Here’s the uncomfortable reality:

When “welfare” is defined broadly—cash aid, food support, and medical coverage—long-term participation can remain high even after many years in the country.  A recent, widely circulated analysis of Minnesota households headed by Somali immigrants, based on pooled American Community Survey data, reported that the vast majority used at least one major means-tested program, and that use among those with 10+ years of residency was only modestly lower.

A congressional hearing exchange amplified a similar figure—claiming 78% remained on “welfare” after ten years—though the claim was presented in questioning and depends heavily on definitions and measurement choices.

That’s precisely why we need to slow down and define the problem correctly—because the solution depends on what we mean by “welfare.”

What Counts as “Welfare” (and Why It Matters)

In everyday conversation, “welfare” can mean one thing.

In policy, it can mean several:

• Cash assistance (e.g., TANF or state equivalents) is the classic “welfare check.”
• Food benefits (SNAP) help households buy groceries.
• Health coverage (Medicaid) is often the most significant benefit by dollar value.

But these programs do not operate the same way.

TANF is time-limited at the federal level: states generally can’t use federal TANF funds for adults beyond 60 cumulative months (with limited hardship exceptions).  Meanwhile, SNAP and Medicaid are typically eligibility-based rather than lifetime—capped, meaning a household can qualify on and off for years as income fluctuates.

“If we treat every benefit like a ‘check,’ we’ll misdiagnose the problem—and prescribe the wrong cure.”

The Trap Isn’t Always Laziness.  Sometimes It’s Math.

For a family of 4, some welfare systems pay more after taxation and benefits than a college graduate’s teacher’s salary.  So, why would an enterprising parasite be foolish enough to expend energy on productive work when the handouts are so luscious?

A serious welfare discussion must separate two issues that often get mashed together:

·       Work avoidance (a behavioral problem)

·       Low earning power (a human capital problem)

In many long-duration cases, the core issue is the second.  When households face low formal education, limited English proficiency, weak credential recognition, or concentrated poverty, assistance can remain “necessary” even with employment, because wages don’t cover the basics.

The Minnesota Somali case study illustrates this dynamic: the same analysis reporting high benefit usage also documents high poverty rates and significant language barriers.  Whether or not you agree with every conclusion of that report, it underscores an important point: high benefit usage often follows predictable inputs—poverty, family size, and limited earnings capacity.

“A job is not the same as a living wage—and policy fails when it pretends otherwise.”

What Research Says About Refugees Over Time

Zooming out from individual communities reveals a more nuanced, broader body of evidence.

A major federal study from HHS (ASPE) examining refugees and asylees over a 15-year window found that, in aggregate, they contributed a net positive fiscal Impact over time—even though public costs are often front-loaded during early resettlement.  The report also notes that refugees and asylees with 10+ years of residency had income levels that approximated those of the overall U.S. population, consistent with the idea that integration can improve substantially over time.

Academic work using ACS data finds that refugees tend to arrive with lower educational attainment and higher initial use of benefits, but outcomes can improve considerably.  One well-known NBER working paper reports that refugee employment rises over time and can eventually exceed native employment rates, even if earnings remain lower.

Importantly, the federal government’s Annual Survey of Refugees is designed to measure early integration outcomes—employment, education, well-being, and public assistance—over the first five years, because the early period is when policy choices matter most.

Front-loaded costs don’t prove permanent dependence.  But persistent reliance does prove policy failure.

So Why Does “Long-Term Welfare” Persist?

If we want welfare to be temporary, we need to identify the mechanisms that make it stick.  In practice, several repeat offenders show up:

1) The Benefit Cliff

Many programs phase out quickly as earnings rise.  A modest raise can trigger the loss of childcare help, health coverage, or food support—leaving a household worse off for working more.

“When a raise makes you poorer, the system is paying people to stand still.”

2) Credential and Language Bottlenecks-employers pay for productivity, not effort.  When credentials aren’t recognized, or English proficiency remains limited, workers get stuck in low-wage lanes where assistance remains necessary, underscoring the importance of targeted language and credential recognition policies.

Employers pay for productivity, not for effort.  If credentials aren’t recognized or English proficiency remains limited, workers get stuck in low-wage lanes where assistance remains necessary.

3) Housing and Childcare Costs

In high-cost regions, even full-time work can’t keep up with rent and childcare.  That turns short-term assistance into a semi-permanent supplement.

4) Program Design That Measures Compliance, Not Mobility

Too many systems reward the completion of paperwork rather than wage progression.  “Participation” becomes the metric—not independence.

