Pursue rate adequacy
When it may fit: Claims cost is rising faster than premium.
Evidence needed: Rate need analysis by product.
Trade-off: Rate increases can reduce retention.
Insurance Intelligence Brief
How to Investigate Changes in Loss Ratio
The loss ratio is the first number in underwriting results. When it moves, the work is to find which products, which causes and which part of the movement are real.
1. The business problem
Our loss ratio went up. Where did it come from, and will it stay?
The loss ratio, claims incurred divided by earned premium, summarizes how much of each premium dollar goes to claims. A rise of a few points can erase most of an underwriting profit, so any move deserves attention.
A loss ratio can rise because claims rose or because premiums did not keep pace. It can also reflect timing, reserve changes or a few large losses. Separating these is the first step.
2. What the evidence can reveal
A loss ratio investigation works through several cuts:
- Calculated Loss ratio by product, by quarter and by cause of change: frequency, severity and rate.
- Calculated Claims incurred against earned premium growth.
- Calculated The dollar value of each point of loss ratio.
- Information gap Reserve development, large losses, catastrophe events and changes in business mix, which are not visible in a ratio.
A step change that stays is usually a structural shift. A spike that reverses is more likely an event.
3. Investigate the problem
In the synthetic insurer, Calculated the loss ratio rose from 63.2% to 68.4% and the combined ratio from 92.8% to 97.5%. Earned premium grew 11.5% to $82.5M, but claims incurred grew 20.5%, so the underwriting result fell from $5.30M to $2.10M.
Calculated The quarterly view shows a step, not a drift: the loss ratio was 63.2% in the last quarter of FY25 and 68.4% in the first quarter of FY26, and stayed near 68.3% after that. By product, Personal Auto moved from 65.5% to 74.3%, Homeowners from 62.9% to 65.4%, Small Commercial from 60.7% to 63.4% and Specialty & Umbrella from 58.0% to 61.5%.
Calculated Each point of loss ratio is worth about $825,000 on this premium, and five points is more than the entire underwriting result of $2.10M. Hypothesis A step that persists, concentrated in one product, suggests a structural change such as a rate shortfall or a shift in claim behavior, not a single event. Information gap Reserve changes, large losses and catastrophe events are not in the demo, and any of them could account for part of the movement.
4. Explore the interactive experience
Investigate it yourself.
Insurance Portfolio Explorer: a portfolio dashboard with loss, expense and combined ratios, a scenario simulator for rate, volume, frequency and severity, evidence-linked answers from APIS, and a sample brief, all on synthetic data.
The experience below is built on a synthetic insurer, “Demo Mutual Insurance,” analyzed by product line. It uses no policyholder or claims data of any kind, and it has no distribution-channel data, so product lines are used to show the method. Every figure is illustrative.
- Click the Loss ratio card and read the quarterly trend. Is it a drift or a step?
- In the product table, read the Loss ratio Δ column. Which line moved most?
- Ask APIS why premium growth does not necessarily mean improved underwriting results.
5. Evaluate possible solutions
The response depends on whether the cause is rate, frequency, severity or mix.
6. What leadership should do next
Questions to answer
- Which products drove the move?
- Is it frequency, severity or rate?
- Is it a step or a drift?
- Are reserves involved?
Data to gather
- Loss ratio by product and quarter
- Frequency and severity detail
- Rate changes
- Reserve development
Steps to take
- Decompose the change.
- Rank products by contribution to the move.
- Test rate and mix actions.
- Track quarterly.
7. Engage Two Pillars
Investigate Performance Across Your Insurance Business
Two Pillars can help carriers, agencies and brokerages investigate portfolio, claims operations, agency economics and process improvement using your own data. This demo uses no policyholder data. A good first step is a loss-ratio decomposition that separates rate, frequency, severity and mix effects.
Sources and data notes
- The organization, its figures, and every number in this article and its interactive experience are synthetic and illustrative. They are not real data, a benchmark, or a forecast.
- Reviewed October 2026. This article describes an approach to investigation. It does not replace review by qualified professionals.