Off the Charts Blog | Center on Budget and Policy Priorities

Even as the House prepares this week to cut SNAP (formerly food stamps) by $21 billion and push 2 million low-income people off the program, new research shows that SNAP is the most effective program pushing against the steep rise in extreme poverty.

The number of households with children living on $2 or less per person per day — one definition of poverty the World Bank uses for developing nations — more than doubled between 1996 and 2011, to 1.6 million, according to research by the University of Michigan’s H. Luke Shaefer and Harvard University’s Kathryn Edin.

While these findings are troubling, the authors also show that SNAP kept more households with children out of extreme poverty than any other government program.

Counting SNAP benefits as income cuts the number of households with children in extreme poverty in 2011 by 48 percent, from 1.6 million to 857,000 (see graph).

SNAP also cut, by roughly half, the rise in extreme poverty among households with children between 1996 and 2011, the study found.

The study provides strong evidence that one of the main drivers of rising extreme poverty was the decline in cash assistance since the 1996 welfare reform.  Aid to Families with Dependent Children (AFDC) cut extreme poverty significantly in 1996, the study found, but Temporary Assistance to Needy Families (TANF) — which replaced AFDC after 1996 and reaches many fewer needy families — had a much smaller impact in 2011.

For many of the poorest Americans, SNAP has become the only form of income assistance they receive.

One reason SNAP is so effective in fighting extreme poverty is that it focuses its benefits on many of the poorest households.  Roughly 91 percent of monthly SNAP benefits go to households below the poverty line, and 55 percent go to households below half of the poverty line (about $9,800 for a family of three).  One in five SNAP households lives on cash income of less than $2 per person a day.

The farm bill before the House would move SNAP in the wrong direction by cutting or eliminating food assistance to large numbers of low-income families.

As our new report explains, some senators are proposing amendments to the immigration bill that would make its long and difficult path to citizenship far more difficult — in some cases undermining the fundamental goal of enabling undocumented workers to legalize their status.

One such proposal, from Senators Orrin Hatch (R-UT) and Marco Rubio (R-FL), would require immigrants seeking legal status to prove they have paid all of their taxes since they entered the country before adjusting to a legal status.

On the Senate floor Wednesday, Senator Hatch claimed that the IRS is well-positioned to determine such individuals’ tax liability, even when workers lack earnings and tax records:

The IRS is well experienced at estimating the tax liabilities for people who, for whatever reason, lack the records that normally support a tax return. . . .  Using bank records, credit card statements, housing records, and other evidence of an individual’s lifestyle, the IRS is able to construct returns and estimate tax liabilities for nonfilers who are U.S. citizens and resident aliens.  The same process can be used for immigrants looking to certify they no longer owe any Federal taxes.  That is not a tough thing to do…

But the IRS does not routinely try to use such records to guesstimate people’s incomes.  These are extreme methods, which it resorts to in only a very small number of cases each year.

Senator Hatch is essentially arguing that it would be workable — and a good use of taxpayer money — to require the IRS to conduct extensive field audits of very large numbers of undocumented immigrants seeking legal status.

The audits would be very complicated in many cases; many of these workers’ past employers will be difficult to locate, will not have records for cash transactions made years ago, and (for obvious reasons) may be reluctant to cooperate.  Moreover, many of these immigrants likely will not have bank records or credit cards, will have moved many times, and will have been working in the United States for many years or even decades, so reconstructing their records for this entire period could prove extremely difficult.

The IRS’s experience in estimating a family’s income based on its “lifestyle” is largely restricted to a very small number of generally high-income people and small businesses where there is a large disconnect between their reported income and the person’s lifestyle or the business’s spending and the IRS has reason to believe large-scale tax evasion may have occurred.  The IRS has never used this method for large numbers of low-income workers.

Given that an estimated 11 million people are eligible to legalize under the bill, the increased workload for the IRS would be tremendous.  Of the 187 million individual and business tax returns filed in 2011, the IRS examined just 1.7 million — fewer than 1 percent.  Moreover, over 70 percent of these audits were conducted by mail (where information is requested and provided by mail) and were not the complex, intensive audits that Senator Hatch envisions.

Based on the total cost of current IRS enforcement efforts and the number of returns examined each year, each field examination would clearly cost thousands of dollars.  This means that the cost of conducting vast numbers of complicated audits, as Senator Hatch evidently envisions, would run into the billions.

Senator Hatch has not proposed any new funding for the IRS to conduct such audits or indicated that large numbers of new IRS staff would be hired.  If the IRS tried to implement his proposal, it would have to divert a very large portion of its total enforcement resources, probably for a number of years, from enforcement activities that yield a much higher return for the Treasury to auditing millions of legalizing workers, most of whom have modest incomes.

