Improving Direct Certification Will Help More Low-Income Children Receive School Meals

July 25, 2014
BY

Madeleine Levin (Food Research and Action Center) and Zoë Neuberger (Center on Budget and Policy Priorities)

The National School Lunch and School Breakfast Programs and the Supplemental Nutrition Assistance Program (SNAP, formerly the Food Stamp Program) are powerful tools that alleviate child hunger and improve child well-being.  To make sure that the most vulnerable children have access to much-needed school meals and to make these programs more efficient, Congress has established automatic eligibility for free school meals for certain categories of children whose families are most likely to struggle against hunger.  Moreover, Congress has set high expectations that states will reach the vast majority — if not all — of these children with automatic eligibility through a process called “direct certification.”

Through direct certification, school districts that participate in the National School Lunch Program, as almost all do, match the names of children living in households that receive SNAP, Temporary Assistance for Needy Families cash assistance (TANF), or Food Distribution Program on Indian Reservations (FDPIR) benefits with school enrollment records. This match is then used to certify students for free school meals without the need for their families to complete a school meals application.  School districts are required to directly certify students living in households receiving SNAP benefits and are encouraged to so it for children receiving TANF or FDPIR benefits.

Direct certification benefits students, parents, and school districts.  Millions of eligible low-income children receive free breakfast and lunch through direct certification.  School districts process and verify fewer school meal applications, which allows them to benefit from administrative savings and improves certification accuracy.  The administrative savings give school districts more resources to focus on improving meal quality and service.  Moreover, strong direct certification results in easier implementation of the Community Eligibility Provision, an important new option for high-poverty schools to offer free breakfast and lunch to all students that relies heavily on direct certification for its success.

Direct certification became an option for schools in 1986, and state child nutrition agencies and school districts across the country began developing direct certification systems.  Since then, Congress has taken numerous steps to strengthen and improve direct certification.  In 2004, it instituted a requirement that all school districts conduct direct certification for children living in households receiving SNAP benefits, and it provided funding to states to develop or improve their direct certification systems.  Congress made further improvements through the Healthy, Hunger-Free Kids Act of 2010 (the most recent child nutrition reauthorization), which set performance benchmarks for states, requiring them to directly certify at least 95 percent of children living in households receiving SNAP benefits by the 2013-2014 school year, and provided performance awards to states for improving direct certification and having strong direct certification systems.

States have made significant progress in improving their direct certification rates, yet they can do more to ensure that they meet the federal standard and enroll for free school meals all low-income children eligible for automatic enrollment.  According to data from the U.S. Department of Agriculture (USDA), 28 states fell short of Congress’ benchmark to directly certify at least 90 percent of students living in SNAP households for the 2012-2013 school year.[1]  Twelve states did not even reach 80 percent, missing more than one in five eligible children.

Many state agencies and school districts need to intensify their efforts to improve their direct certification systems.  One key step is for them to adopt the best practices that successful states have implemented in order to reach the benchmarks set by Congress.  By meeting those benchmarks, states and school districts will help individual low-income children and simplify program administration, as well as facilitate broader adoption of community eligibility and increase federal reimbursement to schools using that provision.

How School Districts and States Directly Certify Children for Free School Meals

Children who live in households that receive SNAP, TANF, or FDPIR benefits, as well as children who are migrant, homeless, in foster care, or enrolled in Head Start are considered “categorically eligible” for free school meals and can be certified without submitting any application.

School districts are required to directly certify — through data matching, a minimum of three times per year — children who live in households participating in SNAP.  But not all school districts have adopted effective direct certification systems, leading to uneven enrollment for free school meals, added application burdens for schools and families, and eligible children missing the free, nutritious meals they need.

States and school districts also can work with additional programs to directly certify other groups of “categorically eligible” children for free meals.  For programs with state or county-wide enrollment databases, including TANF cash assistance, FDPIR, and (often) foster care, children can be certified through data matching.  Other groups of categorically eligible children, such as migrant and homeless children, for which central databases are not common, can be certified based on a list provided to school nutrition staff by an appropriate official.

