Audit Report: Maryland Department of Labor Division of Unemployment

Audit Report
Maryland Department of Labor

Division of Unemployment Insurance
Part 2
Unemployment Benefits
November 2022
Public Notice
In compliance with the requirements of the State Government Article Section

2-1224(i), of the Annotated Code of Maryland, the Office of Legislative
Audits has redacted cybersecurity findings and related auditee responses
from this public report.
OFFICE OF LEGISLATIVE AUDITS

DEPARTMENT OF LEGISLATIVE SERVICES
MARYLAND GENERAL ASSEMBLY
Joint Audit and Evaluation Committee
Senator Clarence K. Lam, M.D. (Senate Chair) Delegate Mark S. Chang (House Chair)
Senator Malcolm L. Augustine Delegate Steven J. Arentz
Senator Adelaide C. Eckardt Delegate Nicholas P. Charles II
Senator George C. Edwards Delegate Andrea Fletcher Harrison
Senator Katie Fry Hester Delegate Trent M. Kittleman
Senator Cheryl C. Kagan Delegate Carol L. Krimm
Senator Benjamin F. Kramer Delegate David Moon
Senator Cory V. McCray Delegate Julie Palakovich Carr
Senator Justin D. Ready Delegate Elizabeth G. Proctor
Senator Craig J. Zucker Delegate Geraldine Valentino-Smith
To Obtain Further Information

Office of Legislative Audits
The Warehouse at Camden Yards
351 West Camden Street, Suite 400
Baltimore, Maryland 21201
Phone: 410-946-5900
Maryland Relay: 711
TTY: 410-946-5401 · 301-970-5401
E-mail: [email protected]
Website: www.ola.state.md.us
To Report Fraud
The Office of Legislative Audits operates a Fraud Hotline to report fraud, waste, or abuse involving State
of Maryland government resources. Reports of fraud, waste, or abuse may be communicated anonymously
by a toll-free call to 1-877-FRAUD-11, by mail to the Fraud Hotline, c/o Office of Legislative Audits, or
through the Office’s website.
Nondiscrimination Statement
The Department of Legislative Services does not discriminate on the basis of age, ancestry, color, creed,
marital status, national origin, race, religion, gender, gender identity, sexual orientation, or disability in the
admission or access to its programs, services, or activities. The Department’s Information Officer has been
designated to coordinate compliance with the nondiscrimination requirements contained in Section 35.107
of the United States Department of Justice Regulations. Requests for assistance should be directed to the
Information Officer at 410-946-5400 or 410-970-5400.
November 15, 2022
Senator Clarence K. Lam, M.D., Senate Chair, Joint Audit and Evaluation Committee
Delegate Mark S. Chang, House Chair, Joint Audit and Evaluation Committee
Members of Joint Audit and Evaluation Committee
Annapolis, Maryland
Ladies and Gentlemen:
We have conducted a fiscal compliance audit of the Maryland Department of

Labor (MDL) – Division of Unemployment Insurance (DUI) for the period
beginning April 17, 2017 and ending November 15, 2020. DUI administers the
State’s Unemployment Insurance Program, which is normally funded primarily by
unemployment insurance tax contributions collected from employers. However,
DUI operations were significantly impacted by the COVID-19 pandemic, which
greatly increased unemployment insurance activity and required the expansion
and modification of various DUI operations, which were primarily federally
funded. While we recognize the impact on DUI’s operations, the circumstances
created by the pandemic did not materially affect our decision as to the inclusion
of the specific findings and related recommendations in our report, which
represent our determination of DUI’s statutory compliance or the appropriate and
necessary controls over Program elements audited. However, as a result of the
significant impact of the pandemic on DUI, we did divide our audit into two parts
as further described below. This report addresses the second part of our audit. A
report on the first part of our audit was issued May 4, 2022.
Our audit disclosed that DUI did not conduct certain critical data matches used to
identify potentially fraudulent or improper claims. We conducted three matches
replicating four discontinued DUI matches and identified at least $32.3 million in
potentially improper benefit payments. In addition, DUI did not have sufficient
procedures to ensure that individuals filing claims using a foreign Internet
Protocol (IP) address were eligible for benefits, including 3,724 claimants who
received $3.6 million in benefit payments between September 2017 and April
2020. Similarly, DUI lacked procedures to help prevent and detect duplicate
payments, and our analysis disclosed $43.3 million in potentially duplicate
payments made to 12,500 claimants between April 2020 and December 2021.
With regard to one program in particular, our audit disclosed that DUI did not
conduct timely verifications of income reported by claimants as required,
resulting in potential overpayments for this program, which as of January 2021
had paid $5.9 billion in benefits in Maryland. Furthermore, DUI did not
adequately review regular claims and adjudications, such as decisions regarding
claimant eligibility that were processed by DUI employees and temporary staff.
Finally, the inability of BEACON (DUI’s automated benefits system) to provide
certain critical data regarding entries and adjustments that had to be recorded
manually severely restricted DUI’s ability to verify the propriety of those
transactions.
We also found that DUI did not establish sufficient controls over reissued debit
cards, which totaled 354,445 between July 2017 and January 2021; and did not
ensure the proper disposition of funds remaining on expired and never activated
cards, which totaled $23.1 million as of October 2021. Furthermore, DUI did not
properly account for in BEACON the status of potentially fraudulent benefits
totaling $493.9 million that were removed from claimants’ debit cards.
In addition, we noted that DUI did not ensure that amounts disbursed from the
Unemployment Insurance Trust Fund were properly transferred to the bank
account used to make benefit payments. Furthermore, we also noted information
system security deficiencies. However, in accordance with the State Government
Article, Section 2-1224(i) of the Annotated Code of Maryland, we have redacted
these findings from this audit report. Specifically, State law requires the Office of
Legislative Audits to redact cybersecurity-related findings in a manner consistent
with auditing best practices before the report is made available to the public. The
term “cybersecurity” is defined in the State Finance and Procurement Article,
Section 3A-301(b), and using our professional judgment we have determined that
the redacted findings fall under the referenced definition. The specifics of the
cybersecurity findings were previously communicated to DUI as well as those
parties responsible for acting on our recommendations.
Finally, our audit included a review to determine the status of four of the six
findings contained in our preceding audit report. We determined that DUI
satisfactorily addressed one of these four findings. The remaining three findings
are repeated in this report as four findings. The status of the remaining two
2
findings in our preceding audit report was previously determined during our audit
of DUI Part 1 report dated May 4, 2022.
We determined that DUI’s accountability and compliance level was

unsatisfactory, in accordance with the rating system we established in conformity
with State law. The primary factors contributing to the unsatisfactory rating were
the significance of our audit findings in both audit reports and the number of
findings across all areas of DUI’s operations.
MDL’s response to this audit, on behalf of DUI, is included as an appendix to this

report. We reviewed the response and noted agreement to our findings and
related recommendations. While there are other aspects of MDL’s response
which will require further clarification, including certain comments that are not
consistent with the report analysis, we do not anticipate that these will require the
Joint Audit and Evaluation Committee’s attention to resolve. Finally, we have
edited MDL’s response to remove certain vendor names or products, as allowed
by our policy. Consistent with the requirements of State Law, we have redacted
the elements of MDL’s response related to cybersecurity findings.
We wish to acknowledge the cooperation extended to us during the course of this

audit by DUI. We also wish to acknowledge MDL’s and DUI’s willingness to
address the audit issues and implement appropriate corrective actions.
Respectfully submitted,
Gregory A. Hook, CPA

Legislative Auditor
3
4
Table of Contents
Background Information 8
Agency Responsibilities 8
Maryland Unemployment Insurance Taxes 9
Unemployment Benefits 10
Coronavirus Aid, Relief, and Economic Security Act 11
Unemployment Insurance System 12
Legal Matters 14
Maryland Unemployment Insurance Trust Fund 14
Claims Center Staffing 14
Claims Processing 18
National Trends and Issues Related to Potential Unemployment 20
Insurance Fraud During the COVID-19 Pandemic
Status of Findings From Preceding Audit Report 22
Findings and Recommendations 24
Benefit Payments
Finding 1 – The Division of Unemployment Insurance (DUI) did not 25
conduct certain critical matches used to identify potentially
fraudulent or improper claims. We conducted three matches to
replicate four of the discontinued DUI matches and identified
at least $32.3 million in potentially improper payments.
* Finding 2 – DUI did not have comprehensive procedures to ensure 27

that individuals filing claims using a foreign Internet Protocol
address were eligible to receive benefits, including 3,724
claimants that received benefit payments totaling $3.6 million.
* Finding 3 – DUI did not ensure claimants who were full-time 30

students were eligible for benefits, and that all claimants were
enrolled in the Maryland Workforce Exchange System, as
required.
 Denotes item repeated in full or part from preceding audit report
5
Claims Processing
Finding 4 – DUI did not have procedures to help prevent and detect 31
duplicate benefit payments. Our analysis disclosed $43.3 million
in potentially duplicate payments made to 12,500 claimants
between April 2020 and December 2021 that were not identified or
investigated by DUI.
Finding 5 – DUI did not conduct timely verifications of income reported 33

by applicants for Pandemic Unemployment Assistance benefits and
did not ensure manual adjustments processed by DUI and contract
employees were proper.
* Finding 6 – DUI did not adequately review regular claims and 34

adjudications processed by claims center DUI employees and
temporary staff, and output reports of manual wage entries could
not be generated from BEACON for verification purposes.
* Finding 7 – DUI did not establish sufficient controls over reissued 36

debit cards, and did not ensure the proper disposition of funds
remaining on expired debit cards.
Finding 8 – DUI did not properly account for potentially fraudulent 38

benefits totaling $493.9 million that were removed from claimants’
debit cards.
Unemployment Insurance Trust Fund

Finding 9 – DUI did not ensure amounts disbursed from the 39
Unemployment Insurance Trust Fund were properly transferred to
the bank account used to make benefit payments.
Information Systems Security and Control

Finding 10 – Redacted cybersecurity-related finding 40
Audit Scope, Objectives, and Methodology 41
 Denotes item repeated in full or part from preceding audit report
6
Exhibit 1 – Summary of Unemployment Insurance Modernization 45
(BEACON) Findings in Office of Legislative Audits Audit
Reports Issued from July 1, 2015 to May 31, 2022.
Exhibit 2 – Selected Sources for National Trends 47
Agency Response Appendix
7
Background Information
Agency Responsibilities
The Division of Unemployment Insurance (DUI) is a separate budgetary unit

within the Maryland Department of Labor. According to the State’s records,
DUI’s fiscal year 2020 operating expenditures (excluding unemployment benefit
disbursements) totaled approximately $61.4 million. DUI administers the State’s
Unemployment Insurance Program that includes the following primary
responsibilities.
 Collecting unemployment insurance tax contributions from employers

 Processing applications for, and disbursing unemployment benefits
As further described below, during the audit period, DUI operations were
significantly impacted by the COVID-19 pandemic that greatly increased
unemployment insurance activity and required the expansion and modification of
various DUI operations. In addition, DUI finalized the implementation of its new
information system, BEACON, which further impacted its operations. As a
result, and to provide necessary audit resources and coverage, we have divided
our audit of DUI into the following two parts to address the aforementioned DUI
responsibilities.
Part 1 – Unemployment Insurance Tax Contributions

