SBF season is now an operating model exam

What it is

On 1 September, Phase 1 syndicates submitted their 2027 year of account plans through the Syndicate Business Forecast. It is the most consequential paperwork in the London Market calendar – the SBF drives capital, stamp capacity and Lloyd's oversight categorisation for the year ahead. But the mechanics this year are not the story. The story is what the plan is being tested against, and who is doing the testing.

Three things changed since the last planning cycle. First, the claims hurdle principle, effective from 1 January after a year of soft implementation, made claims management the fifth fundamental under Lloyd's Principles Based Oversight – and under PBO rules, no syndicate can be rated higher than its lowest-rated hurdle. Second, the Corporation stated explicitly that given market conditions, all syndicates should expect an uplift in core oversight for the 2027 cycle. Third, the PRA published a supervision priorities letter that names the exact failure mode it is hunting: internal model assumptions about future underwriting performance that are more optimistic than a firm's own track record, in a softening cycle, with solvency capital requirements at stake.

Read together, these are not three separate pressures. They are one test aimed at the same thing: whether the numbers in your plan can be traced to the reality of how your firm operates.

Why it matters now

A business plan used to be judged on its ambition and its plausibility. A well-argued SBF, a sensible growth story, an experienced underwriter vouching for the book – that was enough to get through the season, and the detail could be reconciled afterwards. That era is ending.

The hurdle principle changes what "planning" means because claims performance is now a plan-level input, not an operational afterthought. A syndicate that wants to grow a class must now be able to show claims outcomes – service quality, maturity of handling, governance over delegated claims arrangements – not just claims intentions. As a Lloyd's executive put it to Insurance Times: "there's a louder voice at the table now because we have an ability to impact whether or not business plans are approved."

The PRA's interest is sharper still. Its letter observed that softening is "particularly pronounced in some wholesale lines in the London Market," and that some firms' internal model assumptions about future profitability look rosier than their history supports. Where the gap between actual and assumed is widest, the regulator says it will engage – and consider further supervisory action where justification is lacking. Under Solvency UK, understated SCRs are not a reporting nuisance; they are capital adequacy.

Here is the uncomfortable part. Both tests are traceability tests. The PRA cannot be answered with a better narrative; it will want to see actuals against assumptions, period by period, class by class. Lloyd's cannot be answered with a claims transformation slide; it will want to see outcomes data. And both will keep asking the same follow-up question: where did this number come from?

For a firm whose plan data lives in a chain of spreadsheets – plan premium from the pricing team, expenses from finance, loss ratios from reserving, assembled by a group of people working through August – that question is the beginning of a very long September.

What firms should do

The firms that will find this season routine are the ones that treat the SBF as an export, not an assembly job. Three differences stand out.

Trace every plan line to a source process

Every number in an SBF should have an origin: the pricing model or tool that produced the plan premium by class, the finance system that produced the expense trajectory, the reserving process that produced the loss ratio assumptions, the capital model that produced the capacity ask. If a number cannot be traced in under an hour, it is an opinion wearing a plan's clothing.

Close the actuals-versus-assumptions loop quarterly, not annually

The PRA's concern is not that firms make optimistic assumptions; it is that assumptions drift away from what the firm is actually achieving, unnoticed. A syndicate that tracks planned versus actual performance – premium, loss ratio, expense ratio, claims maturity – as a standing management rhythm, not a pre-season exercise, will see its own optimism gap before the regulator does.

Map claims as an operating system, not a department

The hurdle principle rates claims management across its whole architecture: who handles what, which third parties touch which processes, what controls sit at each handoff, how outcomes are measured. Firms that know their claims operating model – mapped, owned, evidenced – will be able to demonstrate what the principle requires. Firms that know their claims function only as an org chart and a complaints record will not.

The opmodal perspective

We have spent years arguing that an operating model is not a document – it is the live, connected picture of how an organisation actually works: processes, systems, people, risks and controls in one version-controlled environment. This planning season is the first time the market's own machinery has been tuned to reward exactly that view.

The claims hurdle principle works because claims is the most honest window into an organisation. An underwriting plan can be aspirational; a claims outcome happened. The PRA's optimism-gap concern is the same idea viewed from the capital side: ambition must be tethered to demonstrated reality. Both are, in effect, operating model audits. Both punish firms whose understanding of themselves is distributed across spreadsheets, inboxes and people's heads – because when the questions come, there is nowhere to point.

The syndicates that thrive in this regime will not necessarily be the ones with the best underwriters. They will be the ones who can show their work: where each number came from, which process produced it, who owns it, and what changed since last quarter. That is not a planning capability. It is an operating model – and it compounds.