Streamlining applications to social programs unlocks more support for the U.S. families who need it most

Key takeaways:

U.S. social programs often have overlapping eligibility rules. Many families enrolled in one program do not enroll in other programs for which they almost certainly qualify. Research on participation in these programs has tended to examine them one at a time, which offers valuable insights but can obscure the full picture.
In 2012, Virginia moved its application systems for health, nutrition, and cash assistance programs into a single digital platform, which researchers found meaningfully increased participation in social programs. The enrollment portal not only brought new families into program participation, but also helped families already receiving benefits from one program to add others for which they were eligible but had been missing.
What this means for growth: Because the gains from simplifying access to social programs spill over across these programs, evaluating a reform through any one program on its own can understate its full return. Researchers find, for instance, that a nutrition-assistance-only evaluation understates its complete value by as much as 64 percent, largely because it omits the substantial value families place on the health care coverage they gain when they can apply for multiple programs simultaneously. Given how interconnected today’s social programs are, evaluating them one at a time can lead to potentially misleading calculations about whether a policy investment pays off.

Overview

Millions of low-income families in the United States tap into multiple social programs to help them meet their needs. A single household, for example, might rely on the Supplemental Nutrition Assistance Program for groceries, Medicaid for medical care, and Temporary Assistance for Needy Families for cash assistance. These programs share overlapping eligibility rules, and roughly half of all recipients enroll in two or more of them.

Understanding the network of social programs and how they help U.S. families therefore means looking beyond whether someone receives benefits from any given program to examine how intensively families are connected across programs and whether they are getting all the support for which they are eligible. Research on these programs generally examines them one at a time, which offers valuable insights but can overlook this broader picture.

In our recent study, however, we adopt a multiple-program framework to show that a substantial share of U.S. families who are already enrolled in one social program do not take up others for which they appear to be eligible. We also find that reducing the hassle of applying for different programs can pull families into fuller participation, and that the people pulled in this way are disproportionately the most disadvantaged. Because the gains from simplifying access spill over across programs, evaluating a reform of this kind through the lens of any one program on its own misstates its full return.

This column first details the data showing that many families are not enrolled in all the social programs for which they qualify, leaving needed support unutilized. It then dives into a program in Virginia called CommonHelp that allows families to apply for multiple social programs at the same time, exploring who benefitted from this reform and why looking at participation in multiple programs is an essential new framework for studying social programs. It closes with some lessons for policymakers seeking to broaden access to social programs and increase support for U.S. families in need.

Many U.S. families are not fully enrolled in social programs

To study these questions, we use linked administrative records from Virginia covering the universe of SNAP, Medicaid, and TANF recipients from 2007 to 2022. This data source is a meaningful improvement over the household surveys on which prior research often relies. Surveys substantially undercount the benefits of accessing these social programs, especially for people enrolled in multiple. In Virginia, for example, we find that conventional surveys miss between 30 percent and 60 percent of households jointly receiving nutrition assistance and Medicaid.

We also find that many families already enrolled in one social program do not take up other programs for which they almost certainly qualify—and this take-up gap is far from trivial. After Virginia raised SNAP income limits and eliminated its asset test in 2021, the data show that more than half of Virginia adults receiving Medicaid still were not enrolled in the Supplemental Nutrition Assistance Program, despite appearing eligible based on income and assets. At the same time, we estimate that nearly half of Medicaid-eligible (but not enrolled) children in Virginia already receive benefits from another social program. This implies that incomplete take-up among existing program recipients may account for a large share of overall nonparticipation.

We also find that families living farther away from their county field office participate less in multiple programs—even after accounting for local poverty rates and demographics. This pattern is consistent with the idea that the administrative burdens of physically traveling to an office, learning about program eligibility, and navigating complex paperwork with caseworkers prevents people from claiming all the benefits for which they qualify.

Measuring the effect of reducing administrative burdens on access to social programs

Our study looks at the impact of CommonHelp, an online portal that Virginia launched statewide in 2012 to let residents apply for nutrition assistance, Medicaid, and Temporary Assistance for Needy Families through a single digital platform rather than delivering paper forms to a local field office. Virginia was an early adopter, but it is far from alone. As of 2024, 31 states had built integrated online application platforms for these three social programs, making the question of whether such tools actually improve access a timely one for policymakers across the country.