Includes employer unemployment contributions, reimbursements from
government agencies and certain non-profit organizations, associated
accounts receivable activity, and related changes from a recent system
implementation.
Part 2 – Unemployment Benefits

Includes methods individuals can use to file for unemployment insurance
benefits, eligibility and monetary benefit determinations (for State
unemployment insurance and for additional programs in response to the
COVID-19 pandemic), payment monitoring, prevention of fraudulent claims,
and related changes from a recent system implementation.
This report addresses Part 2 of our audit. Our report on Part 1 was issued May 4,
2022.
8
Maryland Unemployment Insurance Taxes
Maryland employers are required to remit Maryland unemployment insurance

taxes based on a percentage of wages paid to their employees. During calendar
year 2020, the employer tax rate ranged from 0.3 percent to 7.5 percent of the first
$8,500 of employee wages, and for calendar year 2021, the percentage ranged
between 2.2 percent and 13.5 percent. The tax rates are established by State
regulation based on an annual calculation and estimate of the funds necessary to
meet federal requirements regarding the necessary minimum balance to be
maintained in the Maryland Unemployment Insurance Trust Fund.
The particular tax rate paid by an employer is impacted by various factors, most
notably the amount of unemployment benefits paid by DUI and charged to the
employer’s account during a specified period for eligible employee layoffs and
terminations. Certain entities, such as nonprofit organizations and governmental
entities, are exempt from these taxes and, instead, reimburse the State for any
unemployment benefits paid by the State on their behalf. As of November 2020,
an unemployed individual could receive a maximum of $430 per week for 26
weeks of State benefits in one benefit year.1
Historically, employer contributions were the primary source of unemployment

funds. As shown in Figure 1 on the following page, the COVID-19 pandemic
resulted in unprecedented increases in federal funding. Specifically, according to
DUI records, during fiscal year 2021, DUI collected approximately $710.9
million from employers ($637.0 million in unemployment insurance taxes and
$73.9 million in benefit reimbursements) and received $7.6 billion in federal
funding.
1
As a result of the federal Coronavirus Aid, Relief, and Economic Security (CARES) Act, 52
weeks of benefits were available from the combination of standard Maryland benefits, Pandemic
Emergency Unemployment Compensation (PEUC), and extended benefits.
9
Unemployment Benefits
As noted above, the COVID-19 pandemic had a significant impact on

unemployment claims (see Figure 2 on following page). According to DUI
records, there were approximately 103,000 paid claimants in fiscal year 2019
compared to more than 642,000 claimants in fiscal year 2021 which were paid
approximately $8.8 billion in unemployment insurance benefits. Unemployment
insurance claims peaked in May 2020 at over 300,000 claims. According to the
United States Bureau of Labor Statistics, the unemployment rate in Maryland in
May 2020 was 9.0 percent.
10
Benefit amounts are based on the applicant’s earnings and other factors (such as
the number of dependents). State law requires applicants to be able, available,
and actively looking for work (work search requirement) in order to be eligible
for unemployment benefits. The work search requirement, but not the able and
available requirements, was suspended from March 2020 through July 2021 due
to the COVID-19 pandemic.
Coronavirus Aid, Relief, and Economic Security Act
On March 27, 2020, in response to the COVID-19 pandemic, the federal

Coronavirus Aid, Relief, and Economic Security (CARES) Act was signed into
law. The federal law included several provisions affecting the unemployment
insurance program administered by DUI.
 Pandemic Unemployment Assistance (PUA), effective January 27, 2020

through December 31, 2020, provided up to 39 weeks of benefits to covered
individuals who were not eligible for regular benefits or extended benefits.
One of the major populations included under this program were self-employed
individuals who historically were not eligible for unemployment benefits.
 Federal Pandemic Unemployment Compensation (FPUC), effective March 29,
2020 through July 31, 2020, provided $600 in addition to the normal weekly
benefit amount.
11
 Pandemic Emergency Unemployment Compensation (PEUC), effective
March 29, 2020 through December 31, 2020, added up to 13 additional weeks
for claimants who exhausted their initial 26 weeks of regular benefits.
Other federal funding was awarded in addition, and subsequent, to CARES.
 Lost Wage Assistance (LWA) Program (provided by Presidential

Authorization), effective August 1, 2020 through December 27, 2020,
provided claimants an additional $300 per week for six weeks.
 Extended benefits (EB) became available from May 31, 2020 through
December 12, 2020, and provided up to 13 additional weeks of benefits. A
claimant had to exhaust their 39 weeks of benefits under PEUC before
becoming eligible for EB.
 Continued Assistance for Unemployed Workers Act signed into law
December 27, 2020 provided up to 11 additional weeks of the CARES
programs.
 The American Rescue Plan Act was signed into law March 11, 2021 further
extending the CARES programs through September 4, 2021.
Unemployment Insurance System
In 2011, DUI began a major information technology project to replace and

consolidate the following three existing automated systems used to account for
unemployment activities into one system, which was referred to as BEACON.
 Maryland Unemployment Insurance Tax System used to process

employers’ unemployment insurance taxes owed and paid.
 Maryland Automated Benefits System (MABS) used to track payments to
and amounts due from unemployment insurance claimants and wage
information reported by Maryland employers.
 Appeals Case Tracking System used to maintain a record of the status of
each appeal, and notes on postponements, hearings, and other notes.
The new BEACON system was procured as part of a consortium of three states
(Maryland, Vermont, and West Virginia) and was intended to be fully
implemented in 2018. However, significant implementation delays occurred
which DUI asserts were caused by unrealistic timelines, the need to address
vendor quality issues, and certain issues experienced by one of the other states.
As a result, at the onset of the COVID-19 pandemic BEACON had not been fully
implemented in Maryland. In addition, both Vermont and West Virginia
withdrew from the consortium, leaving Maryland to unilaterally implement
BEACON.
12
In April 2020, in response to the COVID-19 pandemic, DUI implemented
BEACON One-Stop to allow claimants to file all claims online, and rigorously
pursued the long-delayed implementation of BEACON, which DUI deemed to be
fully implemented in September 2020. As of January 2021, payments to the
vendor for BEACON implementation since September 2015 totaled $54.4
million, and payments to the same vendor for BEACON One-Stop totaled $2.9
million. The new system (including One-Stop) was paid for primarily with
federal funds.
Our May 1, 2020 report on the Maryland Department of Information Technology

reviewed major information technology development projects including
BEACON. The report addressed several concerns regarding BEACON
development and implementation, such as documentation of project monitoring.
In addition, our January 7, 2021 report on our audit of the Maryland Department
of Labor (MDL) - Office of the Secretary included certain findings relating to
BEACON contract costs. Furthermore, our DUI Part 1 report included additional
findings relating to BEACON deficiencies. See the summary of these findings in
other audit reports related to BEACON in Exhibit 1. Additional deficiencies with
the system are addressed in several of the findings in this report. As of March
2022, MDL had not assessed any permissible damages against the BEACON
vendor for development, implementation, or functionality issues.
As noted in Figure 3, beginning in April 2020 through September 2020, all

claimant information was entered into BEACON One-Stop for processing of
federal claims, such as PUA and EB (but excluding PEUC). MABS was still used
during this period for processing regular State unemployment claims, as well as
PEUC. In late September 2020, DUI began using the fully implemented
BEACON for processing all claim types. Figure 3 indicates a timeline of claim
type and system used.
13
Legal Matters
An indictment was issued on August 23, 2022 by the U.S. District Court v. two
principals of the contractor responsible for the design, development, and
implementation of BEACON. We were advised by MDL management that they
have no reason to believe that the contractor will not abide by the terms and
conditions of its contract with the State of Maryland.
Maryland Unemployment Insurance Trust Fund
DUI maintains the Maryland Unemployment Insurance Trust Fund (UITF) for the
deposit of unemployment taxes collected from employers and the payment of
benefits to the unemployed. The UITF is required by federal regulation to retain a
balance to cover its expected current obligations. As noted in Figure 1, the
average trust fund balance exceeded $1.3 billion for fiscal years 2017 through
2019. The COVID-19 pandemic significantly impacted the balance during fiscal
year 2021 due to its unforeseen nature and the obvious exclusion of its impact
from the preceding tax calculation performed by the State. As a result of the low
balance, DUI borrowed $68.9 million from the federal government to cover
obligations during the period February 2021 through April 2021, which was
subsequently repaid.
Claims Center Staffing
DUI maintains four claims centers with approximately 80 claims processors who
receive calls and provide assistance with any questions on filing, applying for, and
receiving unemployment benefits. Additionally, at these same centers, DUI
maintains approximately 75 adjudicators.2 The aforementioned claims processors
and adjudicators are all State employees. Claims processors generally correspond
with claimants to obtain information required for an unemployment claim,
whereas claims adjudicators research discrepancies with claims filed, such as
income reported by a claimant that does not agree with the corresponding amount
reported by the employer.
During our audit period, DUI awarded two emergency staffing contracts to assist
with the increased volume of unemployment insurance claims resulting from the
COVID-19 pandemic.
2
There are also two adjudication-only centers.
14
Claims Staffing Vendor Contract
In April 2020, DUI entered into a $19.6 million emergency contract with a vendor
for 200 supplemental staff to augment DUI’s claims centers. These vendor
employees were to handle the increase in claim volume and complete certain
tasks, including assisting in the review of PUA proof of income and identity
documentation submitted for claims identified as potentially fraudulent.
As DUI’s needs changed, this contract had multiple change orders amending
staffing levels, total contract amount, and end date. Figure 4 shows change orders
processed from inception of the contract through January 1, 2022. Costs
associated with each change order varied depending on the number of staff
provided at the time. As of November 2021, payments to the vendor totaled
$93.3 million all of which were made using federal funds.
As noted in Figure 5 on the following page, the vendor provided an increasing

number of staff at any given time during calendar year 2021, ranging from 554 in
January 2021 to 2,965 in October 2021. These figures may exceed the contract
totals in Figure 4 because more than one individual may have been provided by
the vendor to cover the required hours for one full time equivalent contract
position.
15
Vendor employees assisted in addressing the signficant number of calls being
received. Figure 6, on the following page, shows total calls received by vendor
employees between May 2020 and January 2022, which peaked at 5.9 million
calls during the month of January 2021, as well as the number of calls actually
handled (answered) by vendor employees during this period. As noted in Figure
6, despite the vendor’s efforts, there was a significant number of calls that were
not handled due to the high call volume. Data regarding the number of calls
received and handled by DUI employees was not available.
16
DUI management asserted to us that the calls received could include multiple
calls from the same individual, and that the number of unique calls (calls
associated to specific individuals) received was more closely aligned with the
number of calls handled. However, DUI could not document that assertion; and
we found that the disparity between calls received and calls handled was still
significant. In addition, the observed disparity is consistent with widely reported
public concerns regarding the inability to contact DUI to resolve issues with filing
a claim and/or the related unemployment insurance benefits. During the period of
our review, our Fraud hotline received numerous calls expressing these concerns,
which our Fraud Investigation Unit was able to refer to appropriate DUI personnel
for resolution.
Adjudication Staffing Contract

