Insurance pricing in Brazil has evolved significantly over the past decade. What was once a process driven largely by experience and professional judgment is becoming a more structured discipline based on data, modelling and risk analysis.
Yet having data does not automatically lead to better decisions. Between data and action lies a critical step: identifying where current tariffs diverge from the expected cost of risk. This is where tariff gap analysis becomes essential, according to Akur8.
By systematically comparing the premium currently charged with the expected loss derived from actuarial modelling, insurers can determine whether pricing accurately reflects risk. The results often reveal surprising discrepancies. In portfolios that have not undergone structured technical review, meaningful misalignment between tariff and risk is common.
In Brazil, this type of analysis is particularly important. Claims volatility, limited historical datasets in certain segments, and the regulatory environment overseen by SUSEP all make disciplined pricing practices essential.
Understanding where pricing diverges from expected loss allows insurers to transform technical insight into commercial strategy.
The four pillars of strategic insurance pricing
Effective insurance pricing requires balancing four core elements.
The current portfolio price reflects the insurer’s historical positioning, past pricing decisions and regulatory constraints.
Operating expenses, including commissions, administrative costs and loadings, determine the premium required to sustain the business.
The expected loss is the most critical component. Derived from actuarial modelling, it represents the projected future cost of risk and provides the technical reference point for pricing decisions.
Finally, market dynamics complete the picture. Competitor pricing, conversion rates, elasticity, retention levels and renewal behaviour all influence whether a technically correct price is commercially viable.
Insurance pricing cannot exist in isolation. Technical analysis must remain aligned with commercial strategy, operational capability and broader growth objectives.
Why expected loss matters
Observed claims data alone cannot support reliable pricing decisions.
Historical loss experience reflects past outcomes, but it can be heavily influenced by randomness. A portfolio may experience unusually low claims in a given year simply due to statistical volatility.
Low observed claims may also reflect pricing effects. Higher prices can drive higher-risk policyholders away, artificially improving loss ratios without necessarily indicating lower risk.
Most importantly, historical claims data does not project the future.
Expected loss modelling addresses these limitations by estimating the future cost of risk based on statistical analysis. By incorporating risk characteristics and actuarial techniques, expected loss provides a more stable and forward-looking reference point for pricing decisions.
A real-world example illustrates the value of this approach. One insurer identified a segment of 500 vehicles that had not experienced severe claims during a given year and considered expanding aggressively within that group. Actuarial modelling revealed that the expected loss for that segment was significantly higher than the observed claims suggested. The apparent performance reflected statistical luck rather than lower risk. Avoiding expansion prevented a costly pricing mistake.
Understanding frequency and severity
Expected loss is built on two components: frequency and severity. Frequency measures how often claims occur, while severity measures the average cost of those claims.
Different segments exhibit different patterns. Lower-cost vehicles may generate frequent claims but relatively inexpensive repairs. Premium vehicles may have fewer claims but significantly higher repair costs.
Analysing these elements together as a single metric can mask important dynamics. Separating frequency and severity improves the accuracy of risk modelling and helps insurers understand where pricing adjustments may be necessary.
Identifying tariff gaps in four steps
A structured tariff review typically follows four practical steps.
First, insurers map the current pricing structure. This involves documenting how premiums are constructed, including base tariffs and factors such as geographic region, vehicle characteristics, driver profile and usage type.
Second, expected loss is calculated using actuarial modelling. Historical claims data is analysed to identify risk patterns, while adjustments account for inflation, claims development and emerging trends.
Third, insurers compare the expected loss with the tariff currently charged. This comparison reveals pricing gaps.
A negative gap indicates pricing below expected risk, creating profitability concerns. A positive gap suggests pricing above expected risk, potentially reducing competitiveness. When the gap approaches zero, pricing is broadly aligned with risk.
Finally, identified gaps are validated and prioritised. Small gaps within large segments may have greater financial impact than large gaps within small segments. Statistical reliability and commercial feasibility must also be considered before adjustments are implemented.
Turning insights into pricing strategy
Once pricing gaps have been identified, insurers can begin implementing targeted adjustments.
Where segments are underpriced, more conservative pricing or underwriting strategies may be required. This can include tighter acceptance criteria, revised discount structures or controlled growth until pricing alignment improves.
Where segments are overpriced, insurers may see opportunities for profitable growth. More competitive pricing, targeted campaigns and stronger distribution support can help capture business from high-performing segments.
Importantly, pricing changes should be introduced gradually to minimise disruption and monitor their impact on renewal behaviour and conversion rates.
Pricing as a continuous process
Tariff review should not be a one-off exercise.
Continuous monitoring allows insurers to track deviations between expected and realised claims performance. Automated alerts and recurring pricing reviews help identify when recalibration is necessary.
Persistent divergence between expected and observed outcomes signals that models or pricing structures may require adjustment.
Read the full blog from Akur8 here.
