Sectors / Insurance

Insurance

Pricing, fraud verification and capital all rest on the same question: what does the data say is likely to happen? StatDec applies its retail risk modelling experience to claims probability and severity, fraud ranking and the estimation work behind Solvency II.

Risk-based pricing Fraud modelling Solvency II

Insurers face the same modelling problems as lenders in a different form — estimating the frequency and size of a future loss, allocating scarce verification resource, and demonstrating that the models used for capital are the models used in the business. We work across all three.

Where we help insurers

Pricing

Risk-based pricing

Accurate and efficient estimation of claims probability, accompanied by severity modelling. The aim is a holistic understanding and measurement of risk, so consultation can extend to the design of the data capture process, data cleaning and mining, correspondence with other internal and external data sources, and the set-up of a validation framework.

Claims

Fraud modelling

Verifying claims for fraud is resource intensive. A model can generate fraud probabilities and rank claims by importance, so that manual verification effort goes where it is most likely to pay. The method is the same one behind our propensity models.

  • Efficient resource allocation, particularly as claim volumes rise
  • Reduction of handling costs
  • Lower premiums, as insurers reduce the cost of covering for fraud
  • Data-driven decisions and monitoring of events
  • Verification resource can be evaluated against model estimates
  • Suitable for champion/challenger testing and continuous improvement
Capital

Solvency II

Solvency II expects that models approved for capital requirements are also used for business decisions, and that technical provisions rest on a best estimate of future cash flows with appropriately homogeneous pools. Robust pricing models can serve that purpose directly.

  • Best estimate of liabilities for the formulation of technical provisions
  • Forecasting liabilities from new business within the following 12 months under the SCR framework
  • Input to the methodology of the unexpected loss analysis

Transferable experience

  • The equivalent banking framework. Our professionals have supported banking institutions on methodology design, risk component estimation, validation framework design and risk pool optimisation for the equivalent Basel II framework — the work Solvency II asks insurers to do.
  • Pools and segmentation. Homogeneous pool formulation is the same discipline whether it supports technical provisions or the capital and provisioning parameters we estimate for banks.
  • Validation built in. A validation framework is designed alongside the model, not bolted on after approval, and can be extended into periodic independent model validation.
  • Champion/challenger. Fraud and pricing models are set up so that a new approach can be tested against the incumbent on live business.

Put a number on the claim before you pay it

Whether the priority is pricing accuracy, fraud verification capacity or Solvency II evidence, we can scope the modelling work against the data you already hold.

Get in touch