower

The most dangerous defects in an accreted tower are not in the pricing but in the wordings. Excess layers are usually sold as “follow form” — they promise to respond on the same terms as the primary. In practice, many excess wordings follow the primary except where they don’t: a narrower definition of claim, an added exclusion, a different notification clause, a lower costs provision. Because the layers were placed at different times, possibly by different brokers, nobody has ever read them against each other.

The consequence is a silent gap: a claim the primary pays but an excess layer does not, discovered at precisely the moment the excess layer matters. Checking follow-form consistency is slow, unglamorous work — clause-by-clause comparison across every layer — and it is the part of a restructure most often skipped. It should never be. A tower whose layers respond differently to the same claim is not really a tower at all; it is a stack of separate arguments waiting to happen.

Panel concentration and whether the limit still fits the contract book

Two structural questions complete the picture. First, insurer concentration: if one carrier sits on multiple layers, or one group’s paper appears throughout the tower under different names, the firm is more exposed to a single change of appetite than the schedule suggests. Diversity across the panel is a form of resilience — it keeps competitive tension in the programme and means no single withdrawal forces an emergency re-placement.

Second, the total limit. The tower’s height was set by the largest contractual requirement at some point in the past. Contract books move. If the firm’s engagements have grown — larger projects, bigger counterparties, aggregation of exposure across a framework — the limit that satisfied the biggest contract of five years ago may be quietly inadequate for the biggest contract of today. The reverse also happens: firms carry limits demanded by a client relationship that ended years ago. Either way, the limit should be re-derived from the current contract book, not rolled forward by habit.

Restructure, re-market, or hold — and the timetable that makes each possible

The analysis leads to one of three broad outcomes. Hold: the structure is sound, relativities are fair, wordings align — renew with the incumbents and negotiate on evidence. Re-market: the structure is right but one or more layers are mispriced or with the wrong carrier — test those layers competitively while keeping the rest stable. Restructure: the tower itself needs rebuilding — different attachment points, consolidated or split layers, a changed retention — which usually means presenting the whole programme afresh to the market.

All three options depend on time. A larger firm should open its renewal three months out at minimum: month one for the analysis above and the assembly of a proper submission; month two for market approaches and underwriter meetings; month three for structuring, negotiation and documentation, with room to absorb a surprise. A programme that starts six weeks before renewal has already chosen “hold” — whatever the analysis might have said — because no serious alternative can be built in the time remaining. The single most common reason substantial firms overpay is not a bad broker or a hard market; it is a timetable that removed every option except renewal as expiring.

Frequently asked questions

Does restructuring mean changing insurers?

Not necessarily. A restructure can be executed entirely with the incumbent panel — moving attachment points, rebalancing layers, aligning wordings — and incumbents will often engage constructively because the alternative is a full re-marketing. Changing carriers is one available tool, not the definition of the exercise.

Is there a risk in disturbing a programme that seems to be working?

There can be, which is why the analysis comes before any market approach. Continuity has real value in a claims-made class: retroactive dates, known claims handling, established relationships. A good restructure weighs what would be given up against what would be gained, and sometimes concludes the programme should be left alone — but on evidence, not inertia.

How far ahead of renewal should we start?

Three months at minimum for a layered programme, and longer where a genuine restructure or full re-marketing is in prospect. The work in the first month — assembling the schedule, the claims record and the contract-book analysis — requires no commitment to change anything, so there is no reason to delay it.

Apex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority (FRN 724952). Every risk is different: nothing on this page is advice on your own programme, and outcomes depend on your firm’s circumstances and the market at the time.