Large Loss Register: Template for One View of Big Claims
On this page
- Key Takeaways
- What Is a Large Loss Register?
- What Should Trigger Entry to the Large Loss Register?
- What Fields Does a Large Loss Register Need?
- How Does One Loss Look in the Register?
- How Do You Keep the Register Current?
- Where Does Insurnest Fit for Large Losses?
- How Should a CRO Start the Register This Quarter?
- Key Terms
- Sources
- Frequently Asked Questions
Last reviewed: November 2026
The biggest claims are often tracked in the most scattered way, with an email here and a spreadsheet there.
When the board asks what one event will cost net of reinsurance, the answer takes days.
A large loss register is one controlled list of every claim above an agreed trigger, with its estimate history, treaty split and expected recoveries.
It gives claims and risk leaders one live view of big claims across treaties.
Key Takeaways
- Set triggers against retentions. A flat amount misses claims that matter to specific treaties.
- Keep every estimate. The history of a loss tells you more than its latest figure.
- Link each loss to every treaty it touches. Recoveries, reinstatements and notices follow from that link.
- Give every entry an owner and a review date. An unreviewed entry is a stale entry.
What Is a Large Loss Register?
A large loss register is a single, controlled list of the claims that could move results, reserves or reinsurance recoveries.
It sits above the claims system, not instead of it.
The register is a management artefact. It answers three questions: what are our big losses, what are they worth now, and what will we recover?
What Should Trigger Entry to the Large Loss Register?
Entry to the large loss register should be triggered by size relative to retention, and by features that make a claim risky whatever its size.
| Trigger | Illustrative rule | Why it matters |
|---|---|---|
| Size against retention | Latest gross estimate at or above 50% of the per risk retention | Catches claims before they reach the layer |
| Event | Any claim coded to a named catastrophe or occurrence | Losses aggregate across policies |
| Litigation | Suit filed or external counsel appointed | Estimates can move sharply |
| Coverage dispute | Reservation of rights or declinature under review | Recovery may be contested |
| Notification clause | Treaty requires notice at a set threshold | Late notice can put recovery at risk |
| Accumulation in group life | Several deaths from one event in one scheme | Catastrophe cover may respond |
Why set the trigger against retention rather than a flat amount?
A trigger set against retention works because the same claim size means different things on different treaties.
A USD 2m claim is routine under a USD 10m retention. Under a USD 2.5m retention, it needs watching now.
What Fields Does a Large Loss Register Need?
A large loss register needs four groups of fields: identification, financials, treaty links and controls.
The four tables below form a reusable template.
Which identification fields come first?
Identification fields come first because every later figure must attach to one unambiguous loss.
| Field | Purpose |
|---|---|
| Register ID | Unique key that never changes |
| Cedent claim reference | Links to the cedent's file and bordereaux |
| Event code | Groups claims from one catastrophe or occurrence |
| Date of loss | Tests the treaty period and any hours clause |
| Class and cause | Supports aggregation and trend review |
| Territory | Tests territorial scope and currency |
| Insured or member reference | Identifies the original policy or group member |
| Status | Open, reopened, closed or disputed |
Which financial fields should the register hold?
Financial fields should hold the latest position and enough history to explain how it got there.
| Field | Purpose |
|---|---|
| Currency | Original currency of the loss |
| Paid to date | Cash already paid on the claim |
| Outstanding reserve | Remaining case estimate |
| Gross incurred | Paid plus outstanding |
| First estimate and date | Baseline for measuring creep |
| Previous estimate and date | Shows the latest movement |
| Reason for movement | One line on why the estimate changed |
| Net retained | Gross incurred less expected recoveries |
How should the register link a loss to treaties?
The register should link each loss to every treaty it touches, with the layer and share that apply.
| Field | Purpose |
|---|---|
| Treaty ID and layer | Identifies the cover that responds |
| Retention and limit | Drives the ceded amount |
| Ceded amount by treaty | Expected recovery per cover |
| Reinstatement premium due | Cost of restoring the limit |
| Notice date to reinsurers | Evidence the notification terms were met |
| Retrocession impact | Onward recovery where it applies |
Notice dates matter most where a claims cooperation clause sets conditions on recovery.
Which control fields keep the register honest?
Control fields keep the register honest by showing who owns each entry and when it was last challenged.
| Field | Purpose |
|---|---|
| Claim owner | Named handler accountable for the entry |
| Last review date | Shows whether the entry is current |
| Next review date | Drives the review cycle |
| Approver of last movement | Evidence of maker-checker control |
| Lead arrangement | Who leads and who follows on the claim |
| Dispute flag | Marks contested coverage or recovery |
In the Lloyd's market, the Claims Lead Arrangements provide for a lead and second lead to act jointly on complex claims.
Recording who leads saves time when a large loss needs a decision.
How Does One Loss Look in the Register?
One loss in the register reads like a short story told in figures.
This hypothetical property claim sits under a USD 15m xs USD 5m per risk excess of loss treaty, fully placed.
| Date | Gross incurred (USD m) | Ceded to treaty (USD m) | Net retained (USD m) | Reason for movement |
|---|---|---|---|---|
| Month 1 | 6.0 | 1.0 | 5.0 | First notice |
| Month 3 | 12.0 | 7.0 | 5.0 | Adjuster's report |
| Month 6 | 18.0 | 13.0 | 5.0 | Business interruption added |
| Month 9 | 21.0 | 15.0 | 6.0 | Layer exhausted |
The ceded amount is the gross loss above USD 5m, capped at USD 15m. At month 9, that is USD 16m capped to USD 15m, leaving USD 6m net.
The latest estimate is 3.5 times the first. Net retained stayed at the retention until the layer ran out.
