Stranded by the Safety Net: Addressing Benefit Cliffs to Assist Families and Businesses — Angela Rachidi

Summary

A statement by Angela Rachidi, Senior Fellow at the American Enterprise Institute, before the U.S. Senate Committee on Small Business and Entrepreneurship for its June 3, 2026 hearing on the burden of federal benefits cliffs on small businesses and workers. The testimony defines benefit cliffs as situations where a small rise in earnings triggers an abrupt loss of government assistance, and uses a SNAP example in which a participant loses approximately $1400 per year in benefits when crossing $45,340 per year. It explains how phaseouts stack when families receive multiple benefits, producing high effective marginal tax rates, and cites survey research on workers declining higher-paying jobs. It closes with a table of administrative and legislative reforms proposed by an AEI working group, including a superwaiver and standardized eligibility rules.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Full text

Statement before the U.S. Senate Committee on Small Business and Entrepreneurship
hearing titled “Examining the Burden of Federal Benefits Cliffs on Small Businesses
and Workers”




Stranded by the Safety Net: Addressing Benefit Cliffs
to Assist Families and Businesses

Angela Rachidi
Senior Fellow and Rowe Scholar in Opportunity and Mobility Studies
American Enterprise Institute



June 3, 2026
Angela Rachidi, PHD
Senior Fellow and Rowe Scholar in Poverty Studies
American Enterprise Institute


Chairman Ernst, Ranking Member Markey, and members of the Small Business Committee. Thank
you for the opportunity to testify. My name is Angela Rachidi, and I am a Senior Fellow in poverty
studies at the American Enterprise Institute. I have spent much of the past 20 years researching U.S.
safety net programs for low-income families. I worked for the New York City Department of Social
Services in various positions from 2002 through 2015, ending my time as the Deputy Commissioner
for policy research. Since 2015 I have continued this work for the American Enterprise Institute or
AEI.

Over the past two years, I have led a working group of policy experts and researchers focused on
policy solutions to benefit cliffs. I use the term benefit cliffs to broadly describe situations where
even a small increase in earnings or hours worked can trigger an abrupt loss of government
assistance for working families. In some cases, the loss of government assistance can completely
offset new earnings, making a family possibly worse off than before. Other times, the loss may only
partially offset new earnings, but in both scenarios, the loss of benefits can influence people’s
employment decisions. Some workers may rationally choose to work less than they otherwise would
in the absence of government assistance. This harms both families by limiting their upward mobility
and employers who rely on these employees.

I want to cover three key points in my testimony: First, benefit cliffs exist despite individual
program rules that appear to phase benefits out as participant income rises. Second, the problem of
benefit cliffs compound when families receive multiple benefits, which is the most common
situation among low-income families. And Third, research shows that the potential of hitting a
benefit cliff influences people’s behavior, resulting in an unwillingness to accept higher-paying jobs
for fear of losing benefits.

What are Benefit Cliffs?
Safety net programs in the U.S. are targeted toward those with the least income. This allows
programs to provide sufficient resources to meet a low-income family’s basic material needs, but
also ensures public funds are used efficiently. By targeting assistance to those most in need, program
rules require benefits to phase out as family income rises. This is how most of our safety net
programs work.

In most programs, the maximum amount of assistance is provided for those with no income, and
the benefit amount slowly decreases as income rises. Ideally, this benefit reduction is small enough
that it does not affect decisions around employment, such as taking on more hours or accepting a
pay raise. Furthermore, the benefits should eventually phase out to zero once a family crosses a
certain income or asset line. When the maximum benefit, the reduction rate, and income exit point
work in tandem, individuals are less likely to change their behavior based on the loss of government
assistance.

However, many program rules do not align and recipients oftentimes experience an abrupt drop in
benefits when a slight increase of earnings or assets pushes them across the line. The figure below
shows an example from the Supplemental Nutrition Assistance Program or SNAP. The blue line

                                                    2
shows the current program rules where benefits decrease at a reasonable rate. However, the gross
and net income eligibility limits (the green and purple lines) prevent the benefit from phasing out to
zero. The result is that a SNAP participant in this scenario will experience a loss of approximately
$1400 per year in SNAP benefits when they cross over $45,340 per year.1 This dynamic encourages
the participant to keep their income just below the income eligibility limit. SNAP is just one
example; many state-led child care assistance programs and some TANF programs operate this way
too. Medicaid adds to this complexity by requiring a change in coverage once income levels reach a
certain limit rather than phasing out like other programs.