Because CommonHelp launched everywhere in Virginia at the same time, there was no obvious group of nonparticipants to compare against. We address this with a framework that combines two complementary research designs. The first compares enrollment changes across ZIP codes that differ in their distance to the assigned field office, isolating the effect of easing travel-related, or “geographic,” burdens. The second focuses on areas physically close to their assigned field offices and compares the jump in enrollment right after CommonHelp launched to seasonal patterns in other years, isolating “fixed” burdens that affect everyone regardless of distance, such as the complexity of the forms and the stigma of an in-person visit. Together, these research designs uncover the reform’s total effect on enrollment in social programs.

We find that CommonHelp meaningfully increased program participation in Virginia, and that the gains were concentrated in multiple-program enrollment. Over a 3-year period, the reform brought in about 57,000 additional multiple-program recipients, a 4.1 percent increase, while single-program enrollment fell by roughly 22,000 recipients. In other words, CommonHelp not only brought new families into social program participation, but also helped families already receiving benefits from one program to add others for which they were eligible but had been missing.

Who the CommonHelp reform reached

We also develop a method to separate two kinds of responses to CommonHelp: people drawn in from no programs at all and people who moved from receiving benefits from a single program to multiple programs. We estimate that 38 percent to 62 percent of the enrollment gains came from this second group, who we call “intensive-margin” responders. This distinction between participants entering one social program and those deepening their participation is not just a matter of accounting. It also tells us something about who benefits from making programs easier to access.

Linking recipients to earnings and criminal justice records, we find that intensive-margin responders were notably more disadvantaged, with weaker labor market attachment and more contact with the justice system. These recipients were more disadvantaged than both the families newly participating in social programs, as well as those who would have enrolled in multiple programs even without the reform.

The implication is counterintuitive but consequential. Administrative burdens were screening out some of the most vulnerable people—not by keeping them off the rolls entirely, but rather by preventing them from claiming the full set of benefits for which they were already eligible.

This points to a lesson that runs against conventional wisdom about program targeting. Policymakers often think about reaching those in need by enrolling people who are not yet in any program. But our results suggest that some of the most disadvantaged families are already in the system—just not fully enrolled.

Why accounting for all programs changes how a policy may be assessed

Accounting for multiple program participation also reshapes the benefit-cost analysis of a policy reform. This matters as policymakers weigh whether to invest in or cut public benefits and the administrative systems that deliver them because the return on such investments depends on benefits that reach beyond any single program.

We evaluate CommonHelp using the Marginal Value of Public Funds framework, which compares how much recipients value a policy against its net cost to government. The levels of these estimates are sensitive to modeling assumptions, so rather than emphasizing any single number, we focus on how much of a policy’s complete value is missed when researchers can only see one program at a time.

We find that a SNAP-only evaluation of CommonHelp would understate its true value by as much as 64 percent, largely because it omits the substantial value families place on the Medicaid coverage they gain when applying simultaneously. Even a weighted average of single-program evaluations—the closest approximation available without linked data across programs—misses between 10 percent and 24 percent of the complete value. Given how interconnected programs are in the vast network of social programs that exist, evaluating them one at a time can lead to potentially misleading calculations about whether a policy pays off.

Lessons for policymakers

The central lesson of our study is that the various social programs available to U.S. families today cannot be understood in isolation. Because these programs overlap so heavily, policies that touch one program ripple across others, and families experience them not as separate benefits but as a bundle they move into and out of together. Conducting analyses as though families interact with a single program at a time can understate both the scope of unclaimed benefits and the returns to making these programs and their applications easier to navigate.

This multiple-program framework carries two key implications. First, it changes how policymakers and researchers alike think about targeting. Some of the most disadvantaged families are already connected to social programs but only partially. Improving and expanding access for existing single-program participants, who program administrators can readily identify and reach, may be an especially practical way to target support toward those who need and value it most.

Second, it changes how policymakers measure the return on a policy. Because the value of simplifying access spills over across programs, an investment that looks to have modest returns through the lens of one program could prove more valuable once its effects across all social programs are taken into account. Conversely, cuts to a single program or to the systems that connect applicants to benefits may forfeit returns that a program-by-program accounting never registers.

As families increasingly rely on several social programs at once, both the analysis and the design of these programs’ policies should reflect how they coexist together rather than viewing each on its own.

The authors of this column used Claude to help refine some language.

Did you find this content informative and engaging?
Get updates and stay in tune with U.S. economic inequality and growth!