Editor’s note: House Republicans and the Trump administration have proposed deep cuts to the Supplemental Nutrition Assistance Program — also known as food stamps — over the next 10 years. Diane Whitmore Schanzenbach, director of the Hamilton Project at the Brookings Institution and professor at Northwestern University, and Robert E. Rector, senior research fellow at the Heritage Foundation, discussed justifications for the proposed cuts. The email discussion was moderated by Post Opinions assistant editor Robert Gebelhoff and has been edited for style and clarity
Robert Gebelhoff: The White House has proposed more than $190 billion in cuts to SNAP over the next 10 years by adjusting eligibility rules and amping up work requirements. To both of you, is this a step in the right direction? Why or why not?
Diane Schanzenbach: First, thanks so much for inviting me to discuss this important issue with you, I’m looking forward to the conversation. In my opinion, the Trump administration’s cuts to the SNAP program are ill-advised. SNAP is a model program and has been a clear policy success. It lifted nearly 5 million people out of poverty in 2014 (the most recent data available), efficiently targets families who need benefits the most, reduces the likelihood that families have trouble affording food, serves as an automatic fiscal stabilizer in times of economic downturns, and has extremely low rates of both error and fraud.
SNAP also has long-term benefits. My own recent research study found that those who had access to SNAP benefits during childhood grew up to be healthier, and women were especially more likely to become economically self-sufficient due to childhood access to SNAP benefits. Access to food stamps also increased high school graduation rates by more than 18 percentage points.
Beyond reducing food insecurity, SNAP improves households’ financial well-being. When food-insecure households have access to the program, they have more resources available for other essential expenses, such as housing, utilities and medical bills.
Robert Rector: Your question mischaracterizes the Trump proposal. President Trump proposes two reforms to SNAP, raising the state contribution share and establishing national work requirements.
The Trump budget recognizes that the food stamp program will become more efficient if the state governments that operate the program have “skin in the game.” Therefore, it raises the required state contribution to food stamps incrementally from 8 percent to 25 percent. By 2027, this would cost state governments an extra $14 billion per year. Half of the so-called cuts in food stamp spending in the Trump budget simply represent this modest shift from federal to state funding.
Government spends around $350 billion per year providing cash, food and housing benefits to low-income persons. Around 90 percent of this spending comes from the federal government. It makes no sense in a federal structure of government for the federal government to pay the full costs of national defense, Social Security and Medicare while also picking up 90 percent of the cost of supporting the poor. States should bear a greater share.
The other element of the Trump proposal would require able-bodied, non-elderly recipients to work or prepare for work as a condition of receiving aid. Polls show nearly 90 percent of the public agree with this principle. They believe welfare should not be a one-way hand-out.
Gebelhoff: I’d like to discuss that last point in more detail. Research shows that most recipients who receive food stamps do work, and rates of SNAP recipients who do work have steadily risen as a result of policy reforms over the past few decades. If able-bodied recipients already show, for the most part, that they’re willing to work, what’s the justification for more work requirements?
Rector: This again is factually incorrect. For example, there are some 4.2 million able-bodied adults without dependents (ABAWDs) receiving food stamps. In a given month very few of these are employed. Also roughly half of parents on foods stamps are not employed while receiving stamps. The point of the work requirement is to nudge employable individuals toward work. This is the same principle that was the basis of welfare reform which replaced Aid to Families with Dependent Children with Temporary Assistance for Needy Families in 1996. The president seeks to apply the principles of welfare reform to the food stamp program.
No recipients will simply have their benefits cut. Instead they will be required to undertake “job activation” including supervised job search, job preparation, training or community service. Experience from the 1996 reform shows this causes a dramatic surge in employment and a substantial drop in poverty while reducing caseloads.
Schanzenbach: I think that Robert [Rector] is mischaracterizing the Trump proposal. In addition to requiring a state contribution and new work requirements, there are several other changes proposed. For example:
Eligibility for low-income working families that have gross incomes above 130 percent of the poverty line, but have high spending on childcare, rent and other deductible expenses, would be terminated by terminating the Broad Based Categorical Eligibility rule.
Benefits will be capped based on household size to families of six, penalizing larger families.
The $16 minimum monthly benefit, which primarily goes to low-income seniors and people with disabilities, would be eliminated.
Also, the work requirement/time limit would restrict time-limit waivers to areas with at least 10 percent unemployment — a much higher threshold than used in the past. It’s one thing to have a work requirement/time limit during good economic times — and indeed I also support a meaningful work requirement for SNAP during normal economic times — but another to continue to enforce work during times with high unemployment rates like we saw during the Great Recession.
Rector: Again, nearly 90 percent of the public agree that “able-bodied adults who receive cash, food, housing, or medical care should be required to work or prepare for work as a condition for receiving that government aid.” Trump is putting that principle into effect. If an individual cannot immediately find a job, they would be expected to undertake constructive activities that will lead to work. For example, the 4.2 million ABAWDs who get food stamps might be asked to perform six hours of community service per week in exchange for benefits. What is wrong with that?