In November 2020, DUI entered into a $70.9 million emergency contract with
another staffing vendor for up to 675 supplemental staff. This vendor was
responsible for augmenting DUI’s adjudication staff by completing certain tasks
including the following:
17
 Investigate potentially disqualifying issues and determine the impact on
eligibility for benefits,
 Conduct fact finding interviews and document results,
 Detect improper and potentially fraudulent payments, and
 Assist in reviewing PUA proof of income.
As of November 2021, DUI payments to this vendor totaled $15.8 million, all of
which were made using federal funds. According to the contract, 575 staff were
to be available by June 2021; however, the vendor was unable to obtain that
number of qualified personnel. See Figure 7 for actual staffing levels provided by
this vendor for calendar year 2021.
Claims Processing
Individuals may file an application for unemployment insurance benefits online or

over the phone (which are handled by the four DUI-operated claims centers or
DUI’s supplemental staffing vendor). Historically, requests for certain benefit
types (such as benefits for claimants who were employed by the federal
government) could not be filed online. In April 2020, with the implementation of
BEACON One-Stop, individuals could apply for all unemployment benefits
online.
The application includes information about their previous employer, wages

earned, and the reason for the individual’s current unemployment. DUI utilizes a
combination of automated system checks, and manual steps as deemed necessary,
18
to verify that the application information is complete and the applicant is eligible
for benefits, and to set up an account for each approved claimant.
Benefit amounts are based on the individual applicant’s earnings during their base
period and other factors, such as the number of the applicant’s dependents. The
base period is generally the first four of the last five calendar quarters completed
before the claimant filed for benefits. DUI historically issued each approved
claimant a Maryland Unemployment Insurance (UI) Benefits Debit Card to access
their benefits. However, beginning May 24, 2021, DUI discontinued use of the
debit card, and claimants began receiving their benefits either through direct
deposit or by check.
State law requires applicants to be able, available, and actively looking for work
in order to continue to be eligible for unemployment benefits. Accordingly,
claimants are required to certify each week that they are able, available, and
actively looking for work, and to disclose any other information that could affect
their eligibility for benefits, such as attending school. Any claimant who obtains
work must notify DUI of the number of days worked and the related
compensation.
As a further condition of eligibility, DUI requires claimants to register in the

Maryland Workforce Exchange System within 10 days of filing a claim. The
System provides services to help individuals gain employment. This requirement
was not in place from March 2020 through July 2021 due to the COVID-19
pandemic. Effective July 2021, DUI began automatically initiating this
registration process for claimants, and the Workforce Exchange System became
the only place where a claimant could log their reemployment activities, such as
work searches.
Certain claims may require manual review by a claims worker, and claim data
may need to be manually adjusted. For example, a claims worker may have to
review documentation supporting income reported by a self-employed claimant,
and manually enter the income amount into BEACON. Manual review is also
required for claims in adjudication. A claim is adjudicated when it requires a
claims worker to investigate certain issues and determine the impact on the
claimant’s eligibility, for example whether the claimant was actually terminated
or voluntarily resigned.
19
National Trends and Issues Related to Potential UI Fraud
During the COVID-19 Pandemic
During our audit, we reviewed available publications issued by federal and state
entities, as well as news media reports to identify national trends and issues that
arose during the pandemic relating to claim volume and potential fraud in UI
programs. See publications reviewed at Exhibit 2.
National Increase in Claim Volume Observed

The COVID-19 pandemic and subsequent expansion of UI benefits authorized by
the CARES Act and other federal legislation placed unprecedented stress on
unemployment insurance programs throughout the country. According to the
United States Department of Labor – Office of the Inspector General (US DOL –
OIG), UI claims nationwide increased from 282,000 initial claims immediately
prior to the implementation of pandemic mitigation measures as of March 15,
2020 to 57.4 million initial claims by August 15, 2020. As of June 30, 2022,
approximately $1.036 trillion has been allocated to pandemic-related UI programs
since the beginning of the COVID-19 pandemic.
National Increase in Improper Payments Estimated

In December 2021, the US DOL – Education and Training Administration (US
DOL – ETA) estimated that, based on available information, at least 18.71
percent of regular and certain pandemic UI program payments may have been
improper, although the US DOL – OIG noted that the actual rate was likely
higher. Using this rate, US DOL – OIG estimated in March 2022 that, based on
the $872.5 billion in pandemic UI benefits paid, at least $163 billion may have
been improper payments to claimants. An improper payment may result from an
individual’s fraudulent action to obtain payments or from other conditions, such
as unintentional errors on the part of a claimant or program employee.
In its testimony to Congress, US DOL – OIG cited the Pandemic Unemployment

Assistance (PUA) program as a significant risk for improper payments since it
required only self-certification by the claimant that they met the program
requirements, and allowed the claimant to backdate their UI claim. Additionally,
because of the reliance on self-certification, states were not required to verify
claimant identities or income until January 2021, after Congress amended the
PUA program requirements in December 2020.
Examples of Findings from National and State Reports

Our review of audit reports, advisories, and testimony issued by US DOL – OIG
and several states disclosed that they generally did not discuss the details of
specific UI frauds being perpetrated across the nation. Rather, most reports we
20
reviewed discussed significant challenges experienced by state workforce
agencies (SWAs) responsible for administering unemployment insurance
programs, and focused on internal control deficiencies or program weaknesses,
some of which were revealed or exacerbated by the pandemic conditions. We
noted common elements in many of the weaknesses, several of which were
consistent with results included in this report.
 States had not implemented processes or technology that would allow them to
prevent or detect improper payments. For example, several states did not
perform, or they suspended, crucial data matches designed to detect ineligible
claimants. A survey of states by the US DOL – OIG confirmed that 40
percent of SWAs did not perform the cross-matches required by US DOL –
ETA, and 88 percent of SWAs did not perform the cross-matches strongly
recommended by US DOL – ETA, including, for example, procedures
intended to detect claimants using foreign Internet Protocol addresses (a
matter addressed elsewhere in this report as it relates to Maryland).
 States that had otherwise adequate existing pre-pandemic prevention and

detection processes did not or could not readily adjust them to the
circumstances created by the pandemic. For example, one state that
performed cross-matches previously did not adjust the frequency of a match in
order to address the significant increase in claims volume, and thus more
quickly detect and stop improper benefit payments being made to ineligible
claimants. Other states that normally required claimants to appear in person,
or conducted manual reviews of questionable claims, were unable to use, or
did not adapt, those procedures when the pandemic began.
 Existing processes were not adequate. For example, some states did not have
input controls in place to limit or otherwise aid in ensuring the propriety of
certain application data entered by claimants. Without these limits, applicants
were able to enter inaccurate critical information (such as applicant birthdates
that occurred in the future). Other states paid claimants without verifying that
they were entitled to the amount awarded (such as, by verifying reported
income to supporting documentation and/or wage databases). Although
certain of these transactions may have been identified after the benefits were
paid, recovering the funds may be difficult and therefore would not mitigate
the need for preventive controls (some of these matters are addressed
elsewhere in this report as it relates to Maryland).
21
We Were Unable to Compare Maryland’s Potential Losses from Fraud to Other
States
Due to the diversity of methods used by US DOL and state oversight agencies to
quantify potentially (emphasis added) improper payments made and improper
claims that were detected and not paid, we could not readily compare Maryland’s
totals for such amounts to other states. We also were unable to readily compare
actual losses from fraud in states that had not implemented effective controls to
states that either had implemented effective controls, or had implemented
effective controls later in the pandemic.
We did note that states that implemented more comprehensive fraud detection or
prevention approaches such as analytical procedures on application data, live
identity verification, or multi-state data matching, generally reported a decrease in
the number of potentially fraudulent claims. In addition, the US DOL – OIG
noted that SWAs with modernized information technology (IT) systems were
typically able to implement the programs authorized by the CARES Act more
quickly, and were able to more readily participate in necessary fraud control
measures (such as multi-agency coordination and participation in national data
banks).
Certain Pandemic-related Issues Were Determined Not to be New or

Unforeseeable
In our opinion, certain issues experienced during the COVID-19 pandemic by
state UI programs (including many identified in this audit report) were not new or
unforeseeable. For example, at the onset of the COVID-19 pandemic, an April
2020 US DOL – OIG report raised concerns about program delivery and integrity
due to the historically troubled performance of UI programs at other times of prior
high claim volumes (for example, the Great Recession in 2007-2009), known
difficulties in implementation of adequate fraud detection and prevention
procedures, and a known lack of modernized IT systems. Consequently, the
report recommended that states take appropriate precautions to prevent improper
payments.
Status of Findings From Preceding Audit Report
Based on our current assessment of significance and risk relative to our audit
objectives, our audit included a review to determine the status of five of the six
findings contained in our preceding audit report dated February 5, 2019 (prior
audit finding 3 was not reviewed). As disclosed in Figure 8, we determined that
DUI satisfactorily addressed two of these five findings. The remaining three
findings are repeated in this report as four findings. The status of two findings in
22
our preceding audit report (Findings 5 and 6 also included in Figure 8) were
previously determined during our audit of DUI Part 1 report dated May 4, 2022.
Figure 8
Status of Preceding Findings
Preceding Implementation
Finding Description
Finding Status
DUI did not always use available data to identify
Repeated
claimants who may not be eligible for benefits, and
Finding 1 (Current Findings 2
did not always conduct timely investigations into
and 3)
the results of certain data matches.
Supervisory reviews of claims and adjustments to
claimant wages on the Maryland Automated Repeated
Finding 2
Benefit Systems (MABS) were not always (Current Finding 6)
conducted or documented.
DUI lacked a formal comprehensive policy for
Not repeated
timely collection of delinquent accounts resulting
Finding 3 (Not followed up
from benefit overpayments and referrals to the
on)
State’s Central Collection Unit.
DUI did not establish sufficient controls over
Repeated
Finding 4 reissued debit cards nor ensure the proper
(Current Finding 7)
disposition of funds remaining on expired cards.
DUI did not adequately follow up or track the
results of computer matches it performed to
Not repeated
Finding 5 identify employers that had not registered with DUI
(Part 1 Report)
and may not be remitting required unemployment
insurance taxes.
DUI did not have a comprehensive or effective
procedure to periodically review user access to the
Maryland Unemployment Insurance Tax System Not repeated
Finding 6
and MABS, which resulted in unnecessary or (Part 1 Report)
incompatible access being granted to certain
individuals.
23
Findings and Recommendations
Benefit Payments
Background
The Division of Unemployment Insurance (DUI) has historically conducted
periodic matches using the Maryland Automated Benefit System (MABS) to help
detect benefit payments made to ineligible claimants and identify potential fraud.
These matches each address a critical element involved in the eligibility for or
calculation of unemployment benefits.
1. Incarcerated Match is a monthly match performed by a vendor to ensure that