States use one of two data matching approaches to directly certify eligible children:  state-level matching, where the state child nutrition agency matches SNAP, TANF, FDPIR, and/or foster care program data with school enrollment records and distributes the results to districts; or district-level matching, where the state agency distributes program data to the district to match with its own enrollment records.  The choice of system is determined by state-specific factors, such as the size of school districts.  For example, large, county-wide districts lend themselves better to district-level matching because larger districts have the capacity and resources to conduct matching that smaller districts may lack, and because SNAP is generally administered by county agencies and would therefore align with school district boundaries.

Interaction between Direct Certification and Community Eligibility

The new Community Eligibility Provision (CEP) allows high-poverty schools to eliminate school meal applications and offer free breakfast and lunch to all students.  Instead of collecting individual applications to determine eligibility and federal school meal reimbursement rates, the percentage of “identified students” — those who are certified for free school meals without submitting an  application, primarily through direct certification — is multiplied by 1.6 to determine the percentage of meals that the federal government will reimburse at the (highest) “free” meal rate.  For example, a school with 50 percent identified students would receive 80 percent of its lunch and breakfast meals reimbursed at the free rate and 20 percent at the “paid” rate.  To be eligible for CEP, a school, group of schools within a district, or a school district must have at least 40 percent identified students out of their total student enrollment.  High-poverty school districts with low direct certification rates will find that their identified student percentage does not accurately reflect the level of poverty within the student population.  By improving their direct certification, these school districts may qualify for CEP or may find it more financially viable to choose CEP, making their meals programs less complicated and eliminating barriers to participation for numerous low-income, hungry children.

Direct Certification Policy and Performance Standards

Congress and USDA have made a number of policy changes aimed at strengthening the direct certification process over the past five years, including:

  • mandating electronic data matching using SNAP records;
  • requiring a minimum of three matches using SNAP records each year, with more frequent matching encouraged;
  • extending direct certification to all children who live in a household receiving SNAP, FDPIR, or TANF; and
  • requiring USDA to issue an annual report analyzing state performance and highlighting best practices.

As mentioned earlier, Congress and USDA also have established performance benchmarks and provided resources to help states reach them and reward high-performing states.  The Healthy, Hunger-Free Kids Act of 2010 established the benchmarks, which have been in place since the 2011-2012 school year, to ensure that school districts automatically enroll low-income children for free school meals.  Beginning with the 2013-2014 school year, states were required to directly certify 95 percent of the school-age children in households receiving SNAP benefits.  States that do not meet the direct certification performance standards are required to develop a Continuous Improvement Plan (CIP) identifying action steps, a timeline for implementing them, and measures to assess progress.  States’ performance meeting the benchmarks can be found in USDA’s annual report to Congress on direct certification.[2]

High-performing states and those that made substantial improvements in their direct certification performance have received bonus awards.  For performance during the 2011-2012 school year, Alaska, Delaware, Nebraska, North Dakota, West Virginia, and Wyoming received Outstanding Performance Awards; Georgia, Idaho, Kentucky, Maryland, Michigan, New Jersey, Ohio, and Utah received Substantial Improvement Awards.  For performance during the 2012-2013 school year, the District of Columbia, Kansas, Kentucky, Michigan, Tennessee, Texas, Vermont, and Virginia received Outstanding Performance Awards; Arizona, Massachusetts, Ohio, South Dakota, and Utah received Substantial Improvement Awards.

Direct Certification Performance

USDA’s annual report on direct certification performance sets out the share of school-age children in households receiving SNAP benefits who are directly certified.  The report shows that efforts to improve direct certification are making a difference.  When state performance was first measured, for the 2007-2008 school year, only 68 percent of children in households receiving SNAP benefits were directly certified nationwide.  By school year 2012-2013, that figure had grown to 89 percent of school-age children in households receiving SNAP benefits, according to the USDA data. Still, the national average is below the congressionally set performance standard of 90 percent for that year (as are just over half the states).

In the 2012-2013 school year (the most recent year of data available), the number of children directly certified based on SNAP data increased by 6 percent from the previous year, reaching about 740,000 additional students.  This increase significantly outpaced the increase in school-age SNAP participants — of about 221,000 — during the same time period.  This represents substantial progress and there is much to learn from the states and regions that have demonstrated success.