Figure 1: SNAP Benefits (Household Size of Three): Hypothetical vs. Current Policy




Source: Angela Rachidi and Erik Randolph, Eliminating the Benefit Cliff and Achieving Savings for Taxpayers: A Reform Proposal
for the Supplemental Nutrition Assistance Program (Washington, DC: American Enterprise Institute, 2025).
https://www.aei.org/wp-content/uploads/2025/01/Eliminating-the-Benefit-Cliff-and-Achieving-Savings-for-Taxpayers.pdf?x97961

Importantly, some individual program phase out rates align with exit points — meaning they phase
out to zero as income rises, such as housing assistance, the EITC, and CTC. However, when
families receive multiple benefits, the phaseout rates stack on top of each other and create a
compounding effect that result in high effective marginal tax rates. Consider a hypothetical example:
A family receives $100 in benefits each from the EITC, CTC, and SNAP for a total of $300 in
benefits. Each program phases out at 20 percent for each additional dollar earned. If this household
can increase earnings by $100, they will lose $20 in benefits from each of the EITC, the CTC, and

1
 Angela Rachidi and Erik Randolph, Eliminating the Benefit Cliff and Achieving Savings for Taxpayers: A Reform Proposal for the
Supplemental Nutrition Assistance Program (Washington, DC: American Enterprise Institute, 2025). https://www.aei.org/wp-
content/uploads/2025/01/Eliminating-the-Benefit-Cliff-and-Achieving-Savings-for-Taxpayers.pdf?x97961

                                                               3
SNAP for a total of $60 lost in benefits. The result is a 60 percent effective marginal tax rate,
meaning that the family would only gain $40 from their $100 pay raise. Thus, we can see how this
could negatively affect this family’s decision to accept the higher pay. Figure 2 demonstrates this
exact dynamic at play when the benefits are not coordinated across programs.

Figure 2: An Example of the Stacking Effect of Hypothetical Safety Net Benefits




Source: Angela Rachidi, Matt Weidinger, Joshua Bandoch, Nic Dunn, Leslie Ford, and Erik Randolph, Stranded by the
Safety Net: How to Fix the Benefit Cliff Problem (Washington, DC: American Enterprise Institute, 2025).

The Georgia Center for Opportunity and the Federal Reserve Bank of Atlanta, among others, have
reviewed program rules in various states and simulated what happens to benefit amounts in actual
scenarios when income rises.2 Figure 3, below, shows the benefit schedule for a single mother with
two children in North Carolina. This figure shows that the benefit phase out does not present a
problem when considering only earnings and refundable tax credits, but as soon as this single
mother also receives SNAP and Medicaid — a common occurrence—she faces a cliff as her
earnings cross into $30,000 to $35,000 per year. In this scenario, she would need to increase her
annual earnings from $30,000 per year to $60,000 per year to completely overcome her loss in
benefits.




2
  See Georgia Center for Opportunity, https://foropportunity.org/benefitcliffs/ and Federal Reserve Bank of Atlanta,
https://www.atlantafed.org/what-we-study/workforce-development/advancing-careers-for-low-income-families/what-are-benefits-
cliffs.

                                                             4
Figure 3: Safety-Net Benefits Based on North Carolina Policies: A Single Mother with Two
Children




The situations I have described are not simply an academic exercise. The misalignment of program
rules affect behavior and discourage safety net program participants from increasing their work
effort and income. At a time when many businesses across the country are struggling to find
workers, these dynamics also create a drag on local economies.

Survey and focus group research confirm these behavioral influences. For example, researchers for
the U.S. Department of Health and Human Services conducted a survey of low-wage workers in
2024. They found that when workers were presented with a scenario that included a high effective
marginal tax rate, they reported being less likely to take a higher-paying job. Specifically, more than
one-third of respondents said that they would not accept a higher-paying job if they faced a 69
percent marginal tax rate – that is, if they lost 69 percent of their additional earnings to benefit losses



                                                    5
they would turn down a pay raise.3 The Sutherland Institute in Utah conducted a representative
survey of Utah residents in 2024 and found similar results: Nearly 60 percent of past safety net
program participants somewhat or strongly agreed that they were concerned that earning extra
income would result in a loss of benefits that would make them overall worse off.4 This sentiment
translated into behavior changes. According to the survey, nearly 30 percent of past program
participants said they had previously decided not to look for a new job, raise, or promotion because
they believed it would make their family financially worse off.5

Conclusion
The safety net for low-income families should offer a path out of poverty toward opportunity. But
as I have argued, our federal safety net policies instead contain benefit cliffs that discourage more
work and higher pay, a problem that compounds when families receive multiple benefits.
Furthermore, these problematic design features negatively affect employment—related behavior,
harming individual workers and the employers who rely on them.

The benefit cliff workgroup I led, proposed several administrative and legislative reforms to address
benefit cliffs.6 Table 1 summarizes the recommendations from our report. Specifically, we proposed
administrative reforms to individual programs by states and federal agencies: States should use the
discretion they have to alter individual program rules, and federal agencies should allow states to
utilize administrative waivers to address cliffs. Additionally, we proposed several legislative reforms
for Congress: First, Congress should legislate individual program changes to better align benefit
rules, such as in SNAP, and second, Congress should legislate additional flexibility at the state level
to allow cross-program reforms. Notably, Senator Husted (R-OH) of Ohio has introduced the
Upward Mobility Act, which would allow states to pilot test the consolidation of federal programs
designed to address benefit cliffs.7 Finally, we proposed several comprehensive legislative reforms to
the safety net, such as consolidating programs and standardizing program rules. These reforms
would address benefit cliffs, as well as other longstanding problems with the overall safety net for
low-income families in the U.S.

