Competition and efficiency in the non-life insurance market in South Africa

Abstract

Purpose The purpose of this paper is to examine the empirical effect of competition on cost and profit efficiency in the South African non-life insurance market in a three-stage analysis. Design/methodology/approach Using annual firm level data on 80 non-life insurance companies from 2007 to 2012, the authors first employ the stochastic frontier analysis (SFA) to estimate cost and profit efficiency scores. In the second stage, the authors measure insurance market competition using the Panzar-Rosse (P-R) H-statistics. In the final stage, the authors estimate a fixed-effects panel regression model which controls for heteroskedasticity to examine the effect of competition on the estimated efficiency scores. Firm size, diversification, age, risk, reinsurance and leverage are employed as control variables. Findings From the SFA, the authors find average cost and profit efficiency of 80.08 and 45.71 per cent, respectively. This suggests that non-life insurers have high levels of efficiency in cost and low efficiency in profit. The annual estimates of the P-R H-statistics also suggest that firms in the market earn revenues under conditions of monopolistic competition. The authors find a positive effect of competition on cost and profit efficiency to validate the “quiet-life” hypothesis which posits that competition improves efficiency. Practical implications Regulatory policies should be directed towards enhancing competition to improve on the low profit earning potential of firms in the non-life market. Originality/value To the best of the authors’ knowledge, this study presents the first application of a non-structural measure of competition to examine the empirical relationship between competition and efficiency in insurance markets.