Further evidence on property-casualty insurance premiums: Do multiple breaks and country characteristics matter?

This paper employs a panel data stationarity test that incorporates multiple structural breaks to investigate whether property-casualty insurance (PCI) premiums per capita among 40 countries are mean-reverting or not. The results suggest that our panel dataset is stationary after we introduce the structural breaks into the model and consider cross-sectional dependence. However, the results for the panel stationarity test vary with regard to different country characteristics. Countries in which the PCI premiums present a panel unit-root property infer that any external shocks can have a permanent effect on the insurance premiums. Accordingly, the associated insurance authorities should take possible structural breaks and cross-sectional dependence into account when implementing related policies.

► The stationarity of property-casualty insurance premiums is investigated sampled as 40 countries. ► Under consider general cross-sectional dependence and finite-sample bias in panel stationarity test. ► Sub-panels based on country characteristics is examined the influence of structural breaks. ► We find each country's date of structural breaks and the event of breaks.