Purpose: The purpose of this paper is to analyse the determinants of workers’ compensation insurance prices in Portugal. Design/methodology/approach: Multiple regression analysis was used to study the insurance price determinants. The independent variables considered are: payroll, number of employees, and incidence rate. In addition, three categorical variables were used: region, classification of economic activities, and enterprise size. A cross-firm panel data sample from the SABI database of 1,435 firms was considered, over a time horizon from 2010 to 2012. Furthermore, the sample split criterion was the enterprise size. Findings: As expected, the results suggest that payroll and number of employees are related with workers’ compensation insurance prices. Furthermore, incidence rate, region, type of economic activities, and enterprise size have a positive and significant influence on premiums. Research limitations/implications: More panel data are needed to allow a greater focus on the impacts of GDP fluctuations and sectoral consolidation on insurance pricing. Further research could also include the impact of capital/reserving cycles, which can be driven both by economic shocks and the competition cycle. It is well known that insurers tend to reduce reserving standards when under pressure, and this can result in inadequate pricing. Practical implications: The process of workers’ compensation insurance price formation is disentangled. The results suggest that the workers’ compensation insurance premiums behave as expected, especially under periods of economic strain. Therefore, workers’ compensation rate regulation should take this evidence into account, specifically through the establishment of minimum rates, which will protect the insurer, the employer, and the employees alike. Originality/value: The paper is part of a considerable literature on insurance pricing in workers’ compensation, most of which has centred on private markets in the USA. This paper is the first empirical work that employs private market data for Portugal.
