Using Behavioral Economics to Inform the Integration of Human Services and Health Programs under the Affordable Care Act . Insights on take-up from neoclassical economics


Expected utility maximization theory can help explain why many eligible individuals do not take up seemingly free public benefits for which they qualify. Two major literature reviews analyze the impact of multiple factors on participation levels in many different programs. These factors include transaction costs (e.g., both time and dollars spent to meet administrative requirements), lack of information on program eligibility or benefits, and “stigma” associated with government programs.15 Both reviews conclude that administrative barriers and consumer confusion or lack of information matter the most and that larger program benefits have a positive effect on participation. By contrast, stigma—defined as the psychological feeling of shame, a social sense of disrespect associated with program participation, or the perception that participation in a welfare-related program is per se undesirable16—is challenging to empirically define and does not appear to have significant effect, at least as shown in these literature reviews. In sum, tangible costs (which can be temporal or explicitly monetary) and confusion or ignorance reduce take-up, and higher benefits increase participation.

Turning to more specific research about SNAP participation, several factors can inhibit take-up. One such factor is confusion: half of eligible nonparticipants (most of whom have somewhat higher incomes than typical participants) mistakenly think they do not qualify.17 Participation is also more likely among those who qualify for higher benefit amounts, with the median benefit level of eligible nonparticipants well below that of participating households18 Researchers also find that individuals who report more serious nutritional hardship and thus who have larger perceived benefits are more likely to participate, as food insecurity is more frequently reported by SNAP recipients than by eligible non-recipients.19 Transaction costs are another factor that can impede participation, particularly for people who qualify for small benefit amounts; in such cases, SNAP’s costs could reasonably be seen as exceeding its benefits. Although slightly dated because of program streamlining, a 1999 study found that, when the minimum SNAP benefit was $10 a month, average out-of-pocket costs (mostly due to transportation) were $10.31 per application, $5.84 for each of several recertifications per year, plus an average of nearly five hours of consumer time per initial application and 2.5 hours per recertification.20 Today, the minimum monthly benefit is $15.21

Those same three factors highlighted as significant by neoclassical economics—confusion about eligibility, differential benefits, and transaction costs—also affect Medicaid and CHIP. One multivariate study found that among a randomly selected sample of community health center patients, those who perceived the Medicaid application as long and complicated or were confused about who can apply were 1.8 times more likely to be eligible for but not enrolled in Medicaid compared to those who did not perceive these barriers.22 Learning over time might also occur as Medicaid participation levels are affected more by past than current eligibility rules.23 In terms of the second factor, degree of gain, those with larger expected benefits are also more likely to participate in Medicaid. For example, elderly people with chronic functional limitations are four times more likely to enroll, compared to those without such limitations.24 Finally, much work can be required to enroll, imposing costs in the form of time. In the early 1990s, requirements for producing documentation (e.g., birth certificate, citizenship papers, proof of residency, and proof of income) and attending multiple interviews were sufficiently burdensome that up to a quarter of Medicaid applicants could not meet those requirements before applicable deadlines and so could not obtain an eligibility determination.25 Following many states’ considerable streamlining of application procedures since then, regression analyses have shown that measures of inconvenience, such as perceived application length, hinder take-up, while policies such as presumptive eligibility, which lower applicant inconvenience, have a significant positive impact on take-up.26

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®