A Better Model: Welfare With an Exit Ramp

If policymakers want to restore welfare’s original moral logic—help in crisis, then forward motion—the solution is not a single blunt instrument.  It’s a package of incentives and supports engineered for exits.

Here are reforms that have bipartisan potential:

1) Smooth the Benefit Cliff

Phase benefits out gradually, so work always pays.  This is an engineering problem, not a sermon.

2) Tie Aid to Progress, Not Punishment

Replace punitive churn with clear milestones:

• verified job placement
• wage gains over time
• credential completion
• English proficiency benchmarks
  (especially for newcomers facing language barriers)

3) Shift Workforce Funding Toward Outcomes

Pay providers for job retention and wage growth, not intake volume.  If the goal is independence, measure independence.

4) Treat Fraud Enforcement as a Scalpel

Accountability matters.  But broad suspicion aimed at an entire community is both unjust and counterproductive.  Investigate fraud where the evidence leads—while preserving the moral purpose of aid.

5) Be Honest About Selection and Resettlement Strategy

If a locality repeatedly absorbs large numbers of arrivals with limited earning power, the welfare burden will rise predictably.  That’s not xenophobia—it’s arithmetic.

The Point Isn’t Less Compassion.  It’s More Mobility.

A decent society provides a safety net.  A wise society designs it with an exit.

If a household remains dependent for ten years, we should resist the lazy explanation (“they’re choosing this”) and the equally lazy denial (“nothing can be done”).  The truth is more complicated—and more hopeful: policy architecture shapes behavior, and better architecture can produce better outcomes.

“The goal isn’t to cut people off.  The goal is to build a system they can climb out of.”

When the Safety Net Becomes a Ceiling

Welfare is meant to help families regain their footing—not quietly redefine what “normal” looks like.

On a cold morning in Minneapolis, you can watch two American stories unfolding at once.  In one, newly arrived families hustle through English classes, job programs, and the maze of appointments that come with rebuilding a life from scratch.  In the other, the state’s safety net—one of the most generous in the country—strains under the weight of long-term need, rising costs, and growing public suspicion.

Both stories are real.  And they collide in a question that’s become harder to avoid:

When does help stop being a bridge—and start becoming a ceiling?

Start with a definition: “welfare” isn’t one thing

The term welfare is used as if it describes a single program, a single check, or a single behavior.  It doesn’t.  In public policy, “welfare” usually means some combination of:

• Cash assistance (often linked to Temporary Assistance for Needy Families, or TANF)
• Food benefits (SNAP)
• Public health insurance (Medicaid or similar coverage)

Those programs behave very differently.  Cash assistance is the most time-limited: federal rules generally prohibit using federal TANF funds to assist a family with an adult who has received federally funded assistance for 60 cumulative months (with limited hardship exceptions).  However, Medicaid and SNAP are primarily eligibility-based, not lifetime-capped in the same way, meaning a household can qualify—sometimes for long periods—if income remains low.

If we treat every benefit like a “check,” we’ll misdiagnose the problem—and prescribe the wrong cure.

Do refugees stay on welfare “as a way of life”?

Here’s what the best broad, nonpartisan integration research suggests: most do not—especially when you define welfare narrowly as cash assistance.

A Migration Policy Institute (MPI) fact sheet summarizing national data found that public benefit use declines with length of residence and that, after 10 years, much of the gap between refugees and the U.S.-born closes.  In the 2009–11 period, the same MPI summary reported:

• Less than one-quarter of refugee households with 10+ years of U.S. experience received food stamps (SNAP).
• Only 3% of refugee households received cash welfare, compared with 2% of U.S.-born households.
• Fewer than 15% of refugee adults had public health insurance after a decade in the U.S., compared with 11% of U.S.-born adults.

Those are not the numbers for a population that uses welfare as a permanent lifestyle, at least not in the cash-assistance sense.  They do, however, underscore an essential point: nutrition and health programs can persist longer than cash aid, particularly for low-income families.

Meanwhile, academic work using American Community Survey data similarly finds that refugees’ early outcomes can include “high welfare use,” but that outcomes improve substantially as time in the U.S. increases.

Cash welfare tends to be a short chapter.  Health and food benefits can become the extended epilogue when wages remain low.

The Minnesota/Somali case: why one place can look so different

National averages can obscure what’s happening in specific communities, particularly where arrivals concentrate, and barriers accumulate.  Minnesota—home to one of the largest Somali communities in the United States—has become a focal point in this discussion.