State performance in 2012-2013 varied widely, with the top ten states (Alaska, District of Columbia, Kansas, Kentucky, Michigan, Nebraska, New York, North Dakota, West Virginia, and Wyoming) directly certifying all eligible children, while the bottom 12 states certified less than 80 percent.  (See Figure 1.)  The bottom 12 states (Arizona, California, Colorado, Louisiana, Maine, Mississippi, Montana, New Hampshire, New Mexico, Oregon, Pennsylvania, and Washington) include a significant share of the SNAP households in the country.  According to the USDA report, “[T]he wide gap between States near the bottom of the chart and those near the top makes clear that some States’ direct certification systems are simply less effective than other States’ systems.”[3]

The USDA report found that nationally 91 percent of school districts complied with the requirement to conduct direct certification, at least to some degree, but those that did not comply served only 1 percent of students participating in the National School Lunch Program.  Some states had very low school district compliance rates; in four states (Alaska, Maryland, Minnesota, and Nevada) fewer than 70 percent of districts complied with the data matching requirement.  The school districts that are not conducting direct certification are mostly very small districts, charter, or private schools and may require additional technical assistance.  This gap points to an area in which to focus future efforts to improve direct certification rates, and is included in action steps explained later in this report.

7-25-14fa-f1.png

Significantly, five of the ten highest-performing states had in one of the previous two school years implemented community eligibility, a new option for high-poverty schools to offer free breakfast and lunch to all students based on direct certification rates (along with the percentage of children who are homeless, migrant, in Head Start, or in foster care).  Moreover, the seven states that had community eligibility in place for the 2012-2013 school year increased their direct certification performance by 28 percent between the 2010-2011 school year and the 2012-2013 school year, compared with a 16 percent increase for all states over the same period.  These findings show the important relationship between community eligibility and strong direct certification systems.  The other states with strong direct certification systems are well positioned to maximize the number of high-poverty schools that will be able to adopt the universal meal provision now that it is available nationwide for the 2014-2015 school year.

The report demonstrates how concerted efforts by USDA and state agencies to improve direct certification data matching can pay off:  when they invest time and funds, performance improves.  Since January 2011, USDA has made $17 million in grants to 24 states to improve their direct certification systems.  These improvements have included increasing match frequency, adding additional data sources (such as foster care data), incorporating non-public schools and charter schools in the matching process, and enhancing matching procedures.  These changes are important:  half of the 14 states that had double-digit improvements in the 2012-2013 school year had received an improvement grant in the previous year.

For many states, there is still room for improvement.  More than half of the states (27) did not meet the performance benchmark that Congress set for the 2012-2013 school year to certify at least 90 percent of students living in SNAP households.  In addition, 12 states directly certified less than 80 percent of such students.  Tremendous variation exists, too, in direct certification rates in different regions of the country, with only two of 13 states in USDA’s West and Southwest regions meeting the 90 percent benchmark.

Call to Action: Steps to Improve Direct Certification

Every state can meet the national performance benchmarks.  Multiple models of successful direct certification systems exist, and federal funds are still available to help states improve systems.  The strategies outlined below can help states and school districts improve direct certification rates and meet Congress’ performance standards.

Raise the visibility of the issue.  In states with low direct certification rates, advocates can elevate the issue and engage the media, state elected officials, the state agency, education advocates and associations, health groups, businesses, and state school nutrition associations.

Refine the data matching process.  Direct certification can be enhanced by regularly reassessing whether the data elements and process used for data matching are successfully identifying children eligible for direct certification.  States have implemented several approaches that increase the share of children eligible for direct certification who are matched.  For example, some states use software that allows for variations in names and spelling when conducting matches.  Some states conduct multiple data matches using different data elements in each match.  Probabilistic matching, which compares multiple data fields from two sources and determines a statistical probability of a match, offers another promising approach.  States and districts may need to try various alternatives and reassess their matching criteria over time to find the right set of identifiers so that direct certification is neither over- nor under-inclusive.

