Simulating the Effects of Employer Contributions on Adverse Selection and Health Plan Choice

Published in: Health Services Research, v. 34, no. 4, Oct. 1999, p. 813-837

Posted on on January 01, 1999

by M. Susan Marquis, Joan L. Buchanan

OBJECTIVE: To investigate the effect of employer contribution policy and adverse selection on employees' health plan choices. STUDY DESIGN: Microsimulation methods to predict employees' choices between two health plan options and to track changes in those choices over time. The simulation predicts choice given premiums, healthcare spending by enrollees in each plan, and premiums for the next period. DATA SOURCES: The simulation model is based on behavioral relationships originally estimated from the RAND Health Insurance Experiment (HIE).The model has been updated and recalibrated. The data processed in the simulation are from the 1993 Current Population Employee Benefits Supplement sample. PRINCIPAL FINDINGS: A higher fraction of employees choose a high-cost, high-benefit plan if employers contribute a proportional share of the premium or adjust their contribution for risk selection than if employees pay the full cost difference out-of-pocket. When employees pay the full cost difference, the extent of adverse selection can be substantial, which leads to a collapse in the market for the high-cost plan. CONCLUSIONS: Adverse selection can undermine the managed competition strategy, indicating the importance of good risk adjusters. A fixed employer contribution policy can encourage selection of more efficient plans. Ironically, however, it can also further adverse selection in the absence of risk adjusters.

This report is part of the RAND Corporation External publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.