A Microeconometric Model of the Demand for Health Care and Health Insurance in Australia

Abstract

Adverse selection is perceived to be a major source of marketfailure in insurance markets. There is little empirical evidence on the extent of the problem. We estimate a structural model of health insurance and health care choices using data on single individuals from the NMES. A robust prediction of adverse-selection models is that riskier types buy more coverage and, on average, end up using more care. We testfor unobsewables linking health insurance status and health care consumption. Wefind no evidence of informational asymmetries.

Inequality in health care utilization in Germany? Theoretical and empirical evidence for specialist consultation

“…Estimating the number of doctor visits is a classical field for the application of count data regression models (Cameron et al 1988;Cameron and Trivedi 1998). Count data are able to exploit the specific information of a positive discrete variable with right skewed distribution, which is the case for our dependent variable number of specialist visits within the last 12 months in the sample distribution for both men and women.…”

“…However, according to Cameron et al (1988) income appears to be much more important in determining health insurance choice than determining health care service use. Similarly, van Doorslaer et al (2002, 2004b highlight the positive correlation of income and private health insurance.…”

Hips and hearts: The variation in incentive effects of insurance across hospital procedures

“…This estimate aggregates over the insurance effects on the elective and non-elective surgeries, as defined above, and hospitalizations for other purposes. This aggregate measure of health care utilization is often used in the 5 As an added check on the robustness of our results, we also estimated 2SLS regressions instrumenting insurance status with a subset of instrumental variables used by Cameron et al (1988) and Srivastava and Zhao (2008) (marital status, country of birth, a dummy for metro areas, education, past hospitalizations for purposes other than elective and non-elective surgeries, a dummy for having a child under 18 years old, as well as its interaction with marital status (to indicate single parent households), smoking status, a dummy for being a heavy drinker, a dummy for being overweight, and a dummy for not exercising). Using the full set of instrumental variables, we reject the null hypothesis that the instruments are uncorrelated with the error term based on overidentification tests for both elective and non-elective surgeries.…”

“…Several studies have used instrumental variables to separate causal effects from selection. Examples include Cameron et al (1988), Srivastava and Zhao (2008), Cheng andVahid (2011) andDoiron (2012). Most of these studies rely to some extent on exclusion restrictions involving socioeconomic or demographic variables and in some cases risk behaviours (smoking, drinking, exercise and BMI).…”

Diferenciais de utilização do cuidado de saúde no sistema suplementar brasileiro

“…A evidência teórica e empírica já apontou, em diversos estudos (CUTLEr e ZECKHAUSEr, 2000;NEWHOUSE, 1993;MANNING et al 1987;CAMErON, 1988;SAPELLI e VIAL, 2003;CHIAPOrrI, 1998) que o risco moral é um dos principais problemas no mercado de seguros de saúde. O risco moral gera perdas de bem-estar na medida em que parte da população não está apta a adquirir um plano ou seguro de saúde devido ao aumento dos custos com consequente elevação dos preços.…”