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


The neoclassical model of economics assumes that each person has fairly consistent preferences over time and maximizes his or her overall well-being based on the best available information. In contrast, behavioral economics focuses on human limitations, empirically testing the ways in which human behavior departs from the rational and objective calculation of self-interest as the basis of decision-making. The latter literature contains useful findings that can inform policymakers' decisions about how to structure two key connections between health and human services programs under the Affordable Care Act (ACA).

The first connection involves Medicaid programs that use targeted enrollment strategies to provide Medicaid to beneficiaries of the Supplemental Nutrition Assistance Program (SNAP), employing waivers that avoid the need for individualized income determinations. Considerable research shows that modest procedural requirements—completing a simple form or even checking a box—can greatly lower participation levels in public and private benefit programs. The first few states to implement targeted enrollment strategies, which sent SNAP recipients mailings with forms that needed to be returned, covered numerous consumers, but many more would likely receive Medicaid if procedures were fine-tuned to eliminate the need to complete paperwork. State Medicaid programs have accomplished this in the past by: (a) asking eligible consumers to select a managed care plan within a specified period and, if they failed to choose, enrolling them into a plan automatically assigned by the state; and (b) sending eligible consumers Medicaid cards which, when used to seek care, triggered mandatory managed care enrollment, with a plan selected by default if consumers did not make a choice. The latter approach is more likely to meet the “Medicaid application” requirement that applies to targeted enrollment waivers, but the former could be modified to meet that requirement.
The second connection involves applicants for health coverage at Health Insurance Marketplaces, some of whom qualify for but do not yet receive SNAP. The behavioral economics literature suggests that the process of demonstrating financial eligibility for subsidies and sorting through premiums, deductibles, co-pays, provider networks, and other factors to choose a health plan could tax many consumers’ cognitive resources, leaving them unwilling or unable to apply for or even process information about SNAP. To address this issue, states interested in reaching additional eligible SNAP households could consider several strategies that focus on consumers whose applications for health coverage indicate a reasonable likelihood of SNAP eligibility.
After such a consumer has finished applying for health coverage and selecting a health plan on-line, the consumer could be told that he or she may qualify for help paying for food and that someone from the state’s food agency would be glad to reach out to the consumer. If the consumer expresses interest, the consumer would be asked for his or her preferred mode of contact (text message, email, phone call, etc.). The health program would forward that contact information to the SNAP agency for follow-up and enrollment into SNAP. Only after consumers have been presented with this method of SNAP connection would they be given opportunities to receive more information about SNAP and to apply for SNAP on-line—opportunities that someone in a state of cognitive depletion could perceive negatively and to which they might respond by terminating their interaction with the Marketplace.
As another option, before beginning the enrollment process, the consumer could be given the choice to submit either an application for health coverage alone or an application for both health programs and human services programs, including SNAP. Eligible consumers who complete a multi-program application would receive SNAP, avoiding the risk of “falling between the cracks” and missing out on SNAP because of failing to receive sufficient follow-up. On the other hand, this approach would require time and effort from state and federal health officials in developing, reviewing, and approving a multi-program application. Such time and effort is in short supply with health agencies primarily focused on meeting the basic requirements of ACA implementation. Also, despite federal requirements that consumers must be informed that they have the option to answer only questions related to health coverage, adding SNAP-related questions before the health application is completed means that some consumers who would otherwise have finished the health application may instead fail to pursue a longer and more complex multi-program application through to completion.
Finally, SNAP could be raised as a possibility as health coverage is renewed, when consumers will likely experience significantly less cognitive depletion than during the initial application for health coverage. Consumers must present much less information to renew than to initially qualify for assistance, and they can continue with their current plan rather than select a new insurance package from among available options. At renewal, consumers may have the time and cognitive resources needed to apply for SNAP without delay, rather than simply furnish contact information for later follow-up by the SNAP agency. However, both options could be presented so that different types of consumers could each have their needs met.
This illustrates the importance of a system that can adapt to different individual preferences in shaping the connection between health coverage and SNAP. Procedures that work well for some consumers may work poorly with others. The most effective system for connecting health applicants to SNAP would allow different types of consumers to find the particular channel of assistance best suited to meeting their needs.
Policymakers would benefit from field-testing the options described above. Many of these choices involve empirical questions about which approach will prove most effective. Randomized controlled experiments, subject to strong ethical safeguards, can go beyond informal “market testing” to rigorously assess the impact of particular methods for smoothly integrating enrollment procedures for health and human services programs.

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®