Aiming for Targets, Saving on Arrows: Recent Insights from Two USDA Food Assistance Programs

Efficiency involves converting the least amount of inputs into the greatest amount of outputs, which is important not only in farming but also in food assistance programs. In farming, physical inputs (land, labor, seeds, fertilizer, and others) are converted into output (crops and livestock). In USDA’s food assistance programs, taxpayer dollars are the inputs. The outputs are the programs’ goals: to provide needy persons with access to a more nutritious diet, to improve the eating habits of the Nation’s children, and to help America’s farmers by providing an outlet for the distribution of food purchased under farmer assistance authorities. Both farmers and USDA strive to operate efficiently.

In program analysis, the term “targeting” is often interchangeable with “efficiency.” In recent years, Congress and USDA have been particularly interested in operational targeting—focusing on how the Nation’s food assistance programs are administered—and benefits targeting—focusing on who is served. Each taxpayer dollar used to fund a program can be thought of as an arrow that policymakers send toward a policy target, or program goal. Metaphorically, operational targeting is the effort to shoot an arrow at a target at a low cost, while benefits targeting is the effort to hit the bull’s-eye—getting program benefits to the most needy.

Over the years, USDA has endeavored to operate food assistance programs efficiently. The Federal Government and the States continually seek to identify policies and procedures by which program participants can be served at a low cost or the needy can be more effectively targeted. USDA has recently initiated innovative targeting efforts in two of its child nutrition programs.

Program Design Has One Pair of Targeting Decisions . . .

In programs designed to serve recipients most in need, benefits may be targeted in two ways—through eligibility guidelines and through the schedule of benefits. Eligibility guidelines are the criteria households must meet to receive program benefits. Eligible households become program participants only if they choose to apply. Household income, adjusted for family size, is a major criterion for USDA food assistance programs. Age, nutritional risk, breastfeeding status, and workforce status are among other factors that can determine eligibility.

If program eligibility guidelines are broad instead of narrow, the numbers of households that qualify for and can participate in the program increase, which can support program goals (such as improved nutrition). However, as participation rises, so too do program expenditures. Policymakers pursue guidelines targeting by balancing the additional cost of broader eligibility guidelines with the gains in terms of program goals.

While guidelines targeting determines which households are eligible for a program, benefits targeting determines whether or not program participants all receive the same level of benefits. Benefits targeting links benefits to income in an effort to provide greater program benefits to households that have the lowest incomes. For example, Food Stamp Program benefits are highest for households with no income (net of certain allowed deductions). Benefits are reduced by 30 cents for each dollar of income. Similarly, the National School Lunch Program provides three different amounts of USDA subsidies for lunch, depending on the income of a child’s household. In contrast, a breastfeeding mother who participates in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) receives a fixed package of specific foods (such as carrots and tuna fish) regardless of her household’s income, so long as the income does not exceed WIC’s income-eligibility threshold. If a low benefit level reduces participation of higher income households, there is a tradeoff between encouraging participation of higher income households and targeting benefits.

. . . While Program Administration Has Another Pair

For a program to serve its intended recipients at low cost, two additional types of targeting may be used in the administration of food assistance programs. Local program offices try to exclude ineligible households from receiving approval, or being certified, when the household applies for benefits. Certification targeting—providing certification only to those households who are intended to be recipients of program benefits—requires local program offices to obtain household-specific information. Information such as income, household size, and other household characteristics is used to determine if a household is eligible. Certification typically lasts from 1 to 12 months, after which information is again required to determine whether participating households continue to meet eligibility guidelines.

A local program office can obtain a household’s information using various methods of increasing thoroughness, such as asking the household, requiring supporting documents (such as pay stubs), and using third-party verification (such as employers) to ensure the authenticity of the documents. An increase in the thoroughness of the application process can be expected to enhance certification targeting by reducing inaccuracies and increasing compliance with eligibility guidelines.

Denying program benefits to ineligible households helps maintain public confidence in USDA food assistance programs. However, increased thoroughness comes with a price: increased burden—on both ineligible and eligible households—and increased administrative expense. A high level of burden may deter some households from applying for benefits for which they are eligible. Policymakers must strike a balance between certification targeting, on the one hand, and both program accessibility and administrative expense on the other.

Operational targeting seeks to minimize administrative and food procurement expenses. At the extreme, the administrative cost of certifying households could be slashed by closing all but one of the local program offices in an entire State. Likewise, administrative expenses could be saved if nothing—not even an application—was required for a household to receive program benefits. Of course, eliminating the application would negate certification targeting. And widespread office closures would greatly inconvenience many eligible households and diminish their program access and participation, thereby countering the goals of the program. Thus, there can be tradeoffs between operational targeting and other desirable outcomes.