individuals reported as incarcerated in Maryland, and certain other states and
local jurisdictions are not receiving unemployment benefits.
2. Other States Wage Match is a quarterly match to ensure that claimants are
not earning wages in another state while receiving unemployment benefits in
Maryland.
3. Maryland Wage Match is a quarterly match, which identifies individuals
earning wages from a Maryland employer while receiving unemployment
benefits.
4. Regular State Employee Address Match is a quarterly match to specifically
identify regular Maryland State employees earning wages and receiving
unemployment benefits.
5. Contractual State Employee Address Match is the same as the match for
regular State employees applied to the State’s contractual payroll.
6. Regular State Employee Social Security Number (SSN) Match is a
quarterly match, also meant to identify regular Maryland State employees
earning wages and receiving benefits.
7. Contractual State Employee SSN Match is the same as the match for
regular State employees applied to the State’s contractual payroll.
8. Vital Statistics Match is a monthly match, which ensures that benefits are not
being paid to deceased individuals.
9. New Hire Match is a quarterly match, which ensures that claimants who were
unemployed and are now employed, are no longer receiving benefits.
24
Finding 1
DUI did not conduct certain critical matches used to identify potentially
fraudulent or improper claims. We conducted three matches to replicate
four of the discontinued DUI matches and identified at least $32.3 million in
potentially improper payments.
Analysis
DUI did not conduct certain critical matches used to identify fraudulent or
improper claims and has no plans to retroactively conduct the matches. As noted
in Figure 9, for at least some of the audit period DUI did not perform all of the
nine critical matches it had historically conducted on a periodic basis. For
example,
 Between April 2020 and September 2020, DUI only performed the matches
using data in MABS, which accounted for just $1.3 billion of the $8.4 billion
in claim payments for the period.
 Between September 2020 and January 2021, DUI only performed one of the
aforementioned nine historical matches; specifically, the quarterly New Hire
Match.
 As of December 2021, DUI had not performed two matches (Other State
Wage and Maryland Wage Matches) since March 2020, and had not
performed a third match (Vital Statistics Match) since August 2020.
25
DUI management advised us that they could not conduct the aforementioned
matches because of BEACON One-Stop and BEACON system deficiencies, but
could not provide the specific deficiencies encountered. We were further advised
that while DUI has been able to restart certain matches in BEACON, there was no
plan to retroactively conduct matches for the periods for which matches were not
performed.
These matches are critical since they have historically identified questionable
claims. We conducted three matches to replicate four of the DUI matches using
data obtained from BEACON of individuals receiving unemployment insurance
payments during the period April 2020 to January 2021 or December 2021
(depending on the match). During this period these matches were either not
performed by DUI or did not include all claims paid. Our results are as follows:
State Employee Match (to replicate DUI Regular and Contractual SSN matches)
We compared the BEACON data to data we obtained from the State’s Central
Payroll Bureau for regular and contractual State employees and identified at
least $22.6 million in payments (including $11.5 million in Pandemic
Unemployment Assistance (PUA) claims) to over 6,200 named recipients that
were active State employees when the benefits were paid. We were advised by
the Department of Budget and Management that during the pandemic, generally
State employees continued to receive their full salaries even if they were not
able to work. As a result, these individuals may not have been eligible for some
or all of the benefits received.
Incarceration Match
We compared the BEACON data to data we obtained from the State’s
Department of Public Safety and Correctional Services and identified at least
$7.1 million in payments (including $6.5 million in PUA claims) to over 1,200
named recipients that were incarcerated when the benefits were paid.
Vital Statistics Match

We compared the BEACON data to data we obtained from the Maryland
Department of Health Vital Statistics Administration and identified at least $2.6
million in payments (including $2.1 million in PUA claims) to 402 named
recipients who were deceased when the benefits were paid.
DUI also did not perform periodic matches to identify employees of vendors with
critical access to DUI systems (such as the BEACON and staffing vendor
employees) that may be improperly receiving unemployment insurance payments.
As noted above, during the audit period DUI started using a significant number of
vendors to supplement its State employees (which would not be covered by the
26
aforementioned State Employee (Regular and Contractual) SSN Matches). Our
review disclosed that DUI only completed one match in November 2020 that
included employees of one of the staffing vendors and did not identify any
improper payments. At the time the match was conducted, the staffing vendor
only had 350 employees working on DUI activity. DUI did not repeat the match
as staffing levels from this vendor increased, and did not conduct any matches of
the other vendor’s employees.
The questionable payments identified by these matches do not necessarily mean

that the named recipient received the payment. For example, the named recipient
may have been a victim of identity theft. DUI was not aware of these results and
accordingly had not investigated the related paid benefits to determine whether
they were proper.
Recommendation 1
We recommend that DUI
a. require the BEACON vendor to address any system deficiencies
preventing completion of matches, and in the future ensure that all
matches are performed;
b. conduct the aforementioned matches for the aforementioned periods
when matches were not performed or did not include all claims; and
c. investigate and resolve any potentially improper payments identified by
the matches, including those noted in this finding.
Finding 2
DUI did not have comprehensive procedures to ensure that individuals filing
claims using a foreign Internet Protocol (IP) address were eligible to receive
benefits, including 3,724 claimants that received benefit payments totaling
$3.6 million.
Analysis
DUI did not have comprehensive procedures in place to ensure that individuals
filing for claims using a foreign IP address3 were eligible to receive benefits.
During the audit period a significant number of fraudulent claims were identified
in Maryland and several other states that were filed by individuals from outside of
the respective States. At a minimum, individuals who file from a foreign IP
3
An IP address is a numerical label such as 192.0.2.1 that is connected to a computer network that
uses the Internet Protocol for communication. An IP address serves two main functions: network
interface identification and location addressing (source; IP Address article, Wikipedia, the Free
Encyclopedia). For the purpose of our audit, a foreign IP address is defined as one originating
from other than the 50 states, the District of Columbia, Puerto Rico, the Virgin Islands, Guam,
American Samoa, Northern Mariana Islands, and Canada.
27
address may be residing outside of the country and may be unavailable for work,
which could affect their eligibility for benefits. In April 2021, the federal
Department of Labor issued an Unemployment Insurance Program Letter, which
recommended that state workforce agencies, which includes DUI, use data
analytics from state developed tools or private vendor services to detect
suspicious activity such as out of country IP addresses.
Based on DUI records for the period from September 2017 through April 2020,
there were 3,724 claimants with benefit payments totaling $3.6 million, whose
weekly certifications were filed from a foreign IP address. Our review of DUI
procedures for monitoring these claims disclosed the following conditions:
 DUI did not use available IP address data in MABS to identify and investigate
claimants filing an initial claim or weekly certifications from a foreign IP
address, nor did they have a proactive process to prevent these potentially
improper claims from being processed without a review. Rather, DUI
generally only used the IP address data to augment existing investigations of
possibly ineligible claimants identified through other means. DUI also did not
investigate 503 claimants we identified in our preceding audit report that filed
five or more consecutive weekly certifications from a foreign IP address
between June 2015 and June 2017.
 Although BEACON One-Stop and BEACON included automated controls to

block certain foreign IP addresses, these controls did not block all foreign IP
addresses. Specifically, we examined the 3,724 claimants identified above as
having previously used a foreign IP address under MABS to determine if they
could file claims in BEACON One-Stop and BEACON. Our review disclosed
that 988 of the 3,724 claimants filed weekly certifications from an IP address
that would not have been blocked by the automated controls in place as of
March 2021. Figure 10 provides a visual presentation of where the IP
addresses for certain of those 988 claimants were located (many from
Caribbean islands, Mexico, and Central America).
28
Figure 10
Map of IP Addresses Where Claims Could Still be Submitted
Source: Agency Records
 In April 2020, DUI stopped retaining a record of IP addresses used to submit

weekly certifications and canceled its external subscription services, for which
it paid approximately $2,900 per year, which used geo-mapping to identify the
country associated with a claimant’s IP address. Canceling this service,
impairs DUI’s ability to identify and investigate claims and certifications from
foreign IP addresses, and the ability to update BEACON with a more
complete and comprehensive list of restricted addresses.
A similar condition regarding not using available IP addresses was commented

upon in our preceding audit report.
Recommendation 2
a. continue to develop its automated controls used to block foreign IP
addresses to ensure they are sufficiently comprehensive,
b. retain a record of foreign IP addresses used and formally re-evaluate the
decision to cease the use of a geo-mapping service to aid in the
identification and investigation of foreign IP addresses that are used for
both initial applications and weekly certifications, and
c. investigate the foreign IP addresses identified in this finding and take
corrective action for any ineligible claimants and benefits identified
(repeat).
29
Finding 3
DUI did not ensure claimants who were full-time students were eligible for
benefits, and that all claimants were enrolled in the Maryland Workforce
Exchange System, as required.
Analysis
DUI did not ensure claimants who were full-time students were eligible for
benefits, and that all claimants were enrolled in the Maryland Workforce
Exchange System as required.
 DUI did not obtain data to identify claimants who may not be eligible for
benefits because they were full-time students and were not available for work.
During the initial application and weekly certification process, applicants were
asked if they were attending school (defined by State law as an institution of
higher education). For applicants who were attending full-time and stated
they were available for work, DUI inquired about the applicants’ course
schedules and clarified the requirement to be able and available for work.
DUI advised that the information provided by these applicants was not
verified nor flagged for follow up in subsequent semesters. In addition, DUI
did not have a process to identify claimants who failed to disclose their school
enrollment on their applications or who may have enrolled in school after
submitting their applications. For example, DUI did not obtain enrollment
data from State universities to match against claimant data. Finally, DUI did
not follow up on 179 claimants identified in our prior report who were
enrolled as full-time students, all of whom stated they were able and available
to work, and who received unemployment benefits totaling approximately
$506,000.
 DUI did not use available data in MABS to ensure that claimants had enrolled
in the Maryland Workforce Exchange System. Specifically, no verification
was performed from April 2017 until the Secretary of Labor suspended the
requirement in March 2020.4 The requirement was reinstated in July 2021
and DUI began automatically enrolling claimants in the Maryland Workforce
Exchange System. DUI also did not follow up on 7,724 claimants identified
in our prior report who had not enrolled in the Maryland Workforce Exchange
system that had received benefits totaling $44.5 million. State law requires
claimants to enroll in the system unless they are receiving benefits to
supplement a temporary lay-off or a decreased work schedule.
4
State law allows for the Secretary to suspend the requirement on an individual basis. Due to the
impact of the COVID-19 pandemic on employment and unemployment activity, the Secretary
took the extraordinary step of suspending this requirement on a global basis.
30
Similar conditions were commented upon in our preceding audit report.
Recommendation 3
a. establish procedures, such as periodic matches to State higher education
institution enrollment records, to identify and follow up on claimants who
are attending school full-time but fail to disclose it (repeat);
b. follow up on all applicants who state they are attending school to
determine whether it impacts eligibility for unemployment benefits
(repeat);
c. verify that all claimants comply with applicable enrollment requirements,
including the Maryland Workforce Exchange system (repeat); and
d. take timely and appropriate corrective action for any potentially
ineligible claimants or benefits identified, including those noted in this
finding (repeat).
Claims Processing
Background
The new BEACON system (and previously MABS) subjects initial claim
applications to certain automated validation rules to help determine eligibility and
benefits due. For example, wages reported by the applicant are automatically
verified to wages reported by the applicable employer. If inconsistencies or other
discrepancies are detected, applications and claims may be suspended from
processing, and require manual review and adjustment by a claims processor. A
manual review is also required for claims designated for adjudication. A claim is
adjudicated when it requires a claims worker to further investigate certain issues
and determine the impact on the claimant’s eligibility, for example whether the
claimant was actually terminated or voluntarily resigned. Manual claims
processing and adjudication are performed by DUI employees at the four DUI
claims centers and by employees of the two aforementioned staffing vendors.
Finding 4
DUI did not have procedures to help prevent and detect duplicate benefit
payments. Our analysis disclosed $43.3 million in potentially duplicate
payments made to 12,500 claimants between April 2020 and December 2021
that were not identified or investigated by DUI.
Analysis
DUI did not have procedures to help prevent and detect duplicate benefit
payments. Our analysis of BEACON records disclosed $43.3 million in
31
potentially duplicate payments made to 12,500 claimants between April 2020 and
December 2021, where we could not determine that DUI identified or investigated
the potentially duplicated payments. Our further analysis of seven of these
payments totaling $22,103, including a review of the related debit card records,
disclosed that six duplicate payments were made totaling $20,350. For example,
one claimant received 19 payments for one benefit week totaling $5,210, and
another received 16 payments for one benefit week totaling $7,616. These
claimants should have received only $374 and $476, respectively, for the
applicable benefit week.5 The remaining payment we reviewed was determined
not to be a duplicate payment.
According to DUI management, duplicate payments occurred primarily because