Use any available data to reach all children in the household.  If one person in the household is identified as participating in SNAP, then the school district must, to the extent possible, directly certify for free school meals all children in the household even if each child has not been individually matched.[4]  States or school districts can take steps to identify additional children in the household.  For example, once the student database has been matched against the SNAP database to identify individually matched children, the student database can be searched to identify additional children in the households of individually matched children.[5]  States can improve the identification of additional children in the household by ensuring that records are available to link students by household.  In lieu of additional matching, the school district may rely on other available data, such as information from school enrollment forms or “point-of-sale” software.[6]

Conduct matches as often as possible and develop the capacity to look up individual children.  States or districts are required to conduct matches with SNAP data at least three times each school year to identify children who are eligible for free school meals.[7]  Conducting matches more frequently or developing the capacity to look up whether an individual child can be directly certified helps prevent schools from missing children who become eligible for SNAP after the start of a school year, or who change school districts during the year.  In each month of 2013, an average of more than 500,000 children were in families that newly enrolled for SNAP benefits.[8]  Total SNAP participation is likely to decline over the next few years, but substantial numbers of new children will still enroll each month.[9]  Frequent direct certification matching helps ensure that these children begin receiving free school meals promptly if they were not already certified and increases the school’s Identified Student Percentage under the Community Eligibility Provision.

Regularly provide training and support for local staff.  Additional training efforts represent a promising opportunity for improving direct certification.  In most districts, handling direct certification is only a small part of a local school administrator’s job.  Even the best data matching system will not be effective if staff do not know how to use it.  Some steps that have proven successful include adopting a simplified interface for the matching system, ensuring that the process for uploading enrollment data is flexible and can work with all local databases, creating a strong instruction manual, and providing training and technical assistance to staff on an ongoing basis.[10]  It is important to include charter and private schools in training because they may be less familiar with state data systems.

Adapt systems to facilitate charter and private school participation in direct certification data matching.  Direct certification is challenging for private and charter schools because of their limited administrative capacity and lack of defined enrollment areas.  Sometimes charter schools are established as part of a Local Educational Agency (LEA) and other times they are their own separate LEA, responsible for all administrative systems.  The most recent USDA direct certification report provides examples of strategies that states have implemented to include this hard-to-reach group in their direct certification systems, including pairing private schools with nearby public schools, using individual look-up systems for these schools, providing training specifically for private and charter school staff, and building capacity for private schools to upload their student lists into statewide data matching systems.[11]

Strengthen systems for directly certifying vulnerable children who are not in households receiving SNAP benefits.  Although federal law requires school districts to establish data matching systems only for children in households receiving SNAP benefits, districts have the opportunity to directly certify other children, including those receiving TANF or FDPIR benefits.  USDA reports that schools certify through an application process nearly 1.7 million of the categorically eligible children, which needlessly burdens families with paperwork, creates unnecessary work for school districts, and introduces greater potential for error.[12]  Additional (uncounted) children who could be directly certified never get through the application process, missing out on the school meals programs.  Children facing unique hardships — those in foster care or Head Start as well as children who are homeless, migrant, or runaway — can be directly certified if an appropriate official documents their status.  Such children can be identified through a data matching process or based on a list signed by a caseworker, homeless liaison, or other appropriate official.

Continuously assess progress.  USDA’s performance report is a critical starting point for assessing state progress, but it is not the final word.  States with successful direct certification systems regularly self-assess the strength of their systems and identify opportunities for improvement.

Michigan’s Direct Certification System:  A Success Story

Michigan was one of the states to adopt community eligibility in the 2011-2012 school year, the first year the option was offered.  By that time, the state already had been improving its direct certification process.  Four state agencies came together to implement a centralized state matching system in 2006 — the Michigan Center for Educational Performance and Information, the Department of Education, the Department of Human Services, and the Department of Technology Management and Budget.  That same year, the state received a USDA grant to expand its direct certification system to include non-public schools and smaller LEAs, as well as to conduct regional trainings on how to use the system.  After implementing community eligibility, they embarked on further improvements, like increasing the frequency of matches, which are now conducted biweekly in August and September and monthly for the remainder of the school year, and adopting probabilistic matching.  For school year 2012-2013, Michigan incorporated TANF and foster care data for the first time and added the capacity for districts to look up individual children to determine if they can be matched.  These efforts have resulted in striking performance improvements, culminating in a $300,000 Outstanding Improvement Award from USDA in 2013.  For the 2010-2011 school year, Michigan directly certified 72 percent of children in households receiving SNAP benefits who were eligible for direct certification.  In 2011-2012, the share rose to 83 percent.  By the 2012-2013 school year, Michigan directly certified 100 percent of such children.