A register showing only the latest figure would hide both facts.
Why keep the estimate history rather than overwrite it?
Estimate history matters because the pattern of revisions is an early warning in its own right.
In June 2024, Reinsurance News reported Swiss Re's call for an industry focus on European catastrophe loss creep.
One Italian hail estimate it cited rose from USD 2.2bn to USD 6bn.
Swiss Re pointed to a lack of up-to-date exposure and risk value data as a key cause. A register that keeps first estimates lets you measure your own creep.
How Do You Keep the Register Current?
The register stays current only when updates are tied to events and reviews are tied to a calendar. The timings below are illustrative.
| When | Who | What |
|---|---|---|
| Any estimate movement above an agreed amount | Claim owner | Update financials and reason within 2 working days |
| Monthly | Claims team leads | Review every open entry and confirm the next review date |
| Quarterly | Large loss committee | Challenge movements, recoveries and disputes |
| Quarter-end close | Finance and actuarial | Reconcile register totals to the claims ledger |
| Treaty renewal | Reinsurance buyer | Use the register's history in renewal data |
The quarter-end check matters most. The register and the reinsurance claims ledger must tell the same story.
Who should sign off on a movement?
A movement above an agreed amount should be signed off by someone other than the person who entered it.
Our guide to maker-checker in reinsurance claims explains how to set that control up.
Where Does Insurnest Fit for Large Losses?
Insurnest fits large loss work for group life and credit life reinsurance claims, which is what the Reinsurance Claim Core Platform is built for.
The platform registers claims against the relevant treaty and member census and attaches supporting documents.
It applies configured treaty, eligibility, validation and claim-quantum rules. Approvals follow maker-checker and configured authority limits.
It configures treaty structures, counterparties, shares and retrocession, and its ledger tracks retrocession.
Original bookings and later adjustments stay visible as separate movements, which is the estimate history this template asks for.
Structured claims and period reporting includes closing balances by currency, much like a period movement report.
For property and casualty large losses, the template still applies, but the platform is not positioned for those classes.
What will the register still need people for?
The register will still need people for every judgement call: reserve adequacy, coverage positions and whether claims aggregate into one event.
Authorised claims professionals make final decisions. The platform records and controls those decisions, but does not make them.
How Should a CRO Start the Register This Quarter?
A CRO can start the register this quarter without waiting for a new system. Five steps are enough.
- Agree the triggers. Set entry rules per treaty, relative to retention.
- Load the current population. Add every open claim that meets a trigger today.
- Back-fill first estimates. Capture the first estimate and its date for each entry.
- Name an owner per entry. No entry goes live without a person and a next review date.
- Run the first committee. Review the top ten by movement, not just by size.
Key Terms
- Large loss: A claim that crosses an agreed size or risk trigger for closer oversight.
- Loss creep: The upward drift of a loss estimate after the first report.
- Retention: The amount kept before an excess of loss treaty responds.
- Reinstatement premium: The premium paid to restore an excess of loss limit after a loss.
- Event code: A reference that groups claims arising from one catastrophe or occurrence.
Sources
- Reinsurance News: Swiss Re calls for industry focus on European catastrophe loss creep issue (5 June 2024)
- Lloyd's: Lloyd's Claims Lead Arrangements
- Insurnest: Reinsurance Claim Core Platform
Frequently Asked Questions
What is a large loss register?
A large loss register is a single, controlled list of every claim above an agreed size or risk trigger, across all treaties and classes. It records the loss, its estimate history, the treaties it touches, expected recoveries and the owner, so claims and risk leaders see big exposures in one place.
Why do reinsurers and cedents need a large loss register?
Reinsurers and cedents need a large loss register because big claims drive volatility, reserve movements and reinsurance recoveries. Without one view, the same loss can be tracked differently by claims, actuarial and finance teams, and estimate creep or a missed treaty notification is spotted too late.
When should a claim enter the large loss register?
A claim should enter the large loss register when it crosses a trigger set relative to the relevant retention, such as half the excess of loss attachment, or when it is litigated, coverage is disputed or it may aggregate into an event. Entry should be based on the latest estimate, not paid amounts.
Who should own the large loss register?
The large loss register is best owned by the Chief Claims Officer, with named claims handlers updating each entry. The CRO uses it for exposure and reserve oversight, finance uses it for recovery and cash planning, and a large loss committee reviews movements on a fixed cycle.
Which metrics should a large loss register report?
A large loss register should report the number of open large losses, total incurred and movement since the last review, the ratio of current to first estimate, expected recoveries by treaty, entries without a review in 90 days, and notices outstanding to reinsurers. Each metric needs a named owner.
Can automation maintain a large loss register?
Automation can maintain much of a large loss register by flagging claims that cross triggers, recording every estimate as a dated movement and linking losses to treaties. Reserve judgement, coverage positions and aggregation decisions still need experienced claims professionals, so automation should feed the register rather than decide what it says.
How does a large loss register differ from a claims ledger?
A large loss register differs from a claims ledger in purpose. The ledger records every financial movement on every claim for accounting, while the register is a management view of the few claims that matter most, adding estimate history, treaty impact, recovery expectations, ownership and review dates.
Should a large loss register be a spreadsheet or a system?
A large loss register can start as a controlled spreadsheet if it has fixed fields, one owner and dated estimate history. Once it must link claims to treaty shares, approvals and retrocession, a claims system with structured records is safer, because spreadsheets lose history when cells are overwritten.

Hitul Mistry
CEO, Insurnest
An InsurTech leader with more than a decade of experience across insurance and technology, focused on solving business problems with the help of technology. Has worked with brokers, insurance carriers, and reinsurance firms across the India, UAE, and US markets.
View LinkedIn profile →