Overview
DTA did not deactivate or remove from the EPPIC system locations that were no longer in use. For instance, when the North Adams TAO merged with the Pittsfield TAO in 2010, the North Adams TAO was not removed from the list of location options.
The table below shows that 35,543 EBT cards issued to eligible households were assigned to inactive TAOs, out of the total population of 1,062,750 EBT cards that were issued during the audit period.
| Location | EBT Card Issued |
|---|---|
| Dudley Square* | 29,522 |
| Springfield Center | 5,346 |
| Newmarket Square | 571 |
| North Adams | 90 |
| Revere Supplemental Security Income | 8 |
| Inactive Springfield | 3 |
| Milford | 1 |
| Plymouth | 1 |
| Malden Supplemental Security Income | 1 |
| Total | 35,543 |
* The name of Dudley Square changed to Nubian Square in 2019.
Furthermore, while DTA issued a total of 1,062,750 EBT cards, 97,654 (9.2%) EBT cards were issued from non-TAO locations. These locations were identified as Centralized Eligibility, No Association, Unassigned, Centralized TAFDC, or SNAP only. See the table below for EBT cards issued during the audit period from non-TAOs.
| Non-TAO Location | EBT Card Issued |
|---|---|
| Centralized Eligibility | 84,258 |
| SNAP Only | 11,809 |
| No Association | 1,576 |
| Unassigned | 6 |
| Centralized TAFDC | 5 |
| Total | 97,654 |
If DTA does not ensure that all EBT cards are recorded as being issued from the correct TAOs and does not ensure that inactive TAOs are deactivated in the EPPIC system, then staff members could choose the wrong locations when processing EBT cards. This could result in an erosion of public trust and undermine the validity and accuracy of DTA’s EBT card data.
Authoritative Guidance
According to DTA officials, each TAO is assigned a unique prefix embedded in the EBT card, which serves as an identifier, allowing staff members and management to recognize which TAO is associated with the card. This step is essential for maintaining accurate records and ensuring that each card is accounted for within the respective TAO. By following this procedure, DTA staff members help facilitate better management and tracking of benefits and services provided to clients.
Reasons for Issue
According to DTA officials, DTA does not conduct reviews of the EBT card issuance reports generated by the EPPIC system. Additionally, DTA does not have documented policies and procedures on reconciling EBT cards issued from each TAO to the EPPIC system. DTA did not update the TAOs within the EPPIC system.
Recommendations
- DTA should contact its third-party vendor to deactivate all inactive and non-TAO locations from the EPPIC system.
- DTA should develop and implement policies and procedures to conduct regular EBT card issuance data reconciliations to ensure that all issued EBT cards are assigned to an accurate TAO.
- DTA should separately create policies and procedures for periodic review of TAOs in the EPPIC system to ensure that the system is consistently updated.
Auditee’s Response
Regarding the first recommendation, DTA agrees that TAO locations should be accurate and up to date in the EBT vendor’s electronic system. As previously noted, DTA no longer utilizes . . . [the] EPPIC system. It will work with its new vendor . . . to ensure that ebtEDGE is up to date. DTA does not dispute that the EPPIC data from the audit period reflects that 35,543 EBT cards were assigned to inactive TAOs. However, DTA disagrees with [the Office of the State Auditor’s (OSA’s)] finding that staff could choose one of these inactive locations when issuing EBT cards and undermine the accuracy of DTA’s EBT card data. While the EPPIC data reflects that those EBT cards were assigned to those inactive locations, EBT cards were not printed from those inactive locations.
In the authoritative guidance section of the draft report, OSA notes that DTA officials told it about the unique prefix embedded in EBT cards that identifies the TAO associated with the card. DTA apologies for any misunderstanding that led to OSA believing this prefix is essential for recordkeeping or accountability. Instead, the office code in the EBT system reflects the geographic catchment area with which the case is associated; it corresponds with the client’s zip code. The location is assigned by the system, and DTA staff cannot choose it. That code also does not play a role in DTA’s contemporary EBT card inventory results, nor does it inform how DTA tracks benefits and services provided to clients.
In addition, while DTA does not dispute that the EPPIC data may reflect 97,654 EBT cards issued from non-TAO locations, DTA asserts this is an immaterial finding for the reasons set forth above. In the new . . . system, DTA also utilizes reports to track EBT card printing based on the office assignment of the DTA staff person printing the card. Therefore, to the extent OSA is concerned that an inaccurate office location in the EBT vendor’s electronic database may undermine the public’s trust in DTA’s EBT card data, DTA tracks and maintains EBT card printing data internally in a comprehensive and accurate manner outside of its EBT vendor’s administrative terminal.
Regarding the second recommendation, DTA agrees it is important that DTA is properly inventorying its EBT cards, and accurately tracking the location at which they are printed. However, DTA does not believe that it is essential for EBT cards to be assigned to a specific TAO. While the data in the EBT vendor’s database should be current and accurate, there is no significant effect on the agency nor clients if it is not. DTA already has a system in place to ensure that EBT cards are properly inventoried and tracked by printing location. In short, TAOs must submit an order request for EBT cards, and the Benefit Issuance team creates an Adobe document for a member of the TAO to sign when the order is received. Any discrepancy in the order must be annotated on the form and returned to the Benefit Issuance team. In addition, every month, the Benefit Issuance team reviews and reconciles all TAO EBT card shipments.
Regarding the third recommendation, DTA agrees that the TAO locations in its EBT vendor’s electronic system should be regularly updated. Building on the experience the Department had with its previous vendor, DTA specified certain enhanced levels of service in . . . ebtEDGE. However, DTA will confer with [the new vendor] to ensure that this data point is consistently updated and accurately maintained.
Auditor’s Reply
We acknowledge DTA’s response and its agreement with the recommendations to ensure that TAO location data in the EPPIC system is accurate and regularly updated. However, we maintain our findings regarding inaccurate TAO information and the assignment of EBT cards to inactive or non-TAO locations during the audit period.
DTA stated that staff members are unable to issue cards from inactive TAO locations and that the location codes in the EPPIC system are determined solely by client zip codes. However, during our meeting with DTA officials on July 18, 2024, we were provided with TAO locations that have specific prefixes on the cards. For example, prefix 1375 was assigned to the Central EBT Processing Center, prefix 1322 was assigned to the Chelsea TAO, and prefix 1302 was assigned to the Brockton TAO. These prefixes were embedded on each EBT card to serve as identifiers for the respective TAO locations. In our analysis of EBT card issuance data, we matched the prefixes to the specific TAOs they represented.
DTA also stated that the issuance of 97,654 cards from non-TAO locations is immaterial; it is important to note that, for performance audits, we focus on program efficiency rather than materiality. The lack of accurate location data in the EPPIC system undermines the completeness and reliability of the records used to monitor the EBT program. We acknowledge that DTA has implemented internal tracking mechanisms; however, those tools cannot replace the need for its system to have accurate and up-to-date location information.
We acknowledge DTA’s transition to a new EBT vendor to improve reporting capabilities as described. This occurred after the audit period and does not affect the validity of the findings related to inaccuracies in the EPPIC system data during the audit period. Based on its response, DTA is taking steps to address our concerns in this area. We will monitor progress on this matter as part of our post-audit review process in approximately six months.
| Date published: | January 30, 2026 |
|---|