Evaluation of SSI/FSP Joint Processing Alternatives Demonstration: Final Report

EXECUTIVE SUMMARY

In 1995, the Food and Nutrition Service (FNS) of the U.S. Department of Agriculture (USDA) entered into an agreement with the South Carolina Department of Social Services (SCDSS) to implement a demonstration project to improve the delivery of food assistance to elderly and disabled Supplemental Security Income (SSI) recipients. The South Carolina Combined Application Project (SCCAP) was designed to test the effects of an alternative to the current Food Stamp Program (FSP) regulations governing the operation of an SSI/FSP joint processing system. In conjunction with this effort, the FNS contracted with Decision Information Resources, Inc. (DIR) to evaluate the alternative program including its effect on: FSP participation and benefits, timeliness and accuracy of application processing, administrative costs, and customer satisfaction. This report details the findings of a two-year evaluation of SCCAP, from October 1995 through October 1997.

Study Background

The Food Stamp Program, administered at the federal level by the FNS, provides monthly benefits to help low-income households buy food. To be eligible for food stamps, households must meet certain income and resource criteria. Households in which all members are receiving SSI or Temporary Assistance for Needy Families are categorically eligible for food stamps and are thus exempt from these income and asset eligibility tests.

According to current FSP regulations, SSI recipients are, by definition, automatically eligible for food stamp benefits. Despite this categorical eligibility, studies have shown that the SSI population (i.e., aged, blind or disabled) has not participated in the FSP to the fullest extent possible. To improve the coordination of food stamp services for SSI clients, the Food Stamp Act of 1977 required "joint processing" of SSI and FSP applications, whereby the SSA would provide "pure" SSI households the opportunity to apply for food stamps at the local SSA office. "Pure" SSI households were defined as those in which all members were applicants for or recipients of SSI.

While FSP regulations have allowed pure SSI households to apply for food stamp benefits at the SSA office for many years, joint processing has proved to be ineffective. Consequently, food stamp participation among this categorically eligible population has remained lower than expected. In 1992, the SCDSS reported that only 33,000 (42 percent) of the 78,000 SSI recipients who were categorically eligible to receive food stamp benefits were participating in the FSP (South Carolina SSI Standard Individualized Benefit Project, 1992). To address this issue, South Carolina initiated a two-year statewide demonstration project in October 1995. The South Carolina Combined Application Project (SCCAP), was designed to test the effectiveness of using a single application and information source to:

• Increase participation of SSI clients in the Food Stamp Program

• Limit administrative costs by minimizing duplication of intake and application procedures for two different federal programs

• Improve customer satisfaction with the services received

The SCCAP demonstration was designed to streamline the application process through increased automation and eliminate the need for local involvement or face-to-face interviews with Food Stamp Program staff, unless warranted by special circumstances. Data collected at the time of initial SSI application were to be used for both SCCAP eligibility determination and food stamp benefit calculation. Standard income amounts, shelter expenses, and benefit allotments were also used to simplify the application process.

SSI recipients who met the following criteria were automatically deemed eligible to participate in the demonstration:

• Designated as living arrangement code of "A" (which specifies SSI individuals who report home ownership or rental liability, pay a pro rata share of household expenses, and do not receive both food and shelter from others)

• Reported no earned income

• Declared that they purchase and prepare food separately from other members of the household

To ensure food stamp eligibility, only one-person SSI households were eligible to participate in SCCAP.

Individuals who had excess medical or shelter expenses could chose to apply for benefits through the regular food stamp application process but could not participate in the regular FSP and SCCAP simultaneously. If SSI was denied or terminated at any time during the demonstration, individuals would no longer be eligible for SCCAP but could still apply for or continue to receive benefits through the regular FSP.

Methodology

The evaluation of the SCCAP demonstration focuses on seven study objectives and related research questions. Specifically, the study seeks to address the following objectives:

• Objective 1: Describe the processes involved in implementing the demonstration.

• Objective 2: Assess the effect of the demonstration on FSP and SSI administration.

• Objective 3: Assess the effect of the demonstration on food stamp participation of SSI households.

• Objective 4: Assess the effect of the demonstration on the level of household benefits.

• Objective 5: Assess the effect of the demonstration on error rates.

• Objective 6: Quantify, to the extent possible, the administrative costs of the demonstration.

• Objective 7: Assess the effect of the demonstration on client satisfaction.

The SCCAP evaluation was primarily designed to explore the differences between demonstration-eligible SSI clients who participated in the FSP and those who did not participate in the FSP. Demonstration-eligible food stamp participants were further subdivided into two groups: (1) those who participated in the SCCAP demonstration, and (2) those who received food stamps through regular processing, not SCCAP, due to excess shelter or medical expenses.