of deficiencies in BEACON. For example, BEACON accepted multiple weekly
certifications for the same benefit week which resulted in more than one payment
being made, and BEACON allowed duplicate Lost Wage Assistance Program
payments. In addition, claimants were able to improperly file for multiple
programs for the same benefit week, such as PUA and regular unemployment
insurance.
As noted in our testing, certain data in BEACON that appeared to indicate a

duplicate payment was made was incorrect. As a result, we could not readily
determine how much of the aforementioned $43.3 million represented actual
duplicate payments.
Recommendation 4
a. require the BEACON vendor to correct BEACON to prevent duplicate
payments;
b. use available BEACON records to identify duplicate payments; and
c. take appropriate corrective action for the duplicate payments, including
those noted in this finding.
5
A named payee may not have actually received payment, but may have been a victim of identity
theft.
32
Finding 5
DUI did not conduct timely verifications of income reported by applicants
for PUA benefits and did not ensure manual adjustments processed by DUI
and contract employees were proper.
Analysis
DUI did not conduct timely verifications of income reported by applicants for
PUA benefits and did not ensure manual adjustments processed by DUI and
contract employees were proper. During the period from January 27, 2020
through December 31, 2020 federal PUA benefits were available for up to 39
weeks to individuals ineligible for regular benefits (such as self-employed
individuals). Individuals applying for PUA had to self-report their income in
BEACON at the time of application which was used as the basis for claim
payments. Federal regulations required these individuals to submit
documentation to support the reported income within 21 days and for DUI to
“promptly” review and adjust the claim if necessary. As of January 2021, $5.9
billion was paid to claimants under the PUA program in Maryland.
DUI Did Not Conduct Timely Income Verifications

According to BEACON records as of January 31, 2021, DUI had not verified
income for 40,265 of the 198,990 individuals who received PUA benefits between
May and November 2020. In all of these cases, documentation had been
submitted by the applicant and was awaiting review. Although most of these
cases were verified by January 2022, there was a significant delay, which resulted
in certain overpayments. Specifically, our test of 10 of these cases disclosed that
it took between 274 and 288 business days to complete the review despite the
documentation being submitted within 10 days of the initial claim.
The untimely verifications resulted in significant delays in processing adjustments

to the benefits due to discrepancies between the reported income and the support.
For example, the verification of one claimant 281 days after the claim submission
resulted in an adjustment of the weekly benefit from $430 to $278. Due to the
untimely verification, DUI had already overpaid the claimant $6,080. DUI could
not readily provide us with the total amount of overpayments identified as a result
of delayed verifications.
DUI Did Not Conduct Verifications of Manual Adjustments

DUI did not have any process to verify the propriety of manual adjustments made
by DUI and contract employees. As noted above, DUI and contract employees
were responsible for reviewing supporting documentation and adjusting applicant
information on BEACON. According to agency personnel, the BEACON system
does not have the capability to show after the fact (either on a screen or in a
33
generated report) the income amount that was entered by the employee who
verified the income. This is a significant control deficiency since the amount
entered serves as the basis for any future benefit payments, and there was no way
to readily verify the amounts entered resulting in certain errors going undetected.
Specifically, our review of income documentation submitted by 25 claimants

receiving PUA, disclosed 7 claimants who received benefits totaling
approximately $216,000 as of January 2021 that were being overpaid $10 to $169
per week based on the submitted documentation. For example, one claimant was
receiving $430 per week when based on provided income documentation they
should have received $261, an overpayment of $169 per week. After our
inquiries, DUI, with the assistance of the BEACON vendor, determined that the
employee who verified this applicant’s income entered an incorrect income
amount into BEACON.
Recommendation 5
a. ensure that critical applicant data, such as income, is verified and
accurately adjusted if necessary, in a timely manner;
b. ensure the aforementioned system deficiency regarding BEACON’s
inability to show the verified income amount is corrected and establish a
documented process to verify that the recorded information was entered
accurately; and
c. investigate and resolve any differences in weekly PUA benefit amounts
disclosed as a result of income verifications, including those noted above.
Finding 6
DUI did not adequately review regular claims and adjudications processed
by claims center DUI employees and temporary staff, and output reports of
manual wage entries could not be generated from BEACON for verification
purposes.
Analysis
DUI did not conduct all required supervisory reviews of regular claims and
adjudications processed by claims center DUI employees and did not ensure that
claims processed by staffing vendor employees were subject to review. In
addition, output reports of manual wage entries could not be generated from
BEACON for verification purposes.
34
DUI Did Not Ensure Required Reviews were performed at Claim Centers
DUI did not ensure supervisors at the claim centers reviewed manual claims and
adjudications as required and therefore, did not take appropriate action when the
reviews at certain centers were not performed. DUI policy requires supervisors at
the claims centers to review seven claims processed by each claims processer on a
weekly basis, and 60 adjudications each week. These reviews are intended to
verify that critical information supporting the legitimacy of the claim is complete
and properly evaluated and recorded.
DUI did not monitor the claims centers to ensure the reviews were performed, and
as of April 2021 DUI had no plans or process to ensure that the reviews, which
are still required by DUI policy, were conducted in the future at these centers.
We requested the most recent reviews from two of the four DUI claims centers,
and noted that the required reviews were not conducted. For example, one claims
center had not conducted any of the required reviews of adjudications since
December 2016 and had not conducted any of the required reviews of manual
claims since October 2019. DUI was not aware that the reviews were not being
performed at these claims centers and accordingly did not take any corrective
action. We could not readily determine the total number of manual claims and
adjustments processed by DUI because BEACON did not have the ability to
generate a report of these transactions. Similar conditions regarding the lack of
supervisory reviews over manual claims were commented upon in our preceding
audit report.
Staffing Vendor
DUI had no procedure to perform, and the related contract did not require,
supervisory reviews of manual claims processed by staffing vendor employees.
We could not readily determine the total number of manual claims and
adjustments processed by the staffing vendor because BEACON did not have the
ability to generate a report of these transactions that we deemed reliable.
Manual Wage Entries

DUI did not review manual adjustments made by DUI headquarter employees for
certain employer wages, for example, military and federal employees. DUI
headquarter employees were responsible for entering into BEACON income
information based on supporting documentation from certain applicants. Since
the implementation of BEACON in September 2020, DUI has been unable to
generate output reports from BEACON to accurately provide claimant wages
manually recorded. In the absence of output reports, supervisors were provided
lists of manually entered wages by each employee to review. DUI employees
creating the lists were also responsible for initiating claim payments as part of
35
their job duties, and, since they were not independent of the process, the risk of
improper payments increases.
Due to the inability of BEACON to generate output reports, we were unable to

quantify the extent of manual wage entries since the system was implemented.
However, from July 1, 2017 through September 15, 2020, approximately 81,000
wage entries were manually recorded in MABS.
Recommendation 6
a. ensure that supervisors at claim centers perform the required reviews of
claims processed (repeat) and adjudications completed;
b. establish a formal process to provide for supervisory review of claims
processed by temporary staff used to assist DUI’s claim center employees;
and
c. require the BEACON vendor to address the aforementioned system
deficiencies preventing the generation of system output reports of manual
claims and adjustments (including those performed by the staffing
vendor), and use those reports to verify the propriety of those entries.
Finding 7
DUI did not establish sufficient controls over reissued debit cards, and did
not ensure the proper disposition of funds remaining on expired debit cards.
Analysis
DUI did not establish sufficient controls over reissued debit cards, and did not
ensure the proper disposition of funds remaining on expired cards. Prior to May
2021, DUI issued a UI Benefits Debit Card for each approved claimant as a
means to access the claimant’s unemployment insurance benefits. Between July
2017 and January 2021, there were 354,445 debit cards reissued to claimants, and
as of October 2020, the value remaining on expired or never activated debit cards
totaled $23.1 million.
 There was no independent review of reissued debit cards to ensure they were
proper. A new debit card can be issued if the original card is lost or damaged.
The lack of review is significant because the employee responsible for
reissuing cards was also responsible for updating claimant mailing addresses
in the sponsoring bank’s records, and had access to update a claimant’s
address in BEACON. As a result, the employee was in a position to identify
an inactive card, reissue a new card to a different address (such as a PO box),
and misappropriate the funds.
36
According to records we obtained from the sponsoring bank, of the
aforementioned 354,445 reissued debit cards, 1,970 relating to 1,170
claimants were reissued to an address not included in DUI’s unemployment
insurance system. We asked agency personnel for documentation to support
30 of the debit cards that were mailed to addresses that appeared questionable,
such as out of State, or where multiple cards were mailed to the same address.
DUI could not provide us with documentation or adequate explanations for 23
of these reissued cards, including whether an address change was made by
DUI or the sponsoring bank. Benefits paid through these 23 reissued cards
totaled $315,000.
 DUI did not have procedures to ensure the proper disposition of funds
remaining on debit cards that were expired or that were never activated. As
previously noted, there were approximately 30,000 cards that were expired or
never activated with $23.1 million6 remaining in the related accounts at the
sponsoring bank as of October 2020. Since debit cards expire three years after
they are issued, these funds should have been either returned to DUI or
reported to the State Comptroller as unclaimed property.
We were advised by agency personnel that DUI requested the sponsoring