Resources to Support Direct Certification Improvements

In conjunction with strengthening performance standards and requiring states that do not meet the performance benchmarks to implement improvement plans, Congress and USDA provide states with substantial support to improve direct certification data matching.

Funding

In October 2009, Congress provided $22 million in federal funds for USDA to distribute in grants to state child nutrition agencies to improve their direct certification processes.[13]  USDA has distributed approximately $17 million to 24 state agencies.[14]  Some of the remaining funding will be distributed to states that applied for funds before the July 4, 2014 deadline, and there likely will be another round of grants.[15]  Grant funds may be used to implement new or revised direct certification systems, make technology improvements, or provide technical assistance to LEAs.  These funds also may be used to implement direct certification using Medicaid data in states that are approved to participate in USDA’s Medicaid direct certification demonstration projects (California, Florida, Illinois, Kentucky, Massachusetts, New York, and Pennsylvania).  USDA works closely with state agencies during the course of their grant period.
States have undertaken the following types of activities with grant funds:[16]

  • conducting direct certification matching more frequently;
  • adding software that checks names for transposed letters, common spelling variations, or names that sound alike;
  • incorporating unique student identification numbers into the matching process to facilitate future matches;
  • centralizing direct certification systems for greater efficiency;
  • adding features for probabilistic matching;
  • simplifying processes and creating tools for accessing match information online and downloading it directly to point of sale systems in local school districts;
  • developing the capacity to directly certify children who are not initially matched, including mechanisms to look up individual students and matches to identify additional children in the household;
  • providing training and support for LEAs, including private and charter schools, to improve the data they enter in student databases and ensure they are able to use the direct certification matching system; and
  • facilitating direct certification when students transfer from one LEA to another.

Technical Assistance

Through a contract with consulting firm Booz Allen Hamilton, USDA provides detailed, state-specific technical assistance to states that need help identifying ways to improve their performance.  The assistance can focus on improving the data matching algorithm, the most cost-effective hardware and software improvements, or how to support LEAs in successfully using the direct certification system.

Technical assistance has already been provided to 35 states, and planning is underway to provide assistance to additional states in the coming year.  States wishing to request technical assistance may do so by contacting the USDA Food and Nutrition Service Regional Office for their area.

Newsletters

USDA publishes a quarterly newsletter to state agencies, called Match to Meal, to highlight successful and promising practices in direct certification.  Stories and tips are collected from state technical assistance site visits and shared with all states.

Continuous Improvement Plan Guide

As noted above, states that do not meet the direct certification performance standards are required to develop a Continuous Improvement Plan (CIP).  A CIP must include the specific steps the state plans to take to improve direct certification results, a timeline for implementing them, and performance measures that will be used to assess progress.  To assist states in developing their CIPs, USDA has issued a comprehensive guide that includes a self-assessment tool.[17]  The guide takes state child nutrition staff though the steps needed to identify the goals, objectives, and performance measures for a CIP.  It also includes a prototype CIP and implementation timeline example.  The self-assessment tool is intended to help any state identify areas where they could improve, even if they are not required to develop a formal CIP.  USDA encourages its use as a tool for all states and welcomes feedback on ways to make it more useful.

Presentations and Video Resources

USDA maintains a directory of guidance materials, slide presentations, and recorded webinars on direct certification topics in its Child Nutrition Programs PartnerWeb, an online community for sharing information with child nutrition state agencies.  State agency staff who do not have access or need assistance locating materials in this shared community may request help by contacting the USDA Food and Nutrition Service Regional Office for their area.

Conclusion

Direct certification ensures that vulnerable children at risk of hunger can count on getting free breakfasts and lunches at school, and, as direct certification systems improve, millions of low-income students across the country will benefit from improved access to school meals.  At the same time, school districts will benefit even more from the simplified program administration and improved program integrity.

