Study of Direct Certification
in the National School Lunch Program

EXECUTIVE SUMMARY

The NSLP offers free and reduced-price school meals to students from eligible households. Households with incomes at or below 130 percent of poverty are eligible for free meals, and households with incomes between 131 percent and 185 percent of poverty are eligible for reduced-price meals. Traditionally, to receive these benefits, households had to complete and submit application forms to schools or be directly certified.

Direct certification, on the other hand, is a method of eligibility determination that does not require families to complete school meal applications. Instead, school officials use documentation from the local or state welfare agency that indicates that a household participates in AFDC or food stamps as the basis for certifying students for free school meals.

Direct certification offers several potential benefits, including increasing the proportion of eligible students certified for free meals and the number of certified students participating in the NSLP and reducing burdens associated with distributing and processing free and reduced-price meal applications. Potential impediments to direct certification include a lack of cooperation and collaboration between NSLP and food stamp/AFDC agencies and problems with maintaining the confidentiality of food stamp/AFDC data.

This study was conducted to:

1. Provide descriptive information on the use of direct certification nationwide from a statewide and local perspective.

2. Estimate the costs and administrative savings of using direct certification.

3. Assess changes in free eligible certification and participation rates after implementation of direct certification, nationwide and within specific jurisdictions.

4. Identify factors, specifically implementation processes and systems and characteristics of jurisdictions (including populations) implementing direct certification, that contribute to successful direct certification efforts.

Information for the study was collected from seven data sources. They were: (1) a survey of all 51 NSLP state administrators; (2) a screening survey of 1,014 School Food Authorities (SFAs); (3) a survey of 148 SFAs that use direct certification; (4) a survey of 157 schools in directly certifying SFAs; (5) a survey of 30 AFDC/food stamp agencies involved in the direct certification process; (6) administrative data from FNS; and (7) demographic educational data from the National Center for Educational Statistics. Survey data represent the fall of 1996. Data collection activities began in November of 1996 and concluded in August of 1997.

Impact of Welfare Reform

The Personal Responsibility and Work Opportunity Reconciliation Act of 1996, Public Law 104-193 (the welfare reform law) eliminated the federal AFDC program and replaced it with the state-run Temporary Assistance to Needy Families (TANF) program. This has changed the direct certification processes in states where TANF eligibility standards are less restrictive than the old AFDC standards, because these states cannot directly certify TANF children. On the other hand, in states where the TANF standards are more restrictive than old AFDC standards, fewer children will be eligible for TANF than were eligible for AFDC. Thus, fewer children will be directly certified through TANF than were directly certified through AFDC.

Prevalence of Direct Certification

As of fall 1996, direct certification was used in 48 states and the District of Columbia and in 63 percent of all NSLP districts. These districts enrolled approximately 31 million students, which represented 72 percent of all students nationwide. These districts certified just over 10 million students for free meals through direct certification and conventional application-based certification during the fall of 1996. This represented 72 percent of all students certified for free meals nationwide.

Direct Certification Types

The in-depth survey of districts identified the three most common methods of direct certification in use during the fall of 1996. Two of these methods involved matching, which is the comparison of AFDC/food stamp lists or databases against student enrollment lists or databases to identify enrolled students who are AFDC/food stamp recipients. The three methods of direct certification that were identified are:

• non-matching method (Type I)

• district matching method (Type II)

• state matching method (Type III)

In the Type I model, matching did not occur. In this model, a state agency, typically the AFDC/food stamp agency, mailed documentation of AFDC/food stamp participation to AFDC and food stamp households with school-age children. These households could submit the documentation in lieu of applications to have their children directly certified for free meals. In Type I districts, districts and schools typically did not know which households received direct certification notices and it was up to the households to take steps to become directly certified. In contrast, in Type II and Type III models, schools did know which households were eligible and households did not need to take any steps to become certified. School districts performed the match for Type II districts and states performed the match for Type III districts. Then schools or school districts directly certified for free meals the food stamp/AFDC students identified through the matching process. Once the matching occurred, households were given the opportunity to reject certification.