A Center for Immigration Studies analysis of Minnesota households headed by Somali immigrants, using pooled American Community Survey data, reports very high participation in at least one means-tested program—including Medicaid and SNAP—and notes that usage among those with more than 10 years of U.S. residency is only “marginally lower” than the overall Somali immigrant population in the dataset.  The same report attributes a significant portion of the story to measurable barriers, including poverty rates, educational disparities, and limited English proficiency.  [

In the political arena, a congressional hearing exchange popularized a claim that 78% of Somali immigrant households remain on “welfare” after 10 years—a figure asserted during questioning and heavily dependent on how “welfare” is defined (often meaning “any” public benefit, not cash assistance alone).

Both narratives can be authentic at once.

• Nationally, the use of refugee benefits tends to decline sharply over time, with cash welfare becoming rare after a decade.
• Locally, some communities can show persistent high participation in means-tested programs—primarily Medicaid and SNAP—when poverty, language barriers, and labor-market segmentation persist.

Integration isn’t one number.  It’s a curve—and some communities get stuck on the flat part.

The real culprit isn’t always “dependency.” Sometimes it’s the math.

If we want to reduce long-term reliance on benefits, the most productive place to look is not character—it’s incentives and constraints.

1) The benefit cliff

Many benefits phase out quickly as income rises, especially for households with children.  That can create a “cliff,” where a small raise leads to a significant loss of support, leaving a family worse off for working more.  While TANF is time-limited, programs such as SNAP and Medicaid are often the supports families rely on as wages increase.

When a raise makes you poorer, the system is paying people to stand still.

2) Low earnings power, not low effort

The Minnesota Somali case study underscores a hard truth: high benefits usage tracks with high poverty, limited credentialing, and language barriers—conditions that suppress earnings even when people are working.  Refugee integration research also shows that refugees often begin with disadvantages in language and education, but can improve substantially over time.

3) Costs that outpaced wages

In many U.S. metros, housing, childcare, and healthcare costs rise faster than entry-level wages.  In that environment, “assistance” becomes less a hammock than a supplement—one that can persist even among working households.

So what should a better system do?  Build an exit ramp.

If welfare is supposed to be temporary, the system must be designed so that exiting is rational.

A few reforms show up again in serious policy discussions:

·       Smooth the cliff.  Phase benefits down gradually, so work always pays more than not working.

·       Measure mobility, not paperwork.  Reward providers and programs for wage gains, job retention, and skills completion—not just enrollment.

·       Invest early where the data shows barriers.  English acquisition, credential pathways, and job matching are not “extras”—they determine whether a household transitions from assistance to independence.

·       Keep the moral purpose—and enforce integrity.  Fraud investigations should follow evidence, not stereotypes; the legitimacy of a safety net depends on public trust.

There’s also a broader fiscal picture that’s easy to miss.  A federal HHS/ASPE study examining refugees and asylees over 2005–2019 estimated a positive net budgetary Impact over time, emphasizing that costs are often front-loaded while contributions rise with years in the country.  That doesn’t negate local burdens, especially for schools and state budgets—but it does argue for better long-term planning rather than simplistic slogans.

A final reality check

It is fair to insist that welfare should not become a retirement plan.  It is also fair to admit that the U.S. economy can trap people in low-wage work where assistance remains necessary.

The data suggests two conclusions a humane society should be able to hold simultaneously:

·       Most refugees do move toward self-sufficiency, and cash welfare use becomes very low after a decade.

·       Some groups and localities struggle for much longer, especially when language, education, and poverty barriers remain entrenched—as the Minnesota Somali case illustrates.

The goal isn’t to cut people off.  The goal is to build a system they can escape from.

Indeed, there are stronger, specific data points you can weave into the article to show how refugee benefit use changes over time and how that national trend contrasts with the Minnesota/Somali case study.

1) Refugee welfare trends — what the data shows (national)

A. Benefit use is highest early, then falls with time in the U.S.

National research summarized by the Migration Policy Institute (MPI) finds that refugees are immediately eligible for key benefits (cash aid, food assistance, health coverage), so usage is higher in the early years, then declines as refugees build earnings and stability.

