The unfortunate regressivity of public natural disaster insurance: Quantifying distributional implications of EQC building cover for New Zealand

<p>This thesis examines the question “What have been the distributional implications of the setup of Earthquake Commission (EQC) building cover for New Zealand homeowners?” In New Zealand, the vast majority of property owners pay identical premiums for the benefit of the first $100,000 tranche of natural disaster cover per dwelling. The research provides a detailed quantification of the degree of regressivity of the scheme created by these flat premiums. Using EQC claims and property datasets relating to the Canterbury Earthquake Series, I test the hypothesis that wealthier homeowners are receiving more benefit. Wealth is identified by property value, income and a range of socio-economic variables collected from the most recent New Zealand Census before the earthquake series. In explaining EQC total dwelling payout by property value and by these socio-economic variables, the research shows there is a distributional implication to EQC’s building cover. This thesis includes a proposed modification to the premium structure of the scheme, whereby regressivity could be avoided. The research concludes with a survey of other public natural disaster insurance schemes worldwide, and identifies those likely to face similar regressivity issues.</p>

The unfortunate regressivity of public natural disaster insurance: Quantifying distributional implications of EQC building cover for New Zealand

<p>This thesis examines the question “What have been the distributional implications of the setup of Earthquake Commission (EQC) building cover for New Zealand homeowners?” In New Zealand, the vast majority of property owners pay identical premiums for the benefit of the first $100,000 tranche of natural disaster cover per dwelling. The research provides a detailed quantification of the degree of regressivity of the scheme created by these flat premiums. Using EQC claims and property datasets relating to the Canterbury Earthquake Series, I test the hypothesis that wealthier homeowners are receiving more benefit. Wealth is identified by property value, income and a range of socio-economic variables collected from the most recent New Zealand Census before the earthquake series. In explaining EQC total dwelling payout by property value and by these socio-economic variables, the research shows there is a distributional implication to EQC’s building cover. This thesis includes a proposed modification to the premium structure of the scheme, whereby regressivity could be avoided. The research concludes with a survey of other public natural disaster insurance schemes worldwide, and identifies those likely to face similar regressivity issues.</p>

Advertising and Risk Selection in Health Insurance Markets

This paper studies the impact of advertising as a channel for risk selection in Medicare Advantage. We provide evidence that insurer advertising is responsive to the gains from risk selection. Then we develop and estimate an equilibrium model of Medicare Advantage with advertising, allowing rich individual heterogeneity. Our estimates show that advertising is effective in attracting healthy individuals who are newly eligible for Medicare, contributing to advantageous selection into Medicare Advantage. Moreover, risk selection through advertising substantially lowers premiums by improving insurers' risk pools. The distributional implication is that unhealthy consumers may be better off through cross-subsidization from healthy individuals. (JEL D81, G22, I13, I18, M37)