oversight

The City of New York, NY, Implemented Policies That Did Not Always Ensure That Community Development Block Grant Disaster Recovery Funds Were Disbursed in Accordance With Its Action Plan and Federal Regulations

Published by the Department of Housing and Urban Development, Office of Inspector General on 2016-11-02.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

      The City of New York, NY,
   Mayor’s Office of Housing Recovery
               Operations
        Community Development Block Grant Disaster
        Recovery-Funded Build It Back Single Family
                        Program




Office of Audit, Region 2      Audit Report Number: 2017-NY-1001
New York - New Jersey                           November 2, 2016
To:            Stanley A. Gimont
               Deputy Assistant Secretary for Grant Programs (Acting), DG

               Michelle Miller
               Deputy Director, Office of Healthy Homes and Lead Hazard Control, LM


                //SIGNED//
From:          Kimberly Greene
               Regional Inspector General for Audit, 2AGA
Subject:       The City of New York, NY, Implemented Policies That Did Not Always Ensure
               That Community Development Block Grant Disaster Recovery Funds Were
               Disbursed in Accordance With Its Action Plan and Federal Requirements


Attached is the U.S. Department of Housing and Urban Development (HUD), Office of Inspector
General’s (OIG) final results of our review of the City of New York, Mayor’s Office of Housing
Recovery Operations’ administration of Community Development Block Grant Disaster
Recovery funds for its Build it Back Single Family Program.
HUD Handbook 2000.06, REV-4, sets specific timeframes for management decisions on
recommended corrective actions. For each recommendation without a management decision,
please respond and provide status reports in accordance with the HUD Handbook. Please furnish
us copies of any correspondence or directives issued because of the audit.
The Inspector General Act, Title 5 United States Code, section 8M, requires that OIG post its
publicly available reports on the OIG Web site. Accordingly, this report will be posted at
http://www.hudoig.gov.
If you have any questions or comments about this report, please do not hesitate to call me at
212-264-4174.
                   Audit Report Number: 2017-NY-1001
                   Date: November 2, 2016

                   The City of New York, NY, Implemented Policies That Did Not Always
                   Ensure That Community Development Block Grant Disaster Recovery Funds
                   Were Disbursed in Accordance With Its Action Plan and Federal
                   Requirements


Highlights

What We Audited and Why
We audited the City of New York, Mayor’s Office of Housing Recovery Operations’
administration of the Build it Back Single Family Program, funded with Community
Development Block Grant Disaster Recovery (CDBG-DR) funds provided by the U.S.
Department of Housing and Urban Development (HUD) to assist in the disaster recovery and
rebuilding efforts resulting from Hurricane Sandy. The objective of the audit was to determine
whether City officials had adequate controls to ensure that the use of CDBG-DR funds was
consistent with the Build it Back Single Family Program guidelines established under the HUD-
approved action plan.

What We Found
City officials implemented policies that did not always ensure that CDBG-DR funds were
disbursed in accordance with the action plan and Federal requirements. Specifically the policies
implemented did not ensure that all eligible homeowners were reimbursed in accordance with the
action plan and the Program and CDBG-DR-assisted homes complied with HUD’s Lead Safe
Housing Rule requirements. In addition, City officials did not maintain complete and accurate
Program files and records. These deficiencies resulted from City officials’ decision to provide
additional CDBG-DR assistance to homeowners with Small Business Administration (SBA)
loans and reduce operational costs. Further, City officials believed that the Program would meet
HUD’s Lead Safe Housing Rule requirements after reimbursement and wished to avoid
recapturing grants from homeowners who did not allow the Program to complete the lead hazard
work. Additionally, City officials did not establish adequate monitoring controls to ensure that
revised record-keeping procedures were consistently followed. As a result, they could not assure
HUD that the use of CDBG-DR funds benefited eligible homeowners in a fair and equitable
manner, assisted homes were lead safe, and Program records were properly maintained.