Characteristics of Districts Using Direct Certification

As of 1996, approximately one-third of the direct certification districts used the Type I model, one-third used the Type II model, 19 percent used Type III, and about 16 percent used some other model. Most students in direct certification districts were enrolled in Type II districts, 41 percent, compared with 25 percent in Type I districts and 26 percent in Type III districts. Type II districts also certified more students for free meals than either Type I or Type III. Forty percent of students in direct certification districts who were certified for free meals were certified by Type II districts. Type I and Type III districts certified 28 percent and 25 percent respectively.

Direct Certification Processes and Systems

As of 1996, most states had used direct certification for four or more years. Districts tended to be relatively new (two years or less experience) to the direct certification process or to have been involved with it for four years or more.

Except in California, all states using direct certification had their welfare agencies generate lists of AFDC/food stamp households with school-age children. In the Type I method, these lists/databases were used to send AFDC/food stamp participation documentation to households. The Type II and Type III methods matched these lists/databases against student enrollment lists/databases to identify students receiving AFDC/food stamp benefits so they could be directly certified. More than 72 percent of states using direct certification used both AFDC and food stamp caseloads to generate these lists. Ninety-four percent of states using direct certification generated these lists /databases through a purely automated process.

Generating the AFDC/FS lists/databases at the state level usually took one to two months, beginning in early to mid-summer (May through July) and ending in late summer to early fall (late August to early September). States containing districts that conducted matching (Type II) took longest to generate these lists/databases, an average of five months or more.

Most states conducting the matching process took four or more months to complete the process, beginning in May and ending in August. Districts, on the other hand, did not begin the process until mid-summer (July) and only took approximately one to two months to complete, ending in August or September. Matching usually occurred only once per year regardless of whether it was conducted at the state or district level.

Matching at the state level was primarily an automated process. Seventy-seven percent of the states containing districts where matching was conducted at the state level indicated that the process was automated. Districts that conducted matching, on the other hand, were less likely to indicate that the matching process was automated. Only 9 percent indicated that the process at the district level was exclusively automated. This compares with 51 percent that indicated that the matching process was exclusively manual and 40 percent that indicated it was a combination of an automated and a manual process.

Notifications to families concerning their children’s pre-approval to receive free school lunches were usually completed within a two-month timeframe. The process usually began in August and was usually complete by September. In most cases, the notification was a letter sent to the appropriate household.

Assessing the Impact of Direct Certification on Free Certification and Participation Rates

The impact of direct certification on free certification and participation rates was analyzed by estimating a district-level model and a state-level model. The district-level model compared certification and participation levels in districts using direct certification with those not using direct certification during the fall of 1996. The state-level model compared certification and participation levels of states using direct certification with states not using direct certification from the fall of 1988 through the fall of 1996.

The district and state level models yielded disparate results. The district-level model found that direct certification had an insignificant effect on certification and participation levels. By analyzing direct certification over time, the state-level model was able to control for non-time varying differences (i.e., fixed effects) between states and certain time-varying characteristics. The state-level model was judged to be more accurate because of its increased ability to explain variance in the data. Additionally, the state-level model yielded results that were very intuitively reasonable, and its findings were robust across different specifications of the model.

Influence of Direct Certification on Non-Direct Certification Activities

Survey respondents indicated direct certification generally had only a slight effect on nondirect certification NSLP activities. No effect was noted at the state, district, or school level for the public notice process or for the development and dissemination of parental letters and applications. However, the free and reduced-price meal application process was affected at the school level and school district level. Direct certification reportedly reduced workloads and increased efficiency in receiving applications, in reviewing applications for completeness, in making application eligibility determinations, and in verifying the eligibility of a sample of applications.

Issues and Challenges Associated with Direct Certification Implementation

The greatest challenges to states in implementing direct certification were procedural issues such as how to conduct matches, how to notify directly certified households, and/or how to coordinate the various direct certification steps. Approximately 29 percent of the states provided such a response. Other noted concerns were confidentiality (25 percent), computer programming/formatting type issues (33 percent), and cooperation among the relevant direct certification actors (22 percent). Challenges were not as significant at the district or school level; however, such issues as procedural concerns (15 percent) and inadequate and/or incomplete information (10 percent) were noted.

Last modified: 05/22/2009