With the start of a new school year, now is the time to re-double efforts to improve direct certification systems.  School districts are conducting their first required direct certification, and high-poverty schools across the country are preparing to offer free meals to all students through the Community Eligibility Provision.  Strengthened and improved systems provide a key to access to free school meals to low-income students across the country.

End notes:

[1] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2012-2013,” U.S. Department of Agriculture, November 2013, p. 13, http://www.fns.usda.gov/sites/default/files/NSLPDirectCertification2013.pdf.

[2] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2012-2013,” p. 12, http://www.fns.usda.gov/sites/default/files/NSLPDirectCertification2013.pdf.

[3] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2012-2013,” p. 13, http://www.fns.usda.gov/sites/default/files/NSLPDirectCertification2013.pdf.

[4] See 7 C.F.R. § 245.6(b)(7) and Food and Nutrition Service Memorandum, “Extending Categorical Eligibility to Additional Children in a Household,” USDA, August 27, 2009, http://www.fns.usda.gov/sites/default/files/SP_38-2009_os.pdf.

[5] Some student databases have a specific household, or head of household, indicator.  For student databases that do not have such an indicator, the child’s address could be used to identify additional children in the household if the address is unique to a single household (for example, an apartment building would need to include unit numbers).  See Food and Nutrition Service Memorandum, “Questions and Answers on Extending Categorical Eligibility to Additional Children in a Household,” USDA, May 3, 2010, question 13, http://www.fns.usda.gov/sites/default/files/SP_25_CACFP_11_SFSP_10-2010_os.pdf

[6] See Food and Nutrition Service Memorandum, “Questions and Answers on Extending Categorical Eligibility to Additional Children in a Household,” USDA, May 3, 2010, question 10, http://www.fns.usda.gov/sites/default/files/SP_25_CACFP_11_SFSP_10-2010_os.pdf.

[7] See 7 C.F.R. § 245.6(b)(3).

[8] Center on Budget and Policy Priorities estimate based on SNAP administrative data, research on SNAP entry rates, and 2012 Household Characteristics data from USDA.  Even though an average of more than 500,000 children were in families that newly enrolled for SNAP benefits, SNAP caseloads did not increase by that amount because families also left SNAP.

[9] Dottie Rosenbaum and Brynne Keith-Jennings, “SNAP Costs Falling, Expected to Fall Further,”
Center on Budget and Policy Priorities, updated May 28, 2014, http://www.cbpp.org/cms/?fa=view&id=4054.

[10] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2010-2011,” U.S. Department of Agriculture, October 2011, Section V, http://www.fns.usda.gov/sites/default/files/DirectCert2011.pdf.

[11] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2012-2013,” pp. 30-31, http://www.fns.usda.gov/sites/default/files/NSLPDirectCertification2013.pdf,.

[12] “Direct Certification in the National School Lunch Program: State Implementation Progress, School Year 2012-2013,” Table 3, http://www.fns.usda.gov/sites/default/files/NSLPDirectCertification2013.pdf.

[13] Section 749(h) of the Agriculture, Rural Development, Food and Drug Administration, and Related Agencies Appropriations Act of 2010 (P.L. 111-80).

[14] Arkansas, Colorado, Connecticut, District of Columbia, Florida, Idaho, Iowa, Kansas, Massachusetts, Missouri, Montana, New Hampshire, New Jersey, New Mexico, North Carolina, North Dakota, Oregon, Pennsylvania, Rhode Island, South Dakota, Utah, Washington, West Virginia, and Wisconsin.

[15] “National School Lunch Program Fiscal Year 2013-2014 Request for Applications for Direct Certification Improvement Grants,” U.S. Department of Agriculture, http://www.fns.usda.gov/sites/default/files/2013directcert_rfa_2.pdf.

[16] For more detailed descriptions of state activities funded by Direct Certification Improvement Grants, see http://www.fns.usda.gov/school-meals/2013-direct-certification-improvement-grants and http://www.fns.usda.gov/school-meals/fy-2010-2012-direct-certification-grant-summaries.

[17] USDA’s CIP Guide is available at http://frac.org/fns_cip_development_guide.docx.