Collective Risk Assessment in Affordable Care Act markets: A Hierarchical Model under the Bayesian framework

Since the passage of the Affordable Care Act in 2010, the US health care system has shown significant changes in utilization and costs. Under this law, health insurers can no longer reject customers based on their pre-existing health conditions. Moreover, individual premiums cannot account for the associated risk. To mitigate the unbalanced financial impact of high risk enrollees on carriers, the federal government has created a risk equalization program, risk corridors and reinsurance. The focus of this paper is on how to anticipate market instability that can cause inadequacies on risk equalization, also known as Risk Adjustment. Under this program, actuarial pricing requires the assessment of statewide risk pools, whose risk composition dynamically changes over time. In the year 2014, the expansion of ACA to individual markets allowed previously uninsured population to get insurance coverage, increasing uncertainty on statewide resource allocation. Currently, the COVID-19 pandemic is generating significant changes in utilization, prices and covered population, motivating the use of a comprehensive probabilistic framework to project insurance rates. In order to achieve this analysis, we propose a Bayesian Hierarchical Collective Risk Model to identify and quantify the uncertainty associated to statewide claim frequency and severity of medical services. The model is applied on the Commercial Health Care Cost dataset published by the Health Care Cost Institute in collaboration with the Society of Actuaries (2019).

Presenter biography:
Juan Ignacio de Oyarbide

My name is Juan Ignacio de Oyarbide. I am 26 years old and I am from Argentina. I've been studying Actuarial Science for the past 8 years. I've worked in Actuarial consulting since 2017. Some of the fields were reserving, pricing, Solvency II, pension funds and actuarial software development. I am an active user of Stan and I participate in the forum and related activities. I am looking forward to be part of the StanCon 2020 and contribute with my work. Health care is a critical field for the upcoming months and I believe Stan should be used by health actuaries during this period.