Low-income working individuals deserve a safety net that helps them escape poverty, rather than
one that traps them in government dependency and stagnation. Employers across the country also
deserve a flexible labor market free from government disincentives to work. To achieve these ideals,
our safety net programs need reform to eliminate benefit cliffs.

Thank you and I look forward to answering your questions.



3
  Spitzer, Ariella, Jesse Chandler, Bernadette Hicks, and Daniel Thal (2024). Understanding Economic Risk for Low Income
Families: Economic Security, Program Benefits, and Decisions about Work. OPRE Report #2024-324, Washington, DC; Office of
the Assistant Secretary for Planning and Evaluation; and Office of Planning, Research, and Evaluation, Administration
for Children and Families, U.S. Department of Health and Human Services.
4 Nic Dunn, “Strengthening the American Dream: Addressing benefits cliffs to empower safety net participants to

pursue work and opportunity”, Sutherland Institute Report, November 9 2024,
https://sutherlandinstitute.org/strengthening-the-american-dream/.
5 Ibid.
6 Angela Rachidi, Matt Weidinger, Joshua Bandoch, Nic Dunn, Leslie Ford, and Erik Randolph, Stranded by the Safety Net:

How to Fix the Benefit Cliff Problem (Washington, DC: American Enterprise Institute, 2025).
7 See legislative text for the Upward Mobility Act, https://www.congress.gov/bill/119th-congress/senate-

bill/3583/text.

                                                            6
         Table 1. Immediate and Comprehensive Reforms to Address Benefit Cliffs
                                     Immediate Reforms
Individual Program     States have some existing administrative authority to reduce benefit
Administrative         cliffs in individual programs. They often can make administrative
Reforms                decisions, such as frequency of benefit reauthorizations that can make
                       benefit cliffs better or worse. They also have existing authority to
                       request waivers of federal rules, which can be used to lessen benefit
                       cliffs. We outline several key steps states can take to use their existing
                       administrative authority to offer immediate relief to families at risk of a
                       benefit cliff.
Individual Program     Where states do not have existing administrative flexibility, Congress
Legislative Reforms    should legislate changes to individual program rules to lessen benefit
                       cliffs. We offer one example for the Supplemental Nutrition Assistance
                       Program (SNAP), recommending that Congress statutorily change
                       SNAP’s benefit level, deductions, and eligibility limit to address benefit
                       cliffs. This can serve as a model for other program reforms.
Cross-Program          While individual program reforms are important, benefit cliffs are worse
Administrative         when households receive multiple benefits. We outline how federal
Reforms                agencies, through existing waiver authorities, can guide states on
                       soliciting federal approval for demonstration projects aimed at reducing
                       cross-program benefit cliffs. We also identify ways federal agencies can
                       revise rules and regulations to make it easier for states to address cross-
                       program benefit cliffs, such as coordinating how states claim costs
                       across programs.
Cross-Program          Where existing waiver authority falls short, Congress should legislate a
Legislative Reforms    “superwaiver,” which would allow states to consolidate and coordinate
                       multiple safety-net programs while maintaining federal accountability
                       for results. This would allow states to innovate and test processes for
                       consolidating programs to improve delivery and efficiency. The results
                       would help inform comprehensive reforms.
                                   Comprehensive Reforms
Consolidate            Congress should legislatively combine safety-net programs and
Programs and           coordinate administrative functions in several key areas. This would
Reduce Redundancy improve efficiency, eliminate overlapping services, and yield long-term
                       taxpayer savings. We recommend that Congress establish select
                       committees to oversee this approach and use “superwaiver”
                       demonstration projects to inform this effort.
Standardize            Congress should standardize eligibility rules across programs. This
Eligibility Rules and would involve aligning benefit levels and taper rates and creating
Benefit Phaseouts      program exit thresholds across programs so that families experience a
Across Programs        consistent, predictable phaseout of government assistance.
Streamline Delivery    Congress should also streamline administrative systems by giving all
Systems Across         states the ability to adopt the “One Door” approach that Utah uses
Programs               under specialized authority from the federal government. A One Door
                       approach creates a more efficient and effective system by allowing
                       participants to seamlessly access benefits and connect to employment


                                                7
                             and job training. It helps address benefit cliffs by better coordinating
                             benefits and informing participants of program rules and regulations.
Source: Angela Rachidi, Matt Weidinger, Joshua Bandoch, Nic Dunn, Leslie Ford, and Erik
Randolph, Stranded by the Safety Net: How to Fix the Benefit Cliff Problem (Washington, DC: American
Enterprise Institute, 2025).




                                                   8


File and source

File

Rachidi_Testimony.pdf

Size

590,047 bytes

SHA-256

103657b46b4c51953de2cae7c2baf9f721725c2d7ebb3f0af37f1ef8aeb9c04f

Our copy

Rachidi_Testimony.pdf

Original

No public link identified.