In 2012, the IRS conducted 31,700 field audits of corporations, leading to a total of $20 billion in recommended additional tax assessments — an average of more than $600,000 per review.  When both individual and business field audits are considered, the average tax assessment is close to $60,000.  Diverting IRS resources away from these high-return audits to try to reconstruct earnings records from modest-earning immigrants almost surely would result in a net loss of revenues to the Treasury, not the gain that Senator Hatch suggests.

More realistically, the IRS would never be able to conduct the required audits, so the process of legalizing undocumented workers would grind to a halt for many — leaving the basic goal of the legislation in tatters.  Undocumented workers would remain undocumented and the Treasury would collect less in taxes going forward than if these workers were allowed to come out of the shadows.

The Urban Institute has published a new commentary collection on the first year of Work Support Strategies (WSS), a five-year, multi-state project that is designing, testing, and implementing more effective, streamlined, and integrated approaches to delivering key supports for low-income working families.  In my contribution, I outline lessons that we’ve learned from past performance and look ahead to the initiative’s future as states implement health reform:

Most guidance from the federal government on the [Affordable Care Act, or] ACA so far has been focused on changes to health programs.  While this information is crucial, it neglects the overwhelming majority of states administering Medicaid and human services jointly. WSS states play a crucial leadership role in discovering how to blend new Medicaid eligibility rules and customer service standards with delivery of other human services programs.  They’re using lessons of the past and assessing how they can achieve new efficiencies to test creative solutions to provide all the necessary benefits to their most vulnerable people.  They will be innovators on the path to creating modern, efficient systems that serve the poorest families comprehensively while delivering first-class customer service at a low cost to the taxpayer.

Click here to read the whole commentary collection on WSS, including my piece and 11 others.

In today’s Senate debate on the farm bill, Senator David Vitter offered — and Senate Democrats accepted — an amendment that would increase hardship and will likely have strongly racially discriminatory effects.

The amendment would bar from SNAP (food stamps), for life, anyone who was ever convicted of one of a specified list of violent crimes at any time — even if they committed the crime decades ago in their youth and have served their sentence, paid their debt to society, and been a good citizen ever since.  In addition, the amendment would mean lower SNAP benefits for their children and other family members.

So, a young man who was convicted of a single crime at age 19 who then reforms and is now elderly, poor, and raising grandchildren would be thrown off SNAP, and his grandchildren’s benefits would be cut.

Given incarceration patterns in the United States, the amendment would have a skewed racial impact.  Poor elderly African Americans convicted of a single crime decades ago by segregated Southern juries would be among those hit.

The amendment essentially says that rehabilitation doesn’t matter and violates basic norms of criminal justice.

It’s also possible that the amendment could contribute to recidivism.  Ex-offenders often have difficulty finding jobs that pay decent wages.  The amendment could pose dilemmas for ex-offenders who are trying to go straight but can neither find jobs nor, as a result of the amendment, obtain enough food to feed their children and families.

Senator Vitter hawked his amendment as one to prevent murderers and rapists from getting food stamps.  Democrats accepted it without trying to modify it to address its most ill-considered aspects.

The farm bill is still on the floor, and the amendment can still be modified.  Senators should gather the courage to step up to the plate and address this matter forthwith.

My latest post for U.S. News & World Report’s Economic Intelligence blog addresses conservative critics’ argument that SNAP (the Supplemental Nutrition Assistance Program, formerly food stamps) is “broken” and must be “reformed.”  In reality, as our recent blog series shows, SNAP expanded as it’s supposed to during the severe recession of 2007-2009 and subsequent slow recovery and will shrink as the economy improves.

But why, as critics note, did SNAP enrollment continue rising after 2009, even as unemployment began to fall?  Because, it’s not unusual for poverty and hardship to continue rising even after unemployment peaks.

Moreover, the last few years have been very different from a typical recession and recovery, as our Legacy of the Great Recession chartbook shows.  The unemployment rate thus has been a relatively poor indicator of the state of the labor market, for two reasons in particular.

First, the unemployment rate doesn’t include the many people who want a job and would likely have one in a stronger labor market but haven’t looked enough to count as officially unemployed.  Nor does it include the many people who would like to work full time but can only find part-time work.

The graph below, which highlights the share of the population with a job (the so-called employment to population ratio), paints a grimmer picture of what’s happened to employment.  Part of the sharp decline reflects higher unemployment, but the rest reflects a decline in labor force participation.

The Labor Department estimates that 22 million Americans who want to work either don’t have a job or are working only part-time when they want to work full-time.

Second, the unemployment rate doesn’t tell us about long-term unemployment — those out of work for at least 27 weeks — which remains historically high (see second graph).  With the deep and prolonged recession and weak recovery, SNAP has become increasingly valuable for the long-term unemployed, since it’s one of the few resources available for people who have exhausted their unemployment benefits.

In short, the number of people qualifying for and receiving SNAP benefits is still high because unusually high unemployment, reduced incomes, and limited job opportunities all persist.  The best way for policymakers to lower SNAP costs would be to aid the economic recovery to create jobs and boost incomes.