What We Recommend
We recommend that HUD instruct City officials to (1) submit an amended action plan for
approval to ensure that it agrees with the City’s policies regarding the use of $4.5 million and
planned use of $1.3 million in additional CDBG-DR assistance to homeowners with SBA loans
and $32,107 in assistance above the Program’s 60 percent reimbursement rate, (2) reimburse
$101,398 in additional grants owed to 11 homeowners, and (3) provide documentation to support
that more than $1 million in CDBG-DR funds was disbursed for lead-safe homes.
Table of Contents
Background and Objective......................................................................................3

Results of Audit ........................................................................................................5
         Finding 1: City Officials Did Not Always Implement Policies and Procedures To
         Ensure That Funds Were Disbursed in Accordance With the Action Plan ................ 5

         Finding 2: City Officials Did Not Always Comply With HUD’s Lead Safe Housing
         Rule Requirements.......................................................................................................... 12

         Finding 3: City Officials Did Not Maintain Complete and Accurate Program Files
         and Records ..................................................................................................................... 15

Scope and Methodology .........................................................................................22

Internal Controls ....................................................................................................25

Appendixes ..............................................................................................................27
         A. Schedule of Questioned Costs and Funds To Be Put to Better Use ...................... 27

         B. Auditee Comments and OIG’s Evaluation ............................................................. 29




                                                                    2
Background and Objectives
On November 13, 2012, approximately 2 weeks after Hurricane Sandy made landfall, damaging and
destroying properties and disrupting the lives of New York City residents by displacing them from
their homes, the City of New York, Mayor’s Office of Housing Recovery Operations, was created.
The purpose of the Office was to coordinate efforts among the various City agencies and State and
Federal Governments for developing long- and short-term housing solutions for these displaced
residents.
On January 29, 2013, Congress enacted the Disaster Relief Appropriations Act of 2013, Public Law
113-2, appropriating $16 billion in Community Development Block Grant Disaster Recovery
(CDBG-DR) funds. The purpose of this funding was to cover necessary expenses occurring from
calendar years 2011 through 2013 that were related to long-term recovery, infrastructure and
housing restoration, and housing and economic revitalization for locations that were declared
disaster areas in accordance with the Robert T. Stafford Disaster Relief and Emergency Assistance
Act of 1974. Of the $16 billion in CDBG-DR funds appropriated, $4.21 billion was awarded
directly to the City. As of September 30, 2015, City officials had allocated $1.7 billion of that
amount to the Build it Back Single Family Program and had disbursed $270 million.
The Mayor’s Office of Housing Recovery Operations administers the Build it Back Program Single
Family Program. The Program provides assistance to homeowners whose primary residence or
one- to four-unit rental property was destroyed or damaged by Hurricane Sandy. Disaster assistance
is offered through several Program options, such as repair, repair with elevation, rebuild,
reimbursement, relocation, and acquisition. Our review was limited to the reimbursement Program
option.1
The reimbursement Program option provides assistance to all eligible homeowners who
completed Hurricane Sandy-related repairs with personal resources and met the national
objectives of having low to moderate income or urgent need. To be eligible for reimbursement
assistance, the homeowner must have completed the rehabilitation work within the same
footprint of the damaged structure, sidewalk, driveway, parking lot, or other developed area and
incurred costs by the Program application date or October 29, 2013, whichever came first.
The Program did not require documentation of completed repair expenses to receive
reimbursement assistance. A Program assessment determined the value of the completed repairs
using a standardized pricing model and multipliers to provide a set cost for reimbursement. The
determined cost was then reduced by the amount of duplicative assistance made available to the
homeowner from other funding sources. Due to funding limitations, the Program reimbursed
homeowners at a rate of 60 percent of their eligible reimbursement amount.




1
    At the time the review was started, City officials had drawn CDBG-DR funds for the reimbursement option only.



                                                         3
The objective of the audit was to determine whether City officials had adequate controls to ensure
that the use of CDBG-DR funds was consistent with the Build it Back Single Family Program
guidelines established under the HUD-approved action plan.