Operational Targeting and WIC Cost Containment

The mission of WIC is to safeguard and improve the health of low-income pregnant, breastfeeding, and postpartum women and infants and children up to age 5 who are at nutritional risk. To achieve its mission, the program provides a package of supplemental foods, nutritional education, and health care referrals.

WIC State agencies adopt various cost-containment practices to reduce food costs. The practices include:

  • Limiting food item selection according to brand, package size, form, or price (for instance, requiring purchase of least-cost items).
  • Limiting authorized food vendors to those with lower food prices.
  • Negotiating rebates with food manufacturers or suppliers.

Some observers have raised concerns that if cost-containment policies are overly restrictive, then WIC participants’ access to and consumption of prescribed foods may be reduced. Others have questioned whether cost-containment practices save enough in food costs to offset their additional administrative costs.

In 1998, Congress instructed ERS to assess the effects of WIC State agencies’ cost-containment practices (other than manufacturers’ rebates on infant formula) on such outcomes as program costs, participant satisfaction, and the purchase and consumption of prescribed WIC foods. The study was conducted in six States (California, Connecticut, North Carolina, Ohio, Oklahoma, and Texas) selected to represent various combinations of cost-containment practices.

The study found that cost-containment practices can be inexpensive to operate. In the four States with substantial food item restrictions, administrative costs for the cost-containment practices averaged less than 1.5 percent of estimated food package savings.

Annual estimated cost savings for a State depend on the State’s particular cost-containment practices and the size of its WIC caseload. California and Texas, two States with large WIC caseloads, had annual cost savings estimated at $40 million and $66 million, respectively, while Oklahoma had annual savings estimated at $6.7 million. Of the six States, Ohio had the smallest cost savings of $148,000, an outcome that is consistent with Ohio’s limited restrictions on the food items WIC participants can purchase.

What were the effects of cost-containment practices on WIC participants? Most surveyed WIC participants reported that they were satisfied with the available brands of food and package sizes approved for WIC by their State. There were exceptions, however. In Connecticut and Ohio, where purchases of cheese are restricted to the least expensive brand available in the store, WIC participants reported lower levels of satisfaction with allowed cheese brands than participants in the four other States. In Oklahoma, cereal purchases are restricted to store- and private-label brands, which reduced participant satisfaction with allowed brands in that State. Nevertheless, when overall satisfaction levels in States with restrictions are compared with levels in the nonrestrictive States, the differences are small and statistically insignificant. Moreover, according to survey responses, cost-containment practices did not diminish the amounts of monthly allotments that WIC participants purchased or consumed.

The single largest cost-containment strategy in WIC is its infant formula rebate program. Although WIC encourages mothers to breastfeed, a majority of participating infants receive infant formula through WIC. WIC State agencies typically use competitive bidding to award a contract to a single manufacturer of infant formula for the exclusive right to provide its product to WIC participants in the State. The contract-winning manufacturer is then billed for the amount of the rebates on the formula issued for WIC infants. In fiscal year 2001, infant formula manufacturers provided States with $1.5 billion in rebates, an amount that supports 28 percent of WIC participants. To support the same number of WIC participants in the absence of these rebates would require an equivalent increase in taxpayer expenditures.

Benefits Targeting in CACFP

The aim of the Child and Adult Care Food Program (CACFP) is to promote healthful meals in child and adult care settings. In the child care portion of CACFP, the program reimburses participating family child care homes and child care centers for meals and snacks. In the mid-1990s, Congress raised concerns about the types of families most often served by CACFP family child care homes. In 1995, only 21 percent of meal reimbursements to CACFP child care homes were for meals served to low-income children. A meal served in CACFP child care homes received the same reimbursement rate irrespective of the child’s family income.

To target program benefits more intensely on low-income children, Congress lowered the per meal subsidies, effective in mid-1997, on meals generally served to higher income children (see Tiering at a Glance). This tiering system represents a compromise between a single-rate system and a system that can create a potential barrier to participation by requiring determination of family income on a child-by-child basis. Before 1980, CACFP required family child care providers to document each family’s income. Care providers complained that the determination of family income was burdensome and too invasive for their relationship with the families whose children they served. Few family child care providers participated in CACFP prior to 1980, possibly due in part to this factor.

In 1996, Congress asked ERS to examine the effects of reduced meal reimbursements for CACFP family child care homes. The study found that, as intended, the subsidy reduction did concentrate benefits more intensely on low-income children, improving benefits targeting. The share of CACFP meal reimbursements to CACFP child care homes for meals served to low-income children more than doubled, from 21 percent in 1995 to 45 percent in 1999. Over the same period, CACFP child care homes served 80 percent more low-income children and 23 percent fewer higher income children. Between 1997 and 1999, following the subsidy reduction, the number of family child care homes reimbursed at the lower Tier 2 rate fell, while the number of family child care homes receiving the higher Tier 1 rate increased.

