Using Behavioral Economics to Inform the Integration of Human Services and Health Programs under the Affordable Care Act . Research findings


This report does not analyze how Marketplaces and Medicaid programs can best structure the application and health plan choice process. Rather, we focus on that process’s cognitive demands to understand the consumer’s mindset when faced with the additional questions posed by possible SNAP enrollment. Economists have started to consider the complex cognitive challenges associated with choices under uncertainty and over multiple time periods, providing a behavioral economics perspective to explain health insurance choices in a way that standard market models cannot.62 Recent research suggests that the number of plans and how they are presented to consumers could have a substantial impact on plan choice and costs.

Many SNAP-eligible consumers will qualify for Medicaid, rather than Marketplace subsidies. This is particularly true in states that expand Medicaid, thereby raising the lowest income threshold for Marketplace subsidies from 100 to 138 percent FPL. And while plan choice is generally much simpler in Medicaid than it is in Marketplaces, and enrollment brokers often help with those decisions, Medicaid plan selection can nevertheless present challenges. For example, in California counties where Medicaid beneficiaries are not offered a choice of health plans, more than 95 percent of new Medicaid beneficiaries enroll in a plan during their first month of eligibility. By contrast, in counties where a choice of plans is offered, fewer than one percent enroll during their first month; roughly half (51 to 56 percent) enroll in a plan within three months of qualifying for Medicaid; and one in six (15 to 17 percent) are still not enrolled seven months after being found eligible.63

Outside the Medicaid program, experimental evidence indicates that consumers left to their own devices have difficulty making choices among even four standardized plans, making significant financial errors. When instructed to choose the most cost-effective plan, unaided subjects selected the best option 42 percent of the time when presented with four choices (with an average mistake exceeding $200). When the number of choices rose to eight, the proportion of correct selections fell to 21 percent.64

Similarly, in 2009, after the Massachusetts Connector—that state’s pre-ACA health insurance exchange—reorganized a list of 25 plans into three tiers of coverage categorized by price, consumers showed improved ability to choose, becoming increasingly effective in identifying and purchasing the most affordable plan options.65 These results, showing the gains achieved when choices are greatly simplified, suggest that consumers may have difficulty making optimal choices in the much more complex, multi-plan environments that they are now facing in ACA Marketplaces and may find it challenging thinking through and deciding from among available options.

Experimental findings from Medicare Part D coverage of prescription drugs also support the notion that individuals have difficulty making optimal plan choices, even when those choices involve only a single benefit:

  • In a nationally representative study of low-income Medicare beneficiaries from the Health and Retirement Survey (HRS), researchers found that individuals who were older and had lower levels of cognition and numeracy skills were less likely to take up fully subsidized Medicare Part D benefits when they were given a range of plan options.66
  • A randomized, controlled experiment asked medical residents and interns to select the Medicare Part D prescription drug plan that would best meet the needs of a beneficiary with a specified drug regimen, taking into account premiums and out-of-pocket costs. When this highly numerate, medically aware cohort was presented with three plans, 32.8 percent made the wrong choice, with an average error of $60. Presented with nine plans, 67.3 percent made wrong choices, and the average error increased to $128.51; both increases were statistically significant.67

Another study examined the impact of cognitive limitations on the level of enrollment in supplemental coverage among Medicare beneficiaries, based on data from the HRS. Despite gaps in standard Medicare coverage that leave most beneficiaries without supplemental coverage susceptible to high out-of-pocket expenses, many seniors lack this coverage. The study found, after controlling for other factors, that beneficiaries in the lower third of cognitive ability and numeracy were 11 percentage points less likely to enroll in supplemental coverage than those in the upper third.68

Behavioral economics research has also analyzed the effects of choice overload, a finding from psychology where individuals choose inaction when the number of available options becomes overwhelming.69 For example, one study used data from approximately 800,000 workers to estimate the impact of 401(k) plan choices on participation rates. All other things equal, researchers found that each additional 10 mutual fund options were associated with an approximate 1.5 to 2 percent drop in participation rates.70,71

The evidence is somewhat mixed as to whether this particular effect applies to health coverage. Despite a preference among seniors for fewer Part D choices, some analyses have not found a link between the number of Part D plan choices and the probability of enrolling into coverage.72,73 On the other hand, the above-described research includes examples of the complexities of plan choices leading to both mistaken and deferred decision-making.

View full report


"rpt_BehavioralEconomics.pdf" (pdf, 805.46Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®