Constructing a valid counterfactual to evaluate an income-targeted policy: means-tested subsidies and tax penalties on private health insurance coverage

Many governments intervene in private health insurance (PHI) markets to incentivise demand, balance efficiency and equity, and counter adverse selection. In the context of rising health care costs, this is a complex task, and understanding the relative effectiveness of interventions can help governments design an optimal policy mix. We evaluate the impact of means-testing a premium rebate and increasing an income tax penalty rate on PHI hospital coverage in Australia. We employ difference-in-difference (DID) analysis on a Household Income and Labour Dynamics in Australia (HILDA) survey panel of 6,179 individuals. To construct a valid control group, individuals below the policy income threshold are matched to the intervention group using entropy balancing on pre-treatment covariates and trends. Our analyses suggest that FPHII increased the probability of holding hospital cover by 1.5 to 1.8 percentage points (p<0.05 to <0.1). The effect was driven by a significant increase in coverage of 2.4 percentage points (p<0.1) for those in the highest income tier. Estimates were relatively robust to sensitivity analyses, including removing those with complicated income streams and removing potential income shifters. Our findings suggest that increased tax penalty rates had a stronger impact than means-testing premium rebates, most likely due to the low price elasticity of PHI demand in Australia. We discuss these findings in the context of developing an efficient policy mix to attract and maintain PHI members in Australia and the US to promote PHI market sustainability.