In addition to stratifying the analysis on the basis of food stamp participation, client data are subdivided into groups based on when the individual applied for and received SSI and food stamps. Individuals are placed in one of three categories: (1) conversions – those who were already participating in SSI and the FSP at the time the demonstration began; (2) outreach cases – those who were participating in SSI but not the FSP in the early months of the demonstration; and (3) new applicants – those who applied for SSI during the demonstration period.

Information needed to address the study objectives and research questions was collected from several different data sources. In addition to using existing data files, additional data were collected on customer satisfaction, staff perceptions of the SCCAP demonstration, and quality control issues. Seven primary sources of data were used to evaluate the SCCAP demonstration:

• SCDSS Client History and Information Profile (CHIP) data files

• SSA’s State Data Exchange (SDX) data files

• Telephone surveys of SCCAP-eligible SSI clients

• Telephone surveys of SSA and SCDSS program managers, caseworkers, and claims representatives

• Site visit interviews with SSA and SCDSS program administrators and staff and on-site observations

• Supplemental Quality Control (QC) reviews conducted by the SCDSS

• National- and state-level aggregate data

SCCAP Implementation

Several steps were required to initially implement SCCAP. Early activities focused on four areas:

• Conversion of existing food stamp participants. SCCAP implementation began in September 1995 when the SCDSS notified current food stamp participants in one-person, SSI households about their pending conversion to SCCAP. Clients were told to contact their local food stamp worker immediately if they believed that they qualified for excess monthly shelter or medical expenses. If the client did not decline benefits or claim excess expenses by September 30, their case was automatically converted to the SCCAP caseload.

• Outreach to inform potentially eligible SSI clients about SCCAP. In addition to converting current SSI/FSP clients to SCCAP, the SCDSS conducted an outreach campaign to attract demonstration-eligible SSI recipients who were not receiving food stamp benefits. Approximately 42,817 outreach brochures and application forms were distributed to SSI recipients. Other outreach activities conducted included news conferences and mass mailings to community agencies and action groups.

• Training of SSA and SCDSS staff. To successfully implement the demonstration, managers and caseworkers had to become knowledgeable about SCCAP and its requirements. Each agency designed and conducted staff training on SCCAP. SSA staff were trained to describe SCCAP and food stamp benefits to SSI applicants, determine the client’s interest in applying for food stamps, and perform slightly modified computer entry procedures to incorporate the SCCAP application process within the SSI application process. SCDSS training included the conversion process, SSA's application process, and methods to determine excess expenses.

• Modification of computer screens and forms. To automate the SCCAP application process, modification of existing computer programs at SSA and the SCDSS were required. Proposed modifications included: (1) new data fields to indicate SCCAP status; (2) incorporation of the SCCAP eligibility and election statements into SSA data entry screens; (3) built-in edit features to disallow income amounts outside of a set range and automatically calculate the SCCAP benefit allotments; and (4) computer-generated notices.

Major Findings

South Carolina has been successful in meeting the objectives set forth by the SCCAP demonstration. The SCDSS, in collaboration with the SSA and the FNS, has implemented a joint processing alternative that has had positive benefits for both the clients and the agency. The major findings of the SCCAP evaluation are summarized below. Results are presented in relation to each of the three primary objectives of SCCAP.

Increase FSP participation of SSI households

• Estimates based on national data suggest that the rate of food stamp participation among SSI recipients in South Carolina increased from 38 percent in 1994 to 50 percent in 1998 while the national rate decreased from 42 percent to 38 percent during the same period.

• SCCAP outreach efforts resulted in over 8,500 new food stamp cases.

• Each year, approximately 840 new SSI recipients take advantage of the streamlined SCCAP application process and enroll in the Food Stamp Program.

Limit administrative costs by minimizing duplication of intake and application procedures

• Initial start-up costs were estimated to be less than $200,000.

• The SCDSS estimates it has been able to reallocate the equivalent of 40 full-time caseworkers (at least $700,000 in labor costs) by centralizing the SCCAP caseload at the state office in Columbia.

• Ongoing administrative costs at the SCDSS are estimated at less than $125,000 per year. The added burden at the SSA is a mere $2,360 annually.

• Net potential savings at the SCDSS are estimated at $575,000 per year.

Improve client satisfaction with the services received

• Almost 80 percent of new SSI applicants report that the food stamp application process at the SSA was "easy" or "neither easy nor hard."

• Overall, the majority of new SSI applicants were either "somewhat satisfied" or "very satisfied" with: (1) the option of applying for SSI and food stamps in the same place, (2) the amount of time SSA staff took to explain the FSP, (3) the accuracy of the information provided by SSA about the FSP, and (4) the ease of completing the food stamp application process at the SSA.

