How To Fix Child Nutrition Programs
Ensuring that children receive adequate daily nutrition, regardless of family income, should not require programs to be structured like a Soviet-era economy. And yet that is more or less how federal school meal programs are organized, with resources allocated according to a central plan enacted by multiple layers of administrative and committee control. As I explained in an earlier post, the significant changes made to child nutrition programs under the Healthy and Hunger-Free Kids Act of 2010 appear to have only made the program worse. Nonetheless, many of the program’s biggest issues long predate the reform.
In particular, the reimbursement structure under which the National School Lunch Program (NSLP) and the School Breakfast Program (SBP) operates is unnecessarily convoluted, and creates weak incentives for schools and parents to monitor how funds are used. As I argue in this piece, a new payment mechanism currently being piloted by the USDA may provide a pathway for deeper reform. It’s called the Summer Electronic Benefit Transfer card, and it provides a vehicle through which the entire program could be overhauled in a more market-oriented direction.
The Broken Reimbursement System
The Office of Management and Budget has designated both the National School Lunch Program and School Breakfast Program as “high-error”— meaning they report high rates of improper payments in a given year. In fact, among federal programs, School Lunch and Breakfast have the second and third-highest improper payment rates just behind the Earned Income Tax Credit. In 2015, the cost of this error across the two nutrition programs was $2.7 billion per year.
The current reimbursement structure is what makes this level of improper payment possible. Under NSLP and SBP, schools charge differing meal fees to students based on their parent’s income, and are reimbursed according to the distribution of children in each income category. In 2016, schools participating in the NSLP received $3.17 in subsidies for each lunch provided to children below 130 percent of the federal poverty guidelines. Schools are prohibited from charging children from this group any additional cost. For children between 130 and 185 percent of the poverty line, children pay a reduced price for lunch, with the federal government reimbursing at a rate of $2.76 per meal. The third tier is composed of children that do not qualify for fully subsidized or reduced-priced meals, for whom the federal government chips in only $0.30 per meal. Additional payments are made to schools in Hawaii and Alaska, as well as districts with particularly high rates of low-income students.
This reimbursement model presents an immediate problem: parents and school meal administrators have a shared interest in inflating the size of the reimbursement claim that the federal government pays. This was confirmed in 2015 by a USDA study revealing the degree to which both parents and school lunch administrators share the blame for high improper payment rates. Specifically, the study identified certification error and meal-claiming error as the two primary areas through which improper payments in the NSLP and SBP occur.
Principal-Agent Problems in the Current Program
The primary avenue for improper payments is the process through which applications for subsidies in the NSLP and SBP are certified. The majority of students are deemed eligible for either free or reduced-price meals through “direct certification,” whereby eligibility for federal subsidies is determined via federal records showing participation in programs like SNAP, while about 13% and 42.8% of participants in the SBP and NSLP respectively are certified through the submission of an application. Families submitting an application must prove participation in another program that confers eligibility. Students certified via an application are over 3 times more likely to result in an improper payment than those certified directly.
The second avenue for improper payments is the meal-claiming process. This refers to reimbursements claims that are either overstated or do not meet federal standards. Between the two major meals program, the annual cost of this form of improper payment totals to about 1 billion dollars. In other words, local school meal administrators from the cash register attendant up are systematically miscounting the number of meals served.
Miscounting is not necessarily surprising as the safeguards against this sort of behavior are relatively weak. In fact, it is likely not even intentional. Instead, it’s the predictable result of a program designed to create myriad Principal-Agent problems. Principal-Agent problems arise when one party—the Principal delegates— control or execution of a task to another, his or her Agent, creating gaps in both motivation and knowledge. Firms that separate ownership and management invest a lot of resources into solving Principal-Agent problems; for example, to prevent managers from shirking or hiding valuable information. Unfortunately, government programs as large and complex as federal school meal programs tend to exacerbate Principal-Agent problems by making it unclear where accountability lies, and being incredibly slow to adjust in order to correct incentives. Thus, while some may be tempted to blame local administrators for improper payments, the meal-claiming error in particular is much more an indictment of the byzantine list of regulations that schools must comply with, which create the possibility for routine miscounting entirely by accident.
Since neither parents nor schools have any skin in the game with regard to excessive federal reimbursements, there is no reason to expect that they will do anything to minimize improper payments on their own. Nonetheless, in some cases, schools may even actively encourage manipulation of the certification process. In a report investigating lunchroom fraud for Education Next, David N. Bass wrote,
Some districts encourage parents to fill out applications even if they are not sure they qualify. One district in Chillicothe, Missouri, offered parents a $10 Walmart gift card for turning in an application. ‘Even if you choose to pay for your child’s lunches and or breakfasts, each qualified application earns $1,025 per child of state money for our school district,’ said Assistant Superintendent Wade Schroeder.
Fiscal discipline cannot be expected given the incentives of the status quo.
How to Reduce Improper Payment
Rampant payment-error has prompted some to call for expanding the Community Eligibility Provision (CEP) because schools that meet the CEP have an improper payment rate of only 2%, the lowest of all the certification methods. However, this reduces payment-error merely by legitimizing payments that were formerly fraudulent. Since CEP schools offer free lunch to all children, regardless of individual need, there is less room for fraudsters to take advantage of the system. Yet child nutrition programs shouldn’t face a choice between administrative efficiency and effective targeting.
Eliminating the separation between payer and consumer would go a long way to solving Principal-Agent problems of school meal programs. The Summer Electronic Benefit Transfer card currently being piloted offers a vehicle to do just this. Importantly, moving towards an individual benefit care would eliminate the convoluted reimbursement framework entirely. The card would be used to purchase school meals year round, rather than just during the summer break. Placing school meal funds into distinct accounts controlled by the customers would reduce both fraud and error.
SNAP’s Electronic Benefit Transfer (EBT) card is a clear example of how an electronic benefit card can reduce payment error. Since the card’s implementation in 2003, SNAPs error rate has been cut in half. For comparison, the improper payment rate for the NSLP and SBP has essentially remained constant.
Put Children and Parents Back In Control
Shifting authority over the certification process from schools to the federal government would also remove the principal-agent problems faced by having school districts determine their own reimbursement level. Given that a majority of all improper payments stem from school lunch administrators, such a move could easily save over 1 billion dollars per year. Schools would benefit as well through lower administrative costs.
Transforming the school nutrition payment structure into a single card would also enable the introduction of market features into an otherwise highly bureaucratic program. Currently, low-income children that choose not to participate in these programs cannot opt-out in favor of a cash payment to purchase food to bring from home. Children, like the general population, have differing tastes and dietary needs. The option of being able to allocate one’s food options elsewhere would be a significant improvement over the status quo.
Of course, the first solution would be to eliminate school meal programs entirely, and use the savings to implement a simple cash-based child benefit for parents to allocate as they see fit. You can read more about that proposal in our paper, Toward a Universal Child Benefit.