bank to periodically provide documentation regarding never activated and
expired cards, but DUI never used this information for monitoring purposes.
According to the Maryland Unemployment Benefits Debit Card Deposit
Agreement between DUI and the sponsoring bank, if a debit card is not
activated within one year of issuance, the account is to be closed and the funds
returned to DUI. If the debit card is activated, any remaining unclaimed funds
should be reported and remitted by the bank as unclaimed property to
Maryland after a period of three years in accordance with State laws for
unclaimed property.
In May 2021, DUI discontinued use of the debit card, and began issuing benefits
only by direct deposit or by check. However, cards in place at the time with
remaining benefits could still be used, and reissued in the event of loss. Similar
conditions were commented upon in our preceding audit report.
Recommendation 7
We recommend that DUI establish procedures to ensure
a. all reissued debit cards are subject to an independent review and
approval (repeat);
6
A debit card may be not-activated if DUI canceled or froze the card due to potential fraud while
determining the legitimacy of the claim. For example, the $23.1 million includes 59 canceled
debit cards, which are part of the population of cards in Finding 8.
37
b. cards reissued to a questionable address are adequately investigated and
resolved, including the 1,970 noted above; and
c. unspent funds remaining on debit cards are returned to DUI or reported
to the State Comptroller as unclaimed property in accordance with the
aforementioned Agreement (repeat).
Finding 8
DUI did not properly account for potentially fraudulent benefits totaling
$493.9 million that were removed from claimants’ debit cards.
Analysis
DUI did not properly account for potentially fraudulent benefits totaling $493.9
million that were removed from claimants’ UI Benefit Debit Cards (debit cards).
In July 2020, DUI canceled debit cards for 46,986 claimants with benefits totaling
$493.9 million because the claims originated from out of State, and accordingly,
were considered potentially fraudulent. By canceling the cards, DUI stopped
those benefits from being drawn by those claimants. DUI instructed these
claimants to provide documentation to support their identity and the validity of
their claim to have the claim reprocessed and repaid.
Our review disclosed that DUI did not update BEACON to reflect the cancelation
of these payments. As a result, claimants who did not submit the requested
documentation received overpayment notices even though they never received the
funds. In addition, DUI was unable to provide documentation of how much, if
any, of the $493.9 million was subsequently repaid to claimants.
We were advised by DUI management that the United States Department of

Labor - Office of Inspector General is assisting DUI in investigating the
potentially fraudulent claims. As of August 2022, DUI could not provide details
regarding the investigation, the funds remain in DUI’s disbursement account, and
no decision has been made on the disposition of these funds.
Recommendation 8
We recommend that DUI ensure that all transactions impacting claimant
accounts are properly recorded in BEACON, including those noted in this
finding.
38
Finding 9
DUI did not ensure amounts disbursed from the Unemployment Insurance
Trust Fund were properly transferred to the bank account used to make
benefit payments.
Analysis
DUI did not ensure amounts disbursed from the Unemployment Insurance Trust
Fund were properly transferred to the bank account used to make benefit
payments. Specifically, as of May 2021, DUI had not performed a reconciliation
of its record of Trust Fund activity to the corresponding bank records since
August 2020. Disbursements from the Trust Fund totaled approximately $5.0
billion during the period from October 2020 through May 2021 and are generally
made daily.
In response to our inquiries, in May 2021 DUI prepared the reconciliation for
September 2020, which showed unresolved reconciling items totaling
approximately $81.4 million dating back to 2019. As of January 2022, DUI had
been unable to resolve these differences, and no additional reconciliations had
been performed.
Recommendation 9
We recommend that DUI prepare periodic reconciliations of its record of
Trust Fund activity to the corresponding bank records, and resolve
differences timely, including those noted in this finding.
We determined that Findings 10 through 13 related to “cybersecurity”, as defined

by the State Finance and Procurement Article, Section 3A-301(b) of the
Annotated Code of Maryland, and therefore are subject to redaction from the
publicly available audit report in accordance with the State Government Article 2-
1224(i). Consequently, the specifics of the following findings, including the
analysis, related recommendation(s), along with the Maryland Department of
Labor’s responses, have been redacted from this report copy.
39
Finding 10
Redacted cybersecurity-related finding.
Finding 11
Finding 12
Finding 13
40
Audit Scope, Objectives, and Methodology
We have conducted two parts of a fiscal compliance audit of the Maryland
Department of Labor (MDL) – Division of Unemployment Insurance (DUI) for
the period beginning April 17, 2017 and ending November 15, 2020. The audit
was conducted 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.
As prescribed by the State Government Article, Section 2-1221 of the Annotated

Code of Maryland, the objectives of this audit were to examine DUI’s financial
transactions, records, and internal control, and to evaluate its compliance with
applicable State laws, rules, and regulations.
In planning and conducting our audit, we focused on the major financial-related

areas of operations based on assessments of significance and risk. The areas
addressed by the audit included benefit payments and related system
implementation. We also determined the status of four of the six findings
contained in our preceding audit report.
Our audit did not include certain support services provided to DUI by MDL –
Office of the Secretary. These support services (such as payroll, purchasing,
maintenance of accounting records, and related fiscal functions) are included
within the scope of our audits of MDL – Office of the Secretary. In addition, our
audit did not include an evaluation of internal controls over compliance with
federal laws and regulations for federal financial assistance programs and an
assessment of DUI’s compliance with those laws and regulations because the
State of Maryland engages an independent accounting firm to annually audit such
programs administered by State agencies, including DUI.
Our assessment of internal controls was based on agency procedures and controls
in place at the time of our fieldwork. Our tests of transactions and other auditing
procedures were generally focused on the transactions occurring during our audit
period of April 17, 2017 to November 15, 2020, but may include transactions
before or after this period as we considered necessary to achieve our audit
objectives.
To accomplish our audit objectives, our audit procedures included inquiries of

appropriate personnel, inspections of documents and records, tests of transactions,
41
and to the extent practicable, observations of DUI’s operations. Generally,
transactions were selected for testing based on auditor judgment, which primarily
considers risk, the timing or dollar amount of the transaction, or the significance
of the transaction to the area of operation reviewed. As a matter of course, we do
not normally use sampling in our tests, so unless otherwise specifically indicated,
neither statistical nor non-statistical audit sampling was used to select the
transactions tested. Therefore, unless sampling is specifically indicated in a
finding, the results from any tests conducted or disclosed by us cannot be used to
project those results to the entire population from which the test items were
selected.
We also performed various data extracts of pertinent information from the State’s
Financial Management Information System (such as revenue and expenditure
data). The extracts are performed as part of ongoing internal processes
established by the Office of Legislative Audits and were subject to various tests to
determine data reliability. We determined that the data extracted from this source
were sufficiently reliable for the purposes the data were used during this audit.
We also extracted data from the Maryland Automated Benefits System, the
Maryland Unemployment Insurance Tax System, and BEACON, as well as from
certain other State records, such as those maintained by the Maryland Department
of Health, for the purpose of testing unemployment tax payments and
reimbursements related to benefit claims and payments. We performed various
tests of the relevant data and determined the data were sufficiently reliable for the
purposes the data were used during the audit. Finally, we performed other
auditing procedures that we considered necessary to achieve our audit objectives.
The reliability of data used in this report for background or informational
purposes was not assessed.
DUI’s management is responsible for establishing and maintaining effective

internal control. Internal control is a process designed to provide reasonable
assurance that objectives pertaining to the reliability of financial records;
effectiveness and efficiency of operations, including safeguarding of assets; and
compliance with applicable laws, rules, and regulations are achieved. As
provided for in Government Auditing Standards, there are five components of
internal control: control environment, risk assessment, control activities,
information and communication, and monitoring. Each of the five components,
when significant to the audit objectives, and as applicable to DUI, were
considered by us during the course of this audit.
Because of inherent limitations in internal control, errors or fraud may

nevertheless occur and not be detected. Also, projections of any evaluation of
42
internal control to future periods are subject to the risk that conditions may
change or compliance with policies and procedures may deteriorate.
Our reports are designed to assist the Maryland General Assembly in exercising
its legislative oversight function and to provide constructive recommendations for
improving State operations. As a result, our reports generally do not address
activities we reviewed that are functioning properly.
This report includes findings relating to conditions that we consider to be

significant deficiencies in the design or operation of internal control that could
adversely affect DUI’s ability to maintain reliable financial records, operate
effectively and efficiently, and/or comply with applicable laws, rules, and
regulations. Our audit also disclosed significant instances of noncompliance with
applicable laws, rules, or regulations. Other less significant findings were
communicated to DUI that did not warrant inclusion in this report.
State Government Article Section 2-1224(i) requires that we redact in a manner

consistent with auditing best practices any cybersecurity findings before a report
is made available to the public. This results in the issuance of two different
versions of an audit report that contains cybersecurity findings – a redacted
version for the public and an unredacted version for government officials
responsible for acting on our audit recommendations.
The State Finance and Procurement Article, Section 3A-301(b), states that
cybersecurity is defined as “processes or capabilities wherein systems,
communications, and information are protected and defended against damage,
unauthorized use or modification, and exploitation”. Based on that definition, and
in our professional judgment, we concluded that certain findings in this report fall
under that definition. Consequently, for the publicly available audit report all
specifics as to the nature of cybersecurity findings and required corrective actions
have been redacted. We have determined that such aforementioned practices, and
government auditing standards, support the redaction of this information from the
public audit report. The specifics of these cybersecurity findings have been
communicated to DUI and those parties responsible for acting on our
recommendations in an unredacted audit report.
As a result of our audit, we determined that DUI’s accountability and compliance

level was unsatisfactory. The primary factors contributing to the unsatisfactory
rating were the significance of our audit findings and how the findings were
pervasive across all areas of DUI’s operations. Our rating conclusion has been
made solely pursuant to State law and rating guidelines approved by the Joint
43
Audit Committee. The rating process is not a practice prescribed by professional
auditing standards.
The response from MDL, on behalf of DUI, to our findings and recommendations
is included as an appendix to this report. Depending on the version of the audit
report, responses to any cybersecurity findings may be redacted in accordance
with State law. As prescribed in the State Government Article, Section 2-1224 of
the Annotated Code of Maryland, we will advise MDL regarding the results of
our review of its response.
44
Exhibit 1
Summary of Unemployment Insurance Modernization (BEACON) Findings in
OLA Audit Reports Issued from July 1, 2015 to May 31, 2022
Maryland Department of Information Technology (DoIT)

Report Issued May 1, 2020
Finding 1
 Information Technology Project Requests (ITPRs) – DoIT did not have a documented
review and approval of the annual BEACON ITPR, which our testing disclosed had not been
updated from the preceding year’s ITPR.
 Monthly Project Monitoring - DoIT did not require oversight project managers hired by a
DoIT vendor, to document their review and verification of the accuracy of information
provided in monthly project monitoring reports provided by the agencies. The review of these
monthly monitoring reports is critical to monitoring project status including scope, schedule,
cost, and risks. Our review of applicable reports discussed during fiscal year 2018 steering
committee meetings disclosed that the actions to be taken to address BEACON identified
risks, such as project delays, were not always included in the reports, and DoIT did not
document that methods to address these risks were discussed in the related meetings.
 DoIT Annual Major Information Technology Development Project Report – DoIT did not
properly report total estimated project costs for BEACON, which we determined were
significantly underestimated.
Maryland Department of Labor (MDL)

Report Issued January 7, 2022
Finding 1
 MDL did not obtain documentation to support $11.7 million in vendor billings for
modernizing DUI’s unemployment insurance system. MDL approved these costs, which were
essentially for the remainder of the contract, but could not provide documentation verifying the
propriety of the amounts invoiced and paid.

