Information to Increase Insurance Take-up and Reduce Market Risk in the United States

Researchers:
Richard Domurat
Isaac Menashe
Location:
California, United States
Sample:
87,394 households
Timeline:
2015 - 2019
Intervention type:
Health care delivery Insurance Nudges and reminders Subsidies
AEA RCT registration number:
AEARCTR-0002115

Partners:

While the Patient Protection and Affordable Care Act in the United States subsidizes health insurance coverage among individuals with low incomes, insurance take-up rates remain surprisingly low. Researchers studied the impact of reminder letters addressing possible barriers related to information and behavior on insurance take-up, as well as their impact on health insurance market risk. The reminder letters reducing informational and behavioral barriers to enrollment increased insurance take-up and lowered the average market risk.

Policy issue

The Patient Protection and Affordable Care Act (ACA) subsidizes health insurance premiums for households with lower incomes in states’ health insurance marketplaces, subsidizing coverage of 10 to 11 million people at any given time. Despite the availability of premium subsidies, insurance take-up rates are surprisingly low: approximately 60 percent of individuals eligible for federal premium subsidies remain uninsured. One reason eligible individuals might not enroll is that they do not value marketplace coverage. In fact, previous research suggests that eligible individuals are only willing to pay for coverage far below their actual health care costs. Another reason why eligible individuals might not enroll and why their willingness to pay is so low is that they face barriers related to information (such as about the plans) or behavior (such as forgetting to enroll), but it is unclear if addressing these barriers would increase enrollment. Additionally, efforts to reduce barriers could improve or worsen health insurance market risk, depending on whether new enrollees are healthier—potentially lowering premiums and improving market stability—or sicker, possibly increasing premiums and lowering well-being. Could addressing informational or behavioral barriers to enrollment in health insurance marketplaces increase enrollment and improve market risk?

Context of the evaluation

Covered California is the state of California’s health insurance marketplace that connects individuals at or below 600 percent of the Federal Poverty Level (FPL) with health insurance coverage at federally subsidized rates. Under the ACA during the years of the study, most US residents were required to purchase health insurance, and nonexempt adults who chose to remain uninsured had to pay a penalty when they filed their taxes. Health insurance plans were sold in four primary levels of coverage with increasing coverage of medical expenses and significant subsidies. For example, the Bronze Plan in California in 2019 for individuals below 250 percent of the FPL had a subsidized premium of $1 to $10 per member per month, reflecting an average subsidy of $371 per month. Yet surveys conducted by Covered California showed that despite the availability of premium subsidies, many uninsured Californians did not understand or were unaware of their eligibility for premium subsidies, suggesting the presence of informational and behavioral barriers to enrollment.

Researchers selected individuals for the study sample who were eligible for Covered California in 2016 and had not yet selected a plan. These individuals either directly applied to Covered California and were deemed eligible yet never selected a plan, or became newly eligible through the state Medicaid program after experiencing a change in their eligibility. The average age of an individual in the study sample was 37 years old and the average income level was 215 percent of the FPL.

Details of the intervention

In partnership with Covered California, researchers conducted a randomized evaluation to study how reducing informational and behavioral barriers impacts individuals’ insurance take-up and market risk.

Individuals from 87,394 households were randomized into one of five groups that received the following communication from Covered California in addition to generic outreach and state-wide marketing activities used by Covered California for all consumers:

  • Control (Arm 1): Nothing additional, representing the status quo.
  • Basic Reminder Letter (Arm 2): Reported the open enrollment deadline, and the Covered California website and telephone number where they could shop for plans.
  • Subsidy and Penalty Letter (Arm 3): Reported the content of Arm 2, plus the household’s estimated monthly subsidy and tax penalty based on their reported income and household size.
  • Price Compare Letter (Arm 4): Reported the content of Arms 2 and 3, plus a table listing plans offered in their market with the cost of their monthly premium after the subsidy is factored in.
  • Price and Quality Compare Letter (Arm 5): Reported the content of Arms 2, 3, and 4, but the table also included plans' quality rating based on access to the right medical care and members’ reported experience of care and services. Quality ratings were based on Consumer Assessment of Healthcare Providers and Systems (CAHPS) measures.

Each of the four letter interventions was designed to reduce one or several behavioral barriers that potentially impede individuals’ take-up of plans: inattention and forgetfulness and/or lack of awareness about Covered California, subsidies and penalties, and plan attributes.

Researchers used Covered California’s administrative data, including data on applications and enrollment, to track plan selection, payment, and cancellations throughout 2016. Researchers also used data from the Office of Statewide Health Planning and Development (OSHPD) and IBM Watson Health to measure health spending risk from 2015 and prospective health spending risk for 2016.

Results and policy lessons

The study documents the existence of informational and behavioral barriers in health care enrollment and found that informational interventions improved take-up of plans and, on average, induced healthier consumers into the market, contributing to lower average market risk.

The letter interventions raised overall plan enrollment by 1.3 percentage points from the control group take-up rate of 8.1 percent (a 16 percent increase). The effect was larger for those who entered through a change in their Medicaid eligibility compared to those who applied directly to Covered California, in part because the former had lower awareness of the marketplace. Overall, there were no detectable differences in plan take-up across the letter interventions.

The intervention lowered average health spending risk by 5.1 percent, which implies that individuals who enrolled due to the intervention were 37 percent less costly than other consumers. Researchers also estimated that the letters raised measured willingness to pay for insurance by $25-$54 per month.

The results highlight the potential for further reducing informational and behavioral barriers to increase enrollment and improve market risk. The evidence suggests that simplifying enrollment processes—such as through reminder emails with easy links to enrollment or auto-enrollment—could increase enrollment, more than through subsidies alone. More intensive outreach and education with personalized information, particularly among highly-subsidized consumers, could also increase enrollment.

Domurat, Richard, Isaac Menashe, and Wesley Yin. “The Role of Behavioral Frictions in Health Insurance Marketplace Enrollment and Risk: Evidence From a Field Experiement.” NBER Working Paper #26153, August 2019.