According to the study, in 1999, Tier 2 CACFP child care homes spent on average $91 per week on food—$19 less than Tier 1 homes. Also, despite this difference in food expenditures, the subsidy reduction apparently had little if any effect on the array of meals or snacks (breakfast, lunch, etc.) offered by a typical Tier 2 home.

To qualify for reimbursement, a CACFP meal must contain specified combinations of four meal components: milk; fruit, vegetables, and juice; bread (and bread alternatives); and meat (and meat alternatives). The study found that the subsidy reduction did not reduce compliance with meal component requirements. The study also compared the nutritional content of the foods served by Tier 2 homes in 1999 with the nutritional content of foods served by similar CACFP homes in 1995. In most respects, there were no significant differences. However, meals served in Tier 2 homes in 1999 contained more calories than meals served in 1995.

Lessons and Cautions

In striving to make efficient use of taxpayer dollars in the design and administration of USDA food assistance programs, policymakers pursue various types of targeting. WIC cost-containment practices implemented by six States were relatively inexpensive to administer and reduced food costs. Operational targeting was improved with few adverse impacts on WIC participants. The subsidy reduction in CACFP meal reimbursements targeted program benefits more intensely on low-income children, as intended. Benefits targeting was improved, with little if any effect on the components or nutritional content of meals served in the reduced-subsidy homes.

Caution should be exercised when using a study’s results to make inferences about possible effects of related policies. What would happen if WIC cost-containment practices in restrictive States were made yet more stringent? Or what would happen if CACFP meal reimbursements were made yet smaller for Tier 2 homes? It is possible that negative outcomes would be more severe than those reviewed here. Moreover, for cost-containment practices to work, they need to be managed well by State officials. The success of cost containment in the six study States was the result of ongoing efforts by the States to find those restrictions that both reduced food costs and were acceptable to participants. Therefore, even if a particular cost-containment practice improves operational targeting in one State, a different State may have a different experience.

Careful research can address issues surrounding the magnitudes of desired outcomes and adverse side effects. Sometimes a negative effect is sufficiently small that—once research obtains a measure of the effect—policymakers may decide it can be ignored. On the other hand, if negative consequences turn out to be large, the response may be to recalibrate policy if policymakers deem the benefits of such adjustment exceed the costs. Indeed, the States in the WIC cost-containment study engaged in a dynamic process of assessing cost savings and participant responses. Crafting food assistance policies is an ongoing process involving the affected groups, policymakers, and researchers who help to measure the sizes of the consequences at stake.

<a name='box'></a>Tiering at a Glance

For child care homes participating in the Child and Adult Care Food Program (CACFP), Congress replaced a single-rate reimbursement system with a two-tiered system that took effect July 1, 1997. Under the tiering system, the rates for Tier 1 meals, meant to be served generally to low-income children, were similar to the pre-existing rates, while the rates for Tier 2 meals, meant to be served generally to higher income children, were reduced. The Tier 1 and Tier 2 rates that took effect in mid-1997 were, respectively, $0.90 and $0.34 for breakfast; $1.65 and $1.00 for lunch/supper; and $0.49 and $0.13for snacks.

In fiscal year 1999, reimbursements to Tier 2 homes averaged $177 per month but would have averaged $326 per month if those homes had received Tier 1 rates for those same meals. Tiering lowered meal reimbursements to Tier 2 homes by 46 percent on average across meals, or by about $33 per week per home.

Congress established two main criteria by which a meal qualifies for Tier 1 reimbursement rates:

  • A CACFP home located in a low-income area qualifies for Tier 1 rates on all meals (an area is considered low-income if 50 percent or more of the children at the local elementary school have been approved for free or reduced-price school meals, or if 50 percent or more of the children in the area are in families with incomes at or below 185 percent of the Federal poverty guidelines as measured by the most recent decennial census); or
  • A CACFP home operated by a low-income care provider qualifies for Tier 1 rates on all meals.

In addition, a CACFP home that is classified as Tier 2 (it does not meet either of the above criteria) can receive the higher Tier 1 rates on meals served to low-income children.

The current reimbursement system is not designed to prevent totally the payment of a Tier 1 rate for a meal served to a higher income child. By the first two criteria above, a home in a low-income area or operated by a low-income provider receives Tier 1 rates on all meals, including those served to higher income children. Nevertheless, the tiering system has concentrated program benefits on children from low-income families relative to the single-rate system it replaced.