Part 1 Unemployment Insurance Tax Contributions
Report Issued May 4, 2022
Finding 2
 DUI had not verified that unemployment tax collections were properly deposited and recorded
since the implementation of BEACON in September 2020, due to the inability to generate
certain required reports from the BEACON system
45
Exhibit 1
Summary of Unemployment Insurance Modernization (BEACON) Findings in
OLA Audit Reports Issued from July 1, 2015 to May 31, 2022
Finding 3
 DUI did not regularly conduct data matches to identify employers who had not registered with
DUI, as required by State law, and did not always follow up on the results of the matches that
were performed. According to DUI management, DUI did not follow up on match results
because BEACON was unable to generate notices to employers that would alert them of the
legal requirement to register.
Finding 4
 DUI did not ensure reimbursable employers provided sufficient collateral to protect the State
in the event claims are paid on their behalf. As of September 2021, BEACON was not able to
display or generate reports of amounts due.
Finding 5
 DUI did not have formal policies for pursuing collection of delinquent employer accounts, and
discontinued pursing delinquent accounts in September 2020 due to BEACON system
deficiencies.
Finding 6
 Access to process critical employer tax related transactions and functions within BEACON
was not adequately restricted.
46
Exhibit 2
Selected Sources for National Trends
1. Testimony of the US Department of Labor (DOL) – Office of the

Inspector General (OIG) to US Senate Committee on Homeland Security
and Governmental Affairs (March 17, 2022)
2. Pandemic Response Accountability Committee (PRAC) “Key Insights:
State Pandemic Unemployment Insurance Programs” (December 16,
2021) (This report contains links to US DOL – OIG publications and
reports from sixteen State oversight agencies, which we also reviewed.)
3. “Best Practices and Lessons Learned from the Administration of
Pandemic-Related Unemployment Benefits Programs” (January 31, 2022),
a report produced by the MITRE Corporation requested by PRAC
4. US DOL – OIG Advisory Report: “CARES Act: Initial Areas of Concern
Regarding Implementation of Unemployment Insurance Provisions”
(April 21, 2020)
5. US DOL – OIG Alert Memorandum: “The Employment and Training
Administration Needs to Ensure State Workforce Agencies Implement
Effective Unemployment Insurance Program Fraud Controls for High Risk
Areas” (February 22, 2021)
6. US DOL – OIG Alert Memorandum: “The Employment and Training
Administration Needs to Issue Guidance to Ensure State Workforce
Agencies Provide Requested Unemployment Insurance Data to the Office
of Inspector General” (June 16, 2021)
7. US DOL – OIG Report: “COVID-19: States Struggled to Implement
CARES Act Unemployment Insurance Programs” (May 28, 2021)
47
APPENDIX
Part 2
Agency Response Form
Benefit Payments
Finding 1
DUI did not conduct certain critical matches used to identify potentially fraudulent or
improper claims. We conducted three matches to replicate four of the discontinued DUI
matches and identified at least $32.3 million in potentially improper payments.

a. require the BEACON vendor to address any system deficiencies preventing completion
of matches, and in the future ensure that all matches are performed;
b. conduct the aforementioned matches for the aforementioned periods when matches
were not performed or did not include all claims; and
c. investigate and resolve any potentially improper payments identified by the matches,
including those noted in this finding.
Agency Response
Analysis
Please provide State Employee Crossmatch contractor provided DUI with a State
additional comments as Employees Crossmatch. DUI performed those matches weekly. In
deemed necessary. performing the matches, DUI concluded that although the State agency
may have reported certain employees as “still employed”, that was not
always correct. The employee may have been separated due to a lack of
work but remained on the agency’s payroll, therefore technically
unemployed and eligible for unemployment benefits.
Recommendation 1a Agree Estimated Completion Date: 07/01/2023

Please provide details of DUI will require the vendor to address any system deficiencies
corrective action or preventing completion of matches, and in the future ensure that all
explain disagreement. matches are performed as specified in crossmatch business rules. Prior to
the finding, DUI had addressed many crossmatch issues through written
Problem Incident Reports (PIR’s) with the vendor. Although most have
been corrected, DUI will continue to work with the vendor to test and
verify that crossmatches are working as designed.
Recommendation 1b Agree Estimated Completion Date: 07/01/2023
Please provide details of DUI has started and will continue to devote staff to conduct the
corrective action or aforementioned matches for the periods where matches were not
explain disagreement. performed or did not include all claims.
Page 1 of 12
Part 2
Recommendation 1c Agree Estimated Completion Date: 07/01/2023

Please provide details of DUI staff has begun and will continue to investigate and fully resolve
corrective action or any improper payments identified by the matches, including those noted
explain disagreement. in this finding. DUI is taking prompt, corrective actions including
setting up overpayments.
Finding 2
DUI did not have comprehensive procedures to ensure that individuals filing claims using a
foreign Internet Protocol (IP) address were eligible to receive benefits, including 3,724
claimants that received benefit payments totaling $3.6 million.

a. continue to develop its automated controls used to block foreign IP addresses to ensure
they are sufficiently comprehensive,
b. retain a record of foreign IP addresses used and formally re-evaluate the decision to
cease the use of a geo-mapping service to aid in the identification and investigation of
foreign IP addresses that are used for both initial applications and weekly certifications,
and
c. investigate the foreign IP addresses identified in this finding and take corrective action
for any ineligible claimants and benefits identified (repeat).
Agency Response
Analysis
Please provide These findings were on the Legacy system MABS. Since implementing
additional comments as the new BEACON system, the Foreign IPs are monitored and blocked.
deemed necessary. Since the foreign IPs are blocked altogether. DUI is investigating to
implement an alternative software product which is more powerful and
has better Foreign IP and geolocation blocking features by the end of
this year, 07/01/2023.

Please provide details of The modernized system has checks and balances in place to prevent this
corrective action or and the tools used in the web hosting provider including the software
explain disagreement. products along with DUI and vendor procedures are constantly being
assessed and updated to block foreign IPs.
Page 2 of 12
Part 2

Please provide details of The BEACON system does not allow claims filed with foreign IP
corrective action or addresses. However, all IP addresses are stored in an Audit log
explain disagreement. irrespective of the IPS being foreign or domestic

Please provide details of DUI will investigate the foreign IP address and claimants related to those
corrective action or addresses. DUI will take the proper corrective action against those not
explain disagreement. eligible for benefits.
Finding 3
DUI did not ensure claimants who were full-time students were eligible for benefits, and
that all claimants were enrolled in the Maryland Workforce Exchange System, as required.

a. establish procedures, such as periodic matches to State higher education institution
enrollment records, to identify and follow up on claimants who are attending school
full-time but fail to disclose it (repeat);
b. follow up on all applicants who state they are attending school to determine whether it
impacts eligibility for unemployment benefits (repeat);
c. verify that all claimants comply with applicable enrollment requirements, including the
Maryland Workforce Exchange system (repeat); and
d. take timely and appropriate corrective action for any potentially ineligible claimants or
benefits identified, including those noted in this finding (repeat).
Agency Response
Analysis
Please provide DUI’s weekly claim certification asks claimants, for each week of
additional comments as unemployment that is requested, if they are a full-time student. If the
deemed necessary. claimant answers “yes.”, an issue is created to adjudicate. However,
during the pandemic, Maryland was faced with an historic volume of
claims. The Social Security Act requires States to administer the
program in such a way that is reasonably calculated to ensure full
payment of unemployment benefits at the earliest state of unemployment
that is administratively feasible. States must balance the dual concerns of
promptness and accuracy. During the pandemic, DUI prioritized
adjudication issues based on the need to strike this balance.
Page 3 of 12
Part 2

Please provide details of DUI understands the importance of this data match and is establishing
corrective action or procedures to incorporate enrollment data from state universities to
explain disagreement. identify claimants who are attending school full time but fail to disclose
it.
Please provide details of
corrective action or DUI will follow up on all applicants who state they are attending school
explain disagreement. to determine whether it impacts eligibility for unemployment benefits.

Please provide details of DUI’s IT system automatically creates a registration for claimants in the
corrective action or Maryland Workforce Exchange (MWE) when they file an initial claim
explain disagreement. for unemployment insurance benefits. DUI will work with DWDAL to
share information when claimants fail to fully complete their
registration. DUI will investigate and, if applicable, adjudicate whether
claimants have complied with enrollment requirements. In addition, DUI
will add a question about MWE enrollment on the weekly certification.
Currently, DUI relies on audits of random claims to verify whether
claimants have complied with applicable enrollment requirements.
Recommendation 3d Agree Estimated Completion Date: 07/01/2023
Please provide details of DUI will investigate the findings and take the proper corrective action
corrective action or against those claimants who are not eligible for benefits, including those
explain disagreement. noted in the finding.
Page 4 of 12
Part 2
Claims Processing
Finding 4
DUI did not have procedures to help prevent and detect duplicate benefit payments. Our
analysis disclosed $43.3 million in potentially duplicate payments made to 12,500 claimants
between April 2020 and December 2021 that were not identified or investigated by DUI.

a. require the BEACON vendor to correct BEACON to prevent duplicate payments;
b. use available BEACON records to identify duplicate payments; and
c. take appropriate corrective action for the duplicate payments, including those noted in
this finding.
Agency Response
Analysis
Please provide
additional comments as While DUI was unable to get the same query used by the auditors for
deemed necessary. this finding, we concur that it is factually accurate.

Please provide details of The Beacon system had some defects that were fixed as soon as they
corrective action or were identified. The backup for the correcting PIRs has been submitted
explain disagreement. to the auditors.