The left does not oppose work requirements directly because they are so overwhelming popular with the public. Instead, they ensure quietly that the requirements are never actually put into effect. That is what is being debated here.
Schanzenbach: Again, Robert is mischaracterizing the work requirements. Job search doesn’t count, and in the vast majority of cases six hours of community service doesn’t count. The rules are: a minimum of 20 hours per week, and states are not required to offer individuals a job or training program, and in fact the law limits the characteristics of training programs that a state can provide.
On the broader point, SNAP’s basic structure supports work. Households that meet the program’s eligibility rules and requirements can qualify for benefits and be served if they apply. As a result, SNAP is able both to respond quickly and effectively in recessions and to act as a longer-term support for low-wage workers. SNAP typically boosts low-wage workers’ income by 10 percent or more.
The number of households that have earnings while participating in SNAP has more than tripled — from about 2 million in 2000 to about 7 million in 2014 (that’s about a 5 percent increase in the total share of households receiving benefits).
Most SNAP recipients who can work do so. Among households with at least one working-age, non-disabled adult, more than half work while receiving SNAP — and more than 80 percent work in the year before or after receiving SNAP. The rates are even higher for families with children — more than 60 percent work while receiving SNAP, and almost 90 percent work in the prior or subsequent year.
More specifically, Robert mischaracterizes the work requirements for ABAWDs. They may receive SNAP benefits for only three months in a three-year period, unless they are either employed for at least 20 hours per week or engaged in a workfare or training activity. This does not include job search. In addition, states are not required to offer those individuals a job or training program, and the law limits the characteristics of training programs that a state can provide. I think that we could do better, allowing states to craft work requirements that are better tailored to their local economic conditions and available education and job training programs.
Rector: President Trump is obviously proposing to change those rules along the lines of reforms in Maine and Nebraska. Bear in mind there are over 4 million ABAWDs idle on the rolls.
Gebelhoff: Diane, I want to get back to your point that SNAP serves as an effective “automatic stabilizer” — that is, the program automatically helps people when the economy goes sour. But over the past two recessions, participation in the program has risen at unprecedented levels. Participation also hasn’t fallen as quickly as the economy recovered. What does this mean for the program as an automatic stabilizer? Why shouldn’t the government be concerned with reducing participation when the economy is strong?
Schanzenbach: SNAP is unquestionably a good economic stabilizer for the economy as a whole. During the height of the Great Recession, every dollar in SNAP spending was estimated to yield $1.73 in increased economic activity. In more normal economic times, the multiplier is $1.22— estimated to have the highest return of any stimulus spending program.
The SNAP caseload response to the Great Recession is in fact a sign that the program works. It is intended to expand in times of greater need, and in fact, these increases were temporary and we have already started to see a reduction in caseloads. Remember, the Great Recession was the deepest downturn our economy has seen since the Great Depression, and it isn’t surprising that many more Americans were eligible for benefits and takeup increased.
As the economy strengthens, data show that the caseloads have started to decrease. Caseloads fell by 2 percent in both 2014 and 2015. According to the Center of Budget and Policy Priorities, in about 40 states, caseloads were lower in fiscal year 2015 than in fiscal year 2013, and caseload declines are continuing, due in part to the stronger economy. The Congressional Budget Office predicts this number will continue to fall in the coming years.
Importantly, the good news is also that food insecurity – the primary problem that SNAP addresses – is also starting to decline from its recession spike.
Rector: The main problem is that 92 percent of food stamp spending is funded by the federal government in good or bad economic times. One of the key lessons from welfare reform 20 years ago is that state governments, whether blue or red, spend their own revenue far more prudently than they spend “free money” from Washington. Requiring state governments to have much more skin in the game in normal economic times means they would operate the program more efficiently.
In very bad economic times the federal share could increase as an automatic stimulator similar to the unemployment insurance program. But ordinarily states should bear a much greater share of costs. Welfare (cash, food and housing for low-income persons) is almost exclusively funded by the federal government. This makes no sense and should be changed.
Gebelhoff: I’d like to ask just one closing question that’s more forward looking: What are your predictions for the proposed cuts? How should Congress approach these proposals?
Rector: I believe that legislation establishing work requirements on able-bodied non-elderly recipients and an enhanced state match in food stamps will be enacted within the next two years. These were the key principles of successful welfare reform 20 years ago and President Trump is seeking to apply them more broadly across the welfare state. The weakness of left-wing opposition to these ideas can be seen in their obfuscation. For 20 years they have supported work requirements in principle while always opposing them in practice.
Schanzenbach: Thanks for this question — people don’t usually ask economists to weigh in on politics! Upon second thought, I guess there is good reason for that, and I’ll demur.
My advice to Congress, though, is not to fix a program that isn’t broken. The program as currently structured is an efficient and effective safety net. It’s a good investment: It reduces food insecurity, lifts people out of poverty, helps the economy and has long-term benefits for the children that it serves.
Thanks for hosting this debate, I enjoyed corresponding with you both, and hope we can continue to engage in productive debates about ways to strengthen the American economy.