• Clients who applied for food stamps at the SCDSS office also reported being satisfied with (1) the amount of time staff took to explain the FSP, and (2) the delay between completing the application and being notified about eligibility.

• Demonstration participants reported higher levels of satisfaction with some aspects of the demonstration than those who chose to go through regular processing to claim excess expenses.

Challenges to SCCAP Implementation

While the SCCAP demonstration has met with some success, the SCDSS faced several challenges in implementing this alternative to joint processing. While some improvements still must be made, program staff have managed to successfully overcome most of the barriers encountered. Based on their experiences, SCDSS and SSA staff offer the following "lessons learned":

• Systems support is a must. South Carolina has experienced several problems and delays due to programming difficulties. From the SSA perspective, the inability to modify the SSA data system has limited the ability to automate the food stamp application process. The SCDSS is also forced to manually enter data that could be automatically transferred if the data system were appropriately programmed. While some computer modifications have streamlined the application process to some degree, further programming is needed to realize the full potential of SCCAP as it was originally envisioned. Although SCCAP is not yet fully automated in South Carolina, program staff report that the current system is still a vast improvement over regular food stamp application processing.

• The use of standardized shelter expenses can result in decreased benefits for some households and increased benefits for others. Based on supplemental QC data, the demonstration resulted in a 17 percent reduction in total benefits paid (a monthly average of $4.47 less per case). Depending on actual expenses, the effect of the SCCAP benefit calculation formula at the individual case level varied: 63 percent received higher benefits under SCCAP compared to what they would have received through regular FSP processing, 36 percent received lower benefits under SCCAP compared to the FSP, and 1 percent received the same amount under SCCAP that they would have received under the FSP. Since federal statute prohibits the use of a standard that increases deductions for households with no or low expenses relative to income, the use of standardized shelter expenses may make future replication of this streamlined application model questionable. Federal regulations also prohibit the loss of benefits as a result of any demonstration project; therefore, the streamlined nature of SCCAP processing makes future replication of this model questionable.

• Avoid the need to restore lost benefits. SCDSS staff report that the restoration of lost benefits and transfer of some cases back to the regular FSP caseload was extremely time consuming and labor intensive. The decrease in benefits suffered by many households created great confusion among both clients and staff. It is important that program options (e.g., claiming excess expenses) and the formula used to calculate benefits (particularly the use of standard amounts in place of actual expenses) be clear to all staff to avoid any misunderstandings that may result in lost benefits.

• Train front-line staff adequately. SCDSS staff report that the training of their caseworkers did not adequately prepare them to answer the questions raised by clients affected by SCCAP. Program staff suggest that sufficient training (5-7 hours) be provided to all front-line staff before the demonstration is fully operational. This will avoid some of the confusion that is likely to occur when clients are converted to a new program or face new application procedures.

• Allow for adequate staffing to ensure that applications are processed in a timely fashion. Because federal statute requires that food stamp applications be processed within a limited time, it is important to have enough staff available to handle the large influx of applications that can result from outreach efforts. SCCAP outreach efforts were delayed in part because of inadequate staffing at the central office and temporary help was hired to clear the backlog of outreach applications waiting to be processed. To avoid this problem, SCDSS staff suggest that sufficient personnel be hired before a major outreach effort is conducted.

• Certain aspects of the electronic benefit transfer system run counter to the purpose of joint processing. Joint processing is intended to eliminate the need for clients to visit both the SSA and food stamp offices. In South Carolina, the change to electronic benefit transfer (EBT) technology (from mail issuance of benefits) meant that SCCAP participants had to go to their local SCDSS office to pick up their EBT card and be trained on how to use it. Program staff report that not only is this trip difficult for many of their elderly and disabled clients, but many SCCAP participants do not even understand that they must visit the local office before they can access their benefits.

In October 1997, the South Carolina Department of Social Services was granted an extension of the SCCAP demonstration. Based on the success demonstrated to date, the FNS agreed to extend SCCAP for a maximum of three additional years (through Sept 30, 2000). During this time, Congress will have a chance to review the findings of this evaluation and determine whether the results warrant amending the Food Stamp Act so that South Carolina may continue to use the special provisions of SCCAP as part of its normal FSP operations.

While the problems faced by different agencies in different states will vary, a lot can be learned from the SCCAP implementation. Although some areas of operation warrant further investigation, the SCCAP evaluation findings indicate that this alternative approach to joint processing is worthy of replication on a larger scale.

January 2000

Last modified: 02/17/2012