B. Concrete numbers by time in the U.S. (2009–11 snapshot)

Using American Community Survey-based refugee identification, MPI reports that during 2009–11:

• Food assistance (SNAP/food stamps): Approximately 42% of refugee households in the U.S. with five years or less of residence received food stamps.
• Cash welfare: refugees are eligible upon arrival, and cash use is much higher early on than later; MPI reports that cash welfare receipt declines sharply with years in the U.S.
• After a decade: the MPI “Ten Facts” summary reports that in the same 2009–11 period, only ~3% of refugee households received cash welfare after 10+ years in the U.S. (vs. ~2% U.S.-born).
• After a decade (SNAP), fewer than one-quarter of refugee households with 10+ years of experience received food stamps, compared with 11% of U.S.-born households.
• Health coverage declines, but more slowly: MPI reports that after 10+ years, fewer than 15% of refugee adults had public health insurance, compared to 11% of U.S.-born adults.

Cash welfare tends to behave like a “temporary bridge,” while food/health programs can remain part of the mix longer—especially for low-income households with children.

2) The Minnesota/Somali contrast — why local patterns can diverge

To show why a national “decline over time” trend doesn’t always Play out locally, Minnesota is a vivid case study.

A Minnesota-specific analysis of Somali immigrant–headed households (based on pooled ACS data) reported:

• ~54% of Somali-headed households received food stamps
• ~73% had at least one Member on Medicaid
• ~27% received cash welfare [
• ~81% used at least one major welfare program (broad definition).
• The same analysis states that welfare consumption among Somali households with 10+ years of residency is only marginally lower than the overall Somali immigrant figure in that dataset.

A frequently repeated “78% after 10 years” figure has also circulated in political discussion; it was asserted during congressional questioning and depends heavily on how “welfare” is defined (often meaning “any benefit,” not cash assistance alone).

Mainstream framing note: Present this as a policy and integration challenge (language, credentialing, wage progression, family size, local labor markets), not a moral verdict about a group.  The Minnesota analysis links higher eligibility and use togreaterr poverty and barriers, including limited English proficiency and limited educational attainment.

1: National trend

Nationally, the pattern is consistent: refugees use public benefits heavily at first—and then less over time.  Refugees are generally eligible for key programs immediately upon arrival, so early usage is elevated.

In the 2009–11 period, for example, the Migration Policy Institute reports that approximately 42 percent of refugee households with five years or less of U.S. residence received food stamps.  However, the use of benefits declines with years of residence, and after about a decade, most of the gap between refugees and the U.S.-born population closes.  By the same 2009–11 snapshot, MPI reports that only about 3 percent of refugee households received cash welfare after 10 years, compared with 2 percent of U.S.-born households—suggesting that cash assistance functions more like a short-term bridge than a permanent destination.

2: Different programs, different persistence

The type of benefit matters.  Time limits and work requirements constrain cash assistance.  In contrast, programs such as food aid and public health insurance are primarily eligibility-based and can persist for low-income families, especially those with children.  MPI notes that even after ten years in the country, less than one-quarter of refugee households received food stamps (versus 11 percent of U.S.-born households), and fewer than 15 percent of refugee adults had public health insurance (versus 11 percent of U.S.-born adults).  In other words, “dependency” isn’t a single switch that flips off; it varies by program, by family structure, and by local economic conditions.

3: Minnesota/Somali case as an outlier example

Yet national averages can hide local realities.  In Minnesota, a detailed ACS-based analysis of Somali immigrant–headed households reports exceptionally high participation in means-tested programs: about 54 percent receiving food stamps and about 73 percent with at least one Member on Medicaid, with about 27 percent receiving cash welfare.  The same analysis finds that overall “any welfare” use is approximately 81 percent, and that households with more than 10 years of residence show only modest reductions.

Whatever one concludes from those numbers, the takeaway is not that a safety net is inherently broken—but that when poverty, language barriers, and low earning power persist, benefits can remain a long-term supplement even in communities where many adults work.

Summary:

A functioning welfare system is not judged by how many people it enrolls, but by how reliably it helps people leave.  Federal rules already reflect that logic for cash aid, limiting federally funded TANF assistance to 60 cumulative months for most adults.

 Nationally, the trend line is encouraging: reliance on benefits generally declines over time, and after 10 years, the gap between refugees and the U.S.-born population narrows substantially, according to MPI’s summary of integration outcomes.  Yet Minnesota’s Somali case shows what happens when the ladder to higher wages breaks—when language, schooling, and job pathways fail to translate into earning power and “assistance” becomes long-term supplementation.

 The correct response isn’t to demonize need; it’s to redesign incentives so that work always pays, benefits taper predictably, and programs are accountable for mobility—not paperwork.  That’s how welfare returns to what it was meant to be: a temporary way forward, not a way of life.