Zero-Price Effects in Health Insurance: Evidence from Colorado

Abstract

The Affordable Care Act’s premium tax credit subsidies provide millions of eligible enrollees with the option to purchase zero-premium health insurance plans, but millions more do not have this option. What difference does a premium of zero make, relative to a slightly positive one? We use regression discontinuity designs to examine zero-price effects in health insurance coverage take-up, plan choice, and coverage duration using administrative data from Colorado’s Health Insurance Marketplace from 2016 through 2019. Unlike previous studies, we isolate zero-price effects from high price sensitivity near zero using rich variation from Marketplace premium tax credits. We find no discontinuity in insurance choices when premiums increase from to a slightly positive amount, suggesting that zero is not a special price for customers. However, zero-premiums plans increase on-time enrollment, leading to longer coverage duration by reducing the transaction cost of making an initial premium payment. As low-income customers are especially sensitive to this transaction cost, making zero-premium plans available may increase targeting efficiency.

Keywords: Health insurance, Affordable Care Act, zero-price effect, regression discontinuity

JEL Classification: I11, I13, D90, I18