Can Public Options Lower ACA Marketplace Premiums? Evidence from Washington’s Cascade Care

• In Event: ACA Insurance Policies

Abstract

The Affordable Care Act (ACA) made many meaningful improvements to the American health care system. Yet the ACA fell short in at least one key respect: cost containment. Perhaps the most prominent cost containment provision, the public option, was defeated during congressional deliberation. The purpose of this research paper is to examine whether the public option, when implemented at the state level, can effectively limit health care insurance premiums without altering private insurer participation in the Marketplace. 

American states have long been recognized as valuable laboratories for social public policy experiments (see Brandeis’ famous opinion in New State Ice Co. v. Liebmann). Three states have now passed public-option style programs: Washington (2021), Colorado (2023), and Nevada (2026). Washington’s program, Cascade Care, began operations in 2021. 

This paper addresses two central questions: 1) Did Cascade Care reduce premiums?, and 2) Did Cascade Care alter private insurer participation in the Marketplace? Our data comes from the Robert Wood Johnson Foundation’s HIX Compare dataset. HIX Compare contains plan-level data on premiums and insurer participation for all health plans offered in ACA Marketplaces from 2014 through 2026. The dataset covers all 50 states plus the District of Columbia. 

Much of the previous research that examines the effects of public options has been derived from advanced simulations (Blumberg et al 2020; Holahan and Simpson 2021) rather than actual state experiences. 

Alternatively, Sen et al (2022) provide a helpful empirical examination of the first year of Cascade Care, whereas Chen (2024) provides some initial observations of the states that have initiated public options. Waldrop (2025) provides the most recent update on public option plans in the various states. 

Yet missing in the literature is a robust examination of the effects of a public option program that utilizes causal inference techniques to establish credible counterfactuals. We do so through two separate identification strategies. The first is centered on the synthetic control method (Abadie, Diamond, and Hainmueller 2010). We construct a weighted combination of comparison states to most closely match Washington’s premiums and insurer participation rates prior to initiation of Cascade Care. We then examine whether premiums and insurer participation rates statewide deviated from this synthetic Washington after initiation. Our second identification strategy utilizes a difference-in-difference analysis of the various counties within Washington during the three phases when Cascade Care was rolled out. 

Our results will be helpful both to those who are currently implementing public option policies (Colorado and Nevada) as well as those that are actively considering doing so (Minnesota, Connecticut, etc.). We answer several questions: Did Washington achieve the 12 to 17% reductions in premium projected by the simulations? How did changes made to Cascade Care during its implementation phases, including the move from voluntary participation to mandatory participation, affect premiums and private insurer participation? In the spirit of Brandeis, should this policy experiment be considered at the national level?

Authors

• Greg Thorson, University of Redlands
  Presenting Author

• Sam Cooper Illich, University of Redlands
  Non-Presenting Co-Author