                                                 4
Results of Audit

Finding 1: City Officials Did Not Always Implement Policies and
Procedures To Ensure That Funds Were Disbursed in Accordance
With the Action Plan
City officials implemented policies and procedures that did not always ensure that CDBG-DR
funds were disbursed in accordance with the action plan. The action plan stated that the Program
would provide assistance that covered a portion of eligible reimbursable expenses. However,
due to policy changes, City officials provided several homeowners with additional assistance that
would cover from 60 to 100 percent of their eligible reimbursable expenses, while they provided
other homeowners with less than the Program’s 60 percent reimbursement rate. We attributed
these deficiencies to City officials’ decision to provide additional assistance to homeowners
receiving SBA loans and to reduce operational costs associated with a second grant agreement
signing. As a result, all homeowners were not reimbursed at the same rate and City officials
could not ensure that all eligible homeowners received fair and equitable treatment.
Accordingly, City officials disburse an additional $4.5 million2 in CDBG-DR assistance to
homeowners solely because they received Small Business Administration (SBA)3 loans. They
also overpaid 5 homeowners by $32,107 and underpaid 11 homeowners by $101,398.
Due to a Change in the Reimbursement Policy, All Homeowners Would Not Be
Reimbursed in Accordance With the Action Plan
Due to a change in the reimbursement policy affecting homeowners receiving SBA loans, all
homeowners would not be reimbursed in accordance with the action plan and at the same
reimbursement rate. While the City’s action plan stated that the Program would provide
assistance that covered a portion of eligible reimbursable expenses, neither the City’s action plan
nor five versions of the policy manual, dated before March 23, 2015, defined what was meant by
a “portion” or the rate at which eligible expenses would be reimbursed. However, public
comments City officials received requesting a reimbursement percentage higher than 60 percent
showed that the public was informed of the Program’s reimbursement rate. Upon applying to the
Program, homeowners were required to sign a “Notice and Acknowledgment of Limited
Availability of Funding,” acknowledging that reimbursement assistance was subject to the
availability of CDBG-DR funds. The Program’s coordination of benefits and reimbursement
worksheet stated that due to funding limitations, a homeowner would be reimbursed at a rate of
60 percent of the eligible reimbursement amount. City officials decided to reduce the Program-
calculated reimbursement amount by 40 percent based on early estimates of available CDBG-DR
funds and to lessen the potential that the Program may have overvalued completed repairs.


2
    While City officials disbursed $5.8 million to homeowners with SBA loans, of that amount, they drew $4.5
    million in CDBG-DR funds as of September 20, 2016.
3
    In the wake of a disaster, the U.S. Small Business Administration provides low-interest disaster loans to
    homeowners and is the Federal Government’s primary funding source for long-term recovery assistance.



                                                           5
Effective July 24, 2015, City officials implemented the new policy bulletin, in which
homeowners with SBA loans would either receive an additional reimbursement equivalent to the
lesser of the remaining 40 percent of the eligible reimbursement amount or the amount of the
SBA loan considered to be a duplication of benefits. City officials updated the policy manual on
September 16, 2015. However, the language in section 14.12 of the policy manual differed from
that in the policy bulletin in that it also stated that homeowners with SBA loans would receive an
additional reimbursement equivalent to the lesser of the remaining 40 percent of the eligible
reimbursement amount or the current disbursed amount of all SBA loans calculated at a point in
time. Homeowners would not be reimbursed for more than 100 percent of the eligible
reimbursement amount. The policy difference is important because SBA offers several loan
types, including loans for real estate (property damage), mitigation, and personal property
(contents). However, according to section 7 of the City’s Coordination of Benefits Standard
Operating Procedures, the Program included only the amount the homeowner received for
property damage in the homeowner’s duplication of benefits calculation. As a result, a
homeowner could potentially receive more in additional assistance than the amount of the SBA
loan considered duplicative.
During our review, City officials provided a list of potential homeowners with SBA loans who
were projected to receive additional reimbursements. An analysis of 817 potential homeowners
showed that 140 homeowners were projected to receive $2.2 million in additional
reimbursements, which was approximately $1 million more than the $1.2 million in SBA loans
considered duplicative. Of the 140 homeowners, 66 were projected to receive additional
reimbursements that were more than the amount of the SBA loans considered duplicative but less
than all disbursed SBA loans, 47 were projected to receive additional reimbursements that were
equivalent to the amount of all disbursed SBA loans, and 27 were projected to receive additional
reimbursements that were equivalent to the remaining 40 percent of the eligible reimbursement
amounts. The details are shown in the table below.
           Projected value of additional reimbursement exceeds the value of SBA loans considered duplicative
             Column A                    Column B         Column C          Column D       Column E        Column F
                                                        Remaining 40                                        Excess
                                         Projected      percent of the      Amount of                       amount
                                         additional        eligible         SBA loans      Amount of      (column B
    Number of homeowners per          reimbursement    reimbursement        considered     disbursed    minus column
              category                    amount           amounts         duplicative     SBA loans          D)
 66 - projected to receive
 additional reimbursements that
 were more than the amount of
 the SBA loans considered
 duplicative but less than all
 disbursed SBA loans                         $963,291         $988,644          $603,000   $1,629,400        $360,291
 47 - projected to receive
 additional reimbursements that
 were equivalent to the amount of
 disbursed SBA loans                          689,500        1,095,200           341,410      689,500         348,090
 27 - projected to receive
 additional reimbursements that
 were equivalent to the remaining
 40 percent of the eligible
 reimbursement amounts                        551,547          551,547           265,300             0        286,247
                               Total       $2,204,338       $2,635,391        $1,209,710   $2,318,900        $994,628




