A loss development triangle shows how losses for each accident year develop across successive reporting dates.
Insurer disclosures commonly present cumulative incurred losses or cumulative paid losses with accident years down one dimension and evaluation years across the other. The resulting triangular shape exists because recent accident years have fewer years of development history.
Reading a triangle
Imagine an insurer initially estimates ultimate incurred losses for accident year 2023 at $900 million. It later reports:
- 2023 evaluation: $900 million;
- 2024 evaluation: $950 million;
- 2025 evaluation: $925 million.
The first revision is $50 million of unfavorable development. The second is $25 million of favorable development relative to the prior estimate.
The final number is still not guaranteed to be ultimate. Long-tail claims can continue to develop for years.
Incurred and paid triangles answer different questions
An incurred-loss triangle reflects paid losses plus remaining reserve estimates. A paid-loss triangle shows only cumulative cash claim payments.
Comparing the two helps investors see both estimate development and settlement timing.
A slow-paid casualty line can show a large gap between incurred and paid losses for many years. That does not by itself mean the reserve is wrong.
Why triangles are useful
A single reserve balance hides history. A triangle preserves how estimates changed through time.
That can reveal patterns such as:
- recent accident years strengthening repeatedly;
- older years producing persistent favorable releases;
- faster or slower payment patterns;
- different development behavior across business lines; and
- unusually immature recent cohorts.
The triangle therefore gives more information about reserve uncertainty than one aggregate Loss Reserves balance.
IBNR and the triangle
Required insurer development disclosures can also show Incurred But Not Reported Reserves for each accident year. Newer years often have more IBNR because a larger share of ultimate claims is still expected to emerge or develop.
Do not compare raw triangles mechanically
Cross-company comparisons require care. Business mix, policy limits, reinsurance, acquisitions, claim definitions, currency, and the length of the disclosed history can differ.
Even within one company, changes in portfolio mix can make one accident year structurally different from another.
Investor interpretation
The most useful question is whether the pattern of revisions is consistent with the insurer's explanation of pricing, claims inflation, and reserving discipline.
Triangles are evidence, not a standalone reserve-adequacy score. Pair them with Reserve Development, case/IBNR mix, and Accident-Year Loss Ratio.
Sources
- W. R. Berkley 2025 Form 10-K loss-development disclosure
- Kinsale Capital 2025 Form 10-K loss-development disclosure
Continue Research
Continue from the concept into the Grizzly Bulls research surface that best matches the next question. These links are research continuations, not recommendations or required steps.
Continue company research
Use broader company research alongside loss-development history rather than treating one triangle as a standalone quality score.
Explore more topics in the Financial Research Encyclopedia.