Please provide details of The system had some defects that were fixed as soon as they were
corrective action or identified. Overpayments and offsets were recorded in the system for the
explain disagreement. population of claimants identified

Please provide details of The system had some defects that were fixed as soon as they were
corrective action or identified. Overpayments and offsets were recorded in the system for the
explain disagreement. population of claimants identified. DUI was not able to replicate 12,500
duplicate payments as noted in the audit.
Page 5 of 12
Part 2
Finding 5
DUI did not conduct timely verifications of income reported by applicants for PUA benefits
and did not ensure manual adjustments processed by DUI and contract employees were
proper.

a. ensure that critical applicant data, such as income, is verified and accurately adjusted if
necessary, in a timely manner;
b. ensure the aforementioned system deficiency regarding BEACON’s inability to show
the verified income amount is corrected and establish a documented process to verify
that the recorded information was entered accurately; and
c. investigate and resolve any differences in weekly PUA benefit amounts disclosed as a
result of income verifications, including those noted above.
Agency Response
Analysis
Please provide The verifications were not conducted in a timely manner due to the
additional comments as backlog of claims and claims-related work due to the pandemic. We are
deemed necessary. actively working to correct any weekly benefit amount (WBA)
miscalculations.
Please provide details of This was an issue due to the pandemic. To receive PUA benefits of
corrective action or greater than the minimum WBA, a claimant had to provide proof of
explain disagreement. income if they were self-employed or an independent contractor.
Traditionally, self-employed or independent contractors do not qualify
for Unemployment Insurance. Due to the number of PUA claims filed in
2020 and 2021, there was a tremendous backlog of PUA Proof of
Income work items. Under normal circumstances this is not an issue as
the weekly benefit amount is based on employer reported wages, not in
the manner that was used to calculate PUA weekly benefit amount.
Please provide details of This defect is specific to the PUA Proof of Income work item. Any
corrective action or change to wages regarding a UI claim is documented in notes and/or
explain disagreement. account activity in Beacon, allowing for review. DUI implemented a
manual solution to document the wages in the claimant’s portal and
verify that the weekly benefit amount is calculated accurately. If
Page 6 of 12
Part 2
USDOL implements a program that is similar to PUA, DUI will

reprogram its IT system to correct this defect.

Please provide details of This effort is ongoing. There was a short time that USDOL allowed the
corrective action or WBA to be based on the claimant's estimated earnings during the base
explain disagreement. period, not on any actual proof. This caused many overpayments as,
once we were able to verify their earnings, we found that claimants had
overestimated their earnings for the PUA base period. DUI will continue
to review PUA Proof of Income work items for accuracy and will correct
any errors that are discovered.
Finding 6
DUI did not adequately review regular claims and adjudications processed by claims center
DUI employees and temporary staff, and output reports of manual wage entries could not
be generated from BEACON for verification purposes.

a. ensure that supervisors at claim centers perform the required reviews of claims
processed (repeat) and adjudications completed;
b. establish a formal process to provide for supervisory review of claims processed by
temporary staff used to assist DUI’s claim center employees; and
c. require the BEACON vendor to address the aforementioned system deficiencies
preventing the generation of system output reports of manual claims and adjustments
(including those performed by the staffing vendor), and use those reports to verify the
propriety of those entries.
Agency Response
Analysis
Please provide
additional comments as While factually accurate, regular review of claims and adjudication was
deemed necessary. placed on hold due to the historic workload brought on by the pandemic
Page 7 of 12
Part 2

Please provide details of DUI is reinstating the regular Supervisory review of UI claims and
corrective action or adjudication. Supervisors have begun reviewing work to ensure quality
explain disagreement. and consistency.

Please provide details of In addition to reinstating Supervisory review for claim center staff, DUI
corrective action or is working to establish a Quality Assurance (QA) Manager position. This
explain disagreement. position will review the entirety of a claim from the initial claims
application to completion of work items and adjudication issues, and the
timely and accurate payment of benefits. The QA Manager and their
team will also listen to calls to ensure excellent customer service. This
team will review not only Maryland staff, but also relevant vendor staff.
Please provide details of DUI will work with the vendor to ensure there is a report of manual
corrective action or wage entries. DUI will then verify the validity of those entries with a
explain disagreement. formalized review process.
Finding 7
DUI did not establish sufficient controls over reissued debit cards, and did not ensure the
proper disposition of funds remaining on expired debit cards.
We recommend that DUI establish procedures to ensure

a. all reissued debit cards are subject to an independent review and approval (repeat);
b. cards reissued to a questionable address are adequately investigated and resolved,
including the 1,970 noted above; and
c. unspent funds remaining on debit cards are returned to DUI or reported to the State
Comptroller as unclaimed property in accordance with the aforementioned Agreement
(repeat).
Agency Response
Analysis
Please provide
additional comments as
deemed necessary.
Page 8 of 12
Part 2

Please provide details of On Friday, May 21, 2021, DUI stopped issuing new debit cards and
corrective action or loading benefit payments onto debit cards. If claimants have a balance
explain disagreement. on their debit card, they must work directly with the debit card vendor to
access those funds in a different way. Also, the deadline to order a
replacement debit card from the debit card vendor was January 5, 2022.
Going forward, if a claimant changes their address in DUI’s IT system,
they must complete an automated identity verification process.
Please provide details of DUI continues to investigate all cases of potential fraud. DUI will
corrective action or review the 1,970 claims noted in the finding and resolve any issues. In
explain disagreement. many instances, our debit card vendor changed the addresses to which
debit cards were sent without informing DUI. Going forward, if a
claimant changes their address in DUI’s IT system, they must complete
an automated identity verification process.
Please provide details of DUI is working with legal counsel to make sure that the Maryland
corrective action or Unemployment Benefits Debit Card Deposit Agreement is enforced with
explain disagreement. appropriate funds returned to DUI or sent to the State Comptroller as
unclaimed property.
Page 9 of 12
Part 2
Finding 8
DUI did not properly account for potentially fraudulent benefits totaling $493.9 million
that were removed from claimants’ debit cards.
We recommend that DUI ensure that all transactions impacting claimant accounts are
properly recorded in BEACON, including those noted in this finding.
Agency Response
Analysis
Please provide
deemed necessary.
Recommendation 8 Agree Estimated Completion Date: 07/01/2023

Please provide details of The agency is in negotiations with the debit card vendor to finalize the
corrective action or contract exit agreement. Once the issue is settled legally and our debit
explain disagreement. card vendor provides the relevant data, DUI will update the status in
BEACON.
Page 10 of 12
Part 2
Finding 9
DUI did not ensure amounts disbursed from the Unemployment Insurance Trust Fund
were properly transferred to the bank account used to make benefit payments.
We recommend that DUI prepare periodic reconciliations of its record of Trust Fund
activity to the corresponding bank records, and resolve differences timely, including those
noted in this finding.
Agency Response
Analysis
Please provide
deemed necessary.
Recommendation 9 Agree Estimated Completion Date: 07/01/2023

Please provide details of DUI acknowledges that a backlog does exist and that a full reconciliation
corrective action or will be completed. DUI expects to address the backlog and bring
explain disagreement. reconciliation up to date by October 1, 2023, or earlier. Once the backlog
is resolved, DUI will continue to conduct reconciliations monthly and
resolve differences in a timely fashion.
Page 11 of 12
Part 2
The Office of Legislative Audits (OLA) has determined that findings 10 through 13 related to
“cybersecurity”, as defined by the State Finance and Procurements Article, Section 3A-301(b) of
the Annotated Code of Maryland, and therefore are subject to redaction from the publicly
available audit report in accordance with State Government Article 2-1224(i). Although the
specifics of these findings including the analysis, related recommendations, along with MDL’s
responses, have been redacted from this report copy, MDL’s response indicated agreement with
these findings and related recommendations.
Finding 10
Agency Response has been redacted by OLA.
Finding 11
Finding 12
Finding 13
Page 12 of 12
AUDIT TEAM
Michael J. Murdzak, CPA

Audit Manager
R. Brendan Coffey, CPA, CISA

Edwin L. Paul, CPA, CISA
Information Systems Audit Managers
Lauren E. Franchak, CPA

Senior Auditor
Michael K. Bliss
Matthew D. Walbert, CISA
Information Systems Senior Auditors
Monisha A. Barnes
Thea A. Chimento, CFE
Mariyum Gill
Owen M. Long-Grant
Staff Auditors
Dominick R. Abril
Charles O. Price
Information Systems Staff Auditors
David R. Fahnestock, CPA

Data Analytics Manager
Charles H. Hinds IV, CPA

Data Analytics Senior Auditor

Audit Report: Maryland Department of Labor Division of Unemployment

Uploaded by

Copyright:

Available Formats

Audit Report: Maryland Department of Labor Division of Unemployment

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Audit Report: Maryland Department of Labor Division of Unemployment

Uploaded by

Copyright:

Available Formats

Audit Report

Maryland Department of Labor

In compliance with the requirements of the State Government Article Section

OFFICE OF LEGISLATIVE AUDITS

To Obtain Further Information

Ladies and Gentlemen:

We have conducted a fiscal compliance audit of the Maryland Department of

We determined that DUI’s accountability and compliance level was

MDL’s response to this audit, on behalf of DUI, is included as an appendix to this

We wish to acknowledge the cooperation extended to us during the course of this

Gregory A. Hook, CPA

Findings and Recommendations 24

* Finding 2 – DUI did not have comprehensive procedures to ensure 27

* Finding 3 – DUI did not ensure claimants who were full-time 30

 Denotes item repeated in full or part from preceding audit report

Finding 5 – DUI did not conduct timely verifications of income reported 33

* Finding 6 – DUI did not adequately review regular claims and 34

* Finding 7 – DUI did not establish sufficient controls over reissued 36

Finding 8 – DUI did not properly account for potentially fraudulent 38

Unemployment Insurance Trust Fund

Information Systems Security and Control

Finding 11 – Redacted cybersecurity-related finding 40

Finding 12 – Redacted cybersecurity-related finding 40

Finding 13 – Redacted cybersecurity-related finding 40

Audit Scope, Objectives, and Methodology 41

 Denotes item repeated in full or part from preceding audit report

Exhibit 2 – Selected Sources for National Trends 47

Agency Response Appendix

The Division of Unemployment Insurance (DUI) is a separate budgetary unit

 Collecting unemployment insurance tax contributions from employers

Part 1 – Unemployment Insurance Tax Contributions

Part 2 – Unemployment Benefits

Maryland employers are required to remit Maryland unemployment insurance

Historically, employer contributions were the primary source of unemployment

As noted above, the COVID-19 pandemic had a significant impact on

Coronavirus Aid, Relief, and Economic Security Act

On March 27, 2020, in response to the COVID-19 pandemic, the federal

 Pandemic Unemployment Assistance (PUA), effective January 27, 2020

Other federal funding was awarded in addition, and subsequent, to CARES.

 Lost Wage Assistance (LWA) Program (provided by Presidential

Unemployment Insurance System

In 2011, DUI began a major information technology project to replace and

 Maryland Unemployment Insurance Tax System used to process

Our May 1, 2020 report on the Maryland Department of Information Technology

As noted in Figure 3, beginning in April 2020 through September 2020, all

Maryland Unemployment Insurance Trust Fund

Claims Center Staffing

As noted in Figure 5 on the following page, the vendor provided an increasing

Adjudication Staffing Contract

Individuals may file an application for unemployment insurance benefits online or

The application includes information about their previous employer, wages

As a further condition of eligibility, DUI requires claimants to register in the

National Increase in Claim Volume Observed

National Increase in Improper Payments Estimated

In its testimony to Congress, US DOL – OIG cited the Pandemic Unemployment

Examples of Findings from National and State Reports

 States that had otherwise adequate existing pre-pandemic prevention and

Certain Pandemic-related Issues Were Determined Not to be New or

Status of Findings From Preceding Audit Report