Premium
Healthcare Coinsurance Elasticity Coefficient Estimation Using Monthly Cross‐sectional, Time‐series Claims Data
Author(s) -
Scoggins John F.,
Weinberg Daniel A.
Publication year - 2017
Publication title -
health economics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.55
H-Index - 109
eISSN - 1099-1050
pISSN - 1057-9230
DOI - 10.1002/hec.3341
Subject(s) - heteroscedasticity , endogeneity , econometrics , elasticity (physics) , health care , standard error , statistics , estimation , economics , inference , panel data , point estimation , actuarial science , mathematics , computer science , materials science , management , artificial intelligence , composite material , economic growth
Published estimates of the healthcare coinsurance elasticity coefficient have typically relied on annual observations of individual healthcare expenditures even though health plan membership and expenditures are traditionally reported in monthly units and several studies have stressed the need for demand models to recognize the episodic nature of healthcare. Summing individual healthcare expenditures into annual observations complicates two common challenges of statistical inference, heteroscedasticity, and regressor endogeneity. This paper estimates the elasticity coefficient using a monthly panel data model that addresses the heteroscedasticity and endogeneity problems with relative ease. Healthcare claims data from employees of King County, Washington, during 2005 to 2011 were used to estimate the mean point elasticity coefficient: −0.314 (0.015 standard error) to −0.145 (0.015 standard error) depending on model specification. These estimates bracket the −0.2 point estimate (range: −0.22 to −0.17) derived from the famous Rand Health Insurance Experiment. Copyright © 2016 John Wiley & Sons, Ltd.