DIRECTIVE	:	UNEMPLOYMENT INSURANCE PROGRAM LETTER NO. 42-93
TO	:	ALL STATE EMPLOYMENT SECURITY AGENCIES
FROM	:	BARBARA ANN FARMER Administrator for Regional Management
SUBJECT	:	Revenue Quality Control, Policy end Program Design

1. Purpose. To provide State agencies with a revised Draft RQC Handbook, which replaces the March 1993 version and to update the status of the Core Revenue Quality control (RQC) project, to address comments in response to Unemployment Insurance Program Letter (UIPL) No. 4-93, and to introduce changes initiated during four National office (NO) training sessions for RQC Reviewers.

2. Reference. UIPL No. 4-93 (November 9, 1992); UIPL No. 24-92 (April 23, 1992); UIPL No. 44-90 (September 21, 1990): UIPL No. 8-90 (November 15, 1989), UIPL No. 14-89 (February 8, 1989).

3. Background. This is the sixth in a series of program letters issued by the Unemployment Insurance Service (UIS) to report on RQC Task Force progress in RQC design and development. The RQC Task Force comprises UIS NO staff, ETA Regional Office (RO) staff and State tax program specialists on temporary IPA assignments with the N0. A panel of State UI administrators and tax managers provided expert advice to the project's technical consultants.

   The RQC program is promised on a foundation of full State involvement in establishing a reliable system to measure the quality and effectiveness of State unemployment compensation, (UC) tax systems. The program encompasses all aspects of State tax operations including employer liability, collection and compliance methods, field tax operations and adherence to Federal and State standards, laws, regulations and operating procedures.

   To recap milestones in the nearly four years of RQC evolution:

   • The ROC program went through a two year design and development phase, identifying the major tax functions that should be subject to review, determining the objectives of each tax function and developing methodologies to measure attainment of objectives.

   • In 1990, volunteers from eight states, recruited and trained by the RQC Task Force, conducted a pre-test of, the draft RQC Program Review Methodology.

   • From experience gained in the pre-test, the design was remodeled extensively and the computer measures were developed and added.

   • An operating draft of a proposed "core" RQC Handbook was completed and circulated.

   • In 1991, Pilot Testing was announced and sight States were selected for voluntary participation. Training was held for reviewers from the States and they conducted full RQC program reviews in their States.

   • Again, significant design changes were made as a result of the pilot tests. Most significantly, the acceptance sample was changed from 46 cases to 60 cases.

   • The RQC Handbook was revised to include the new acceptance sample criteria, reallocation of the computed measures to specific activity chapters and to reflect the experience gained in the Pilot tests. The new Draft "Core" RQC Handbooks were issued to all State agenaia and ETA RO under UIPL 4-93 in November 1995. Comments were requested and were received from several states and ROs.

   • A video of the RQC Overview vas produced to introduce State administrators and tax staff to this now program and to announce voluntary implementation, which would take place in 1993.

   • Four identical RQC reviewer training sessions have base held at the NO (March 22 through 26, 1993; April 5 through 9, 1993; April 19 through 23, 1993; and May 10 through 14, 1993).

   • Forty nine states are participating in voluntary implementation of the RQC design which began in January 1993.

   • Mandatory implementation is planned for late in Calendar Year 1994 (CY 94).

4. Issues and Answers. In addition to the comments received from States and ROs in response to UIPL 4-93, the Task Force has addressed program design recommendations developed in the four RQC Reviewer training sessions. Significant recommendations are reflected in the revised (draft) RQC Operations Handbook (Attachment II). Representative issues raised and the actions taken are addressed in the accompanying Issues/fiction paper (Attachment I).

   Significant among the changes:

   • In the ADP area -- sampling specifications have boon completely rewritten, providing better detail and giving the states wide latitude in selecting samples, using a redesigned No software, creating their own software or using a manual system. The computed measures appendix has been completely rewritten and reformatted providing more information on haw the measures are calculated and what information is needed.

   • A table of sample sizes for universes with less than 1200 records was added to section III of the Appendix. Also, record layouts were added for control records for cashiering, benefit charging, and experience rating as input to the sample Determination (PICNMBR) program, used to determine transactions to select for the sample.

   • A new pass/fail column, to identity quickly whether the case has passed or failed the review, has been added to the Acceptance Sample Coding Sheet.

   • The maximum acceptable error rate for tax transactions reviewed by RQC is 5 percent for all cases. To test against the 5 percent error limit during the pilot, acceptance samples consisting of 46 cases ware drawn, with no exceptions allowed. This gave a 90 percent chance of detecting unacceptable systems, but also carried a high risk of failing acceptable systems. Actually, in the pilot vary few acceptance samples failed after the system review showed no risk. However, for the period of voluntary implementation, a less stringent acceptance sampling was considered advisable to better balance the risks of failing to detect defective systems and erroneously declaring acceptable systems "defective". The new interim rule is: all acceptance samples are of 60 cases, and two exceptions or fewer are acceptable. By taking more cases but allowing up to two exceptions, this rule gives acceptable systems more room to pass (e.g., if the underlying error rates is 2 percent, 90 percent will pass.), although it is weaker at detecting unacceptable systems. The rule will be reviewed in the-light of experience with voluntary implementation. The larger samples will also make it easier to estimate the true expected error rates in tax functions and thus set a new sampling rule.

   • The acceptance sample instructions for debits/ billings for reimbursing employers were modified to accommodate divergent State practices in reimbursing employer accounting billing methods.

   • An acceptance sample of 60 remittances has also been added to the cashiering function to verify the proper posting of remittances to employers' accounts. This is in addition to the three verification tests used in the pilot study.

   • The Employment Security Manual (ESM) , Part V, Field Audit Function, was revised to bring audit documentation requirements in line with professional auditing standards. (See UIPL No. 18-93.) The RQC criteria for measuring compliance with the new EM guidance on audit documentation are included in chapter VII of the RQC Handbook (7/1/93 edition ).

   • The universe for audit sample seleation is the same as the audits reported on the ETA-581 in item 39 (cumulative fox quarters in review year).

   • A model state AQC Annual Report. outline is under development.

   It is anticipated that additional changes will be made in the program prior to mandatory implementation late in CY 94, based on the experience gained by States participating in the voluntary implementation phase now in place.

5. Action Required. State agency Administrators are requested to ensure that this directive and the attachments are made available to UI staff responsible for coordinating and performing RQC reviews, and to appropriate ADP and UC tax managers.

6. Inquiries. Direct all inquiries to the appropriate ETA RO.

7. Attachments.