Audit of the Massachusetts State Retirement Board Objectives, Scope, and Methodology

An overview of the purpose and process of auditing the Massachusetts State Retirement Board.

Overview

In accordance with Section 12 of Chapter 11 of the Massachusetts General Laws, the Office of the State Auditor (OSA) has conducted a performance audit of certain activities of the Massachusetts State Retirement Board (MSRB) for the period July 1, 2017 through June 30, 2019.

We conducted this performance audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Below is a list of our audit objectives, indicating each question we intended our audit to answer, the conclusion we reached regarding each objective, and where each objective is discussed in the audit findings.  

Objective	Conclusion
Does MSRB ensure that members receive their first pension payments within the timeframe established by Section 13(1)(b) of Chapter 32 of the General Laws?	No; see Finding 1
Has MSRB taken corrective measures to address the issues identified in the prior OSA audit (Audit No. 2015-0088-3S) with its benefit eligibility verification procedures?	No; see Finding 2

 

To achieve our audit objectives, we gained an understanding of MSRB’s internal control environment related to our audit objectives by reviewing agency policies and procedures, as well as conducting inquiries with MSRB’s staff and management. We reviewed, and tested the operating effectiveness of, internal controls related to the processing of new retirees’ first pension payments and MSRB’s benefit eligibility verification procedures.

To obtain sufficient, appropriate audit evidence to address our audit objectives, we conducted further audit testing as follows.

• To determine whether MSRB processed first pension payments within the required timeframe, we performed the following procedures.
• MSRB gave us a list of all 6,032 new retirements with effective dates during our audit period from the Massachusetts Retirement Information System (MARIS), MSRB’s benefit-administration system. For each new retiree, we calculated the amount of time between the effective date of retirement and the date of the first pension payment.
• We then grouped, or stratified, the population of new retirements during our audit period into four categories based on the number of days between the effective date of retirement and the date of the first pension payment:

Segment	Days to Process First Payment	Population	Percentage of Total
Noncompliant—General	63 to 179 days	5,559	92.2%
Noncompliant—Outlier	180 days or more	289	4.8%
In Compliance*	62 days or less	175	2.9%
Data Issues	Data fields were blank or missing	9	0.1%
Total		6,032	100%

*     For audit testing purposes, we defined “in compliance” as “processed within 62 days.” By statute, “in compliance” can range from 28 to 62 days depending on the day of the month retirements are effective, as outlined in the “Overview of Audited Entity.” section of this report under “Retirement Application Process.”

• We selected a statistical random sample for testing, with a 95% confidence level and a 0% expected error rate, of 60 of the 5,559 retirements in the Noncompliant—General segment (processing times of 63 to 179 days). We also selected a nonstatistical random sample of 40 of the 289 Noncompliant—Outlier retirements (processing times of 180 days or more) for testing. We analyzed the retirement application processing timeline for the selected cases to determine whether there were any underlying reasons for the delays in processing first payments. We reviewed supporting documentation (such as retirement applications, application checklists, application receipt acknowledgement letters, Salary Request and Release Forms, benefit request sheets, data and annuity sheets, first pay letters, and OnBase⁶ workflow reports) to identify any occurrences of similar circumstances, common causes for delays, or other trends in the data that may have contributed to the delays.
• We randomly selected a nonstatistical sample of 35 of the 175 In Compliance retirements. We reviewed retirement applications for completeness and the content of first pay letters to determine whether there was any evidence of common attributes, data trends, or potential best practices that may have contributed to the favorable processing times.
• We also tested all nine retirements that had incomplete information to determine underlying causes for the data omissions or any irregularities related to the retirement data being incomplete. For all nine retirements, certain data elements such as payment amount were incomplete because the retiree had died shortly after the retirement date.
• To determine whether MSRB had taken corrective measures to address the issues identified with its benefit eligibility verification procedures in our previous audit, we performed the following procedures.
• We reviewed the procedures conducted during MSRB’s last benefit verification process, which was initiated in April 2018. Specifically, we obtained a list, as of March 2018, of 62,650 then-current retirees and Option C beneficiaries used for the mailing of the 2018 Benefit Verification Forms (BVFs). We selected a statistical random sample, with a 95% confidence level and a 0% expected error rate, of 60 of the 62,650 benefit recipients and reviewed their files to determine whether BVFs were signed by the benefit recipients and notarized.

• MSRB gave us a list of all 179 benefit recipients who did not return BVFs after the third, final mailing in January 2019. We selected a nonstatistical, random sample of 20 and reviewed supporting documentation, such as death certificates and pension warrants (lists of monthly benefit payments), to determine whether the recipients had been reported as deceased and whether benefit payments had been terminated.

• To determine whether MSRB promptly identified deceased members and beneficiaries and satisfactorily adjusted benefit payments, we obtained a system-generated list of all 4,427 retirees and Option C beneficiaries whose recorded dates of death occurred during our audit period. We selected a statistical random sample, with a 95% confidence level and a 0% expected error rate, of 60 deceased benefit recipients and reviewed supporting documentation (such as retirement applications, BVFs, death certificates, death notices, obituaries, and correspondence) to determine whether death certificates were on file; dates of death were promptly and accurately recorded in MARIS; and appropriate actions were taken, including accurate and timely adjustments to benefit payments when necessary.

• MSRB gave us all 92 data files it had received during our audit period from its third-party vendors. These data files contained listings of possible deaths⁷ identified by the vendors cross-matching MSRB’s entire retiree payee file with their own death data. We selected a nonstatistical, random sample of 20 of the 92 files for testing. From each of the 20 files, we judgmentally selected three possible deaths and reviewed supporting documentation (such as retirement applications, BVFs, death certificates, death notices, obituaries, and other correspondence) to determine whether the names and Social Security numbers were contained in the MSRB retiree payee files that MSRB provided to the vendors and the list of recorded deaths during the audit period. We also checked that the dates of death were promptly and accurately recorded in MARIS and appropriate actions were taken, including the accurate and timely resolution of any necessary adjustments to benefit payments.

Where nonstatistical sampling was used, we could not project the results of our testing to the overall populations.

Data Reliability

We assessed the reliability of the data obtained from MARIS, MSRB’s benefit administration system, by interviewing knowledgeable personnel at the agency about the system and testing the data for duplicate records and dates outside our audit period. We also verified the number of records in each data population by comparing the total number of records to those in other data sources such as annual reports and pension warrants. Further, we traced samples of records, from the data provided to us, to and from original source documents (such as retirement applications, BVFs, death certificates, death notices, and pension warrants) for completeness and accuracy. We determined that the information obtained from MARIS for our audit period was sufficiently reliable for our audit work.

6.     OnBase is a document imaging and workflow management application.

7.     The vendors’ match produces a file of deceased individuals whose names and Social Security numbers closely match those of individuals in MSRB’s retiree payee file. MSRB conducts additional research to determine whether the decedents are MSRB payees.

Date published:	January 6, 2021