                                                          6
A sample of 84 homeowners with SBA loans showed that approximately 73 percent had a
national objective determination of urgent need and 27 percent had a low- to moderate-income
determination. Therefore, the City’s policy change would primarily benefit higher income
homeowners and provide them with additional reimbursement of up to 100 percent of their
maximum reimbursement amount. However, the City’s action plan stated that the grant amount
would cover only a portion of eligible reimbursable expenses. In accordance with Federal
Register Notice 78 FR 14338 (March 5, 2013), before implementing this policy change, City
officials should have amended the City’s action plan since there was a change in the Program
benefit from 60 to 100 percent of eligible reimbursable expenses solely for homeowners with
SBA loans. The action plan amendment process included publishing the proposed amendment
on the City’s Web site and giving citizens the opportunity to submit comments.
According to City officials’ May 12, 2016, projections and our data analysis,4 additional
reimbursements to be provided to 817 homeowners with SBA loans totaled approximately $10.2
million. We considered the grants yet to be disbursed to be funds that could be put to better use.
The reduction in outlays could be better used to benefit all eligible homeowners in a more fair
and equitable manner.
In response to the draft report, City officials provided an updated analysis of reimbursement
review activity related to the 817 homeowners with SBA loans included in our initial review.
They also provided an analysis of other homeowners with SBA loans who were not a part of our
review. Based on the updated analysis, City officials determined that 549 homeowners with
SBA loans received additional reimbursements totaling $5.8 million. Of the 549 homeowners
with SBA loans, 532 were included in City officials’ May 12, 2016, projections. As of
September 20, 2016, City officials disbursed an additional $5.75 million to these 532
homeowners. This amount was 45 percent less than the initially projected disbursement amount
of $10.2 million. Further, City officials reported that while 135 of 532 homeowners with SBA
loans received additional reimbursements equivalent to 100 percent all disbursed SBA loans, no
homeowner received more than 100 percent of non-duplicative reimbursable expenses.
Our analysis of the updated data found that 393 of the 549 homeowners with SBA loans received
100 percent of the maximum reimbursement amount and 1 homeowner received $1,793 more
than the maximum amount allowed. However, a review of the homeowner’s Program file
showed that the excess amount was reported in error6. The first reimbursement approved amount
was reported as $4,507 when it should have been reported as $5,378. The second reimbursement
approved amount was reported as $5,378 when it should have been reported as $3,585.




4
    A review of Program files relating to homeowners identified through our data analysis found inaccuracies in the
    data. These inaccuracies are discussed in detail in finding 3 of this report.
5
    While City officials disbursed $5.7 million to 532 homeowners with SBA loans, of that amount, they drew $4.4
    million in CDBG-DR funds as of September 20, 2016.
6
    Data errors found in the updated analysis of reimbursement review activity are discussed in detail in finding 3 of
    this report.



                                                            7
Accordingly, the amount of additional reimbursements disbursed to 549 homeowners with SBA
loans was reduced by the amount of the $1,7937 error.
The use of $4.5 million in CDBG-DR assistance to provide additional reimbursement covering
up to 100 percent of eligible expenses incurred by homeowners who received SBA loans was
inconsistent with the HUD-approved action plan. The action plan stated that the Program would
provide assistance that covered a portion of eligible reimbursable expenses. Since there was a
change in program benefit, City officials were required to amend the action plan. In the absence
of an amended action plan, $4.5 million is considered to be an ineligible use of CDBG-DR funds
and $1.3 million yet be to drawn is considered to be funds that could be put to better use.
Some Homeowners Received More While Others Received Less Than the Program’s 60
Percent Reimbursement Rate
Due to a policy City officials implemented on January 29, 2015, some homeowners received
more while others received less than the Program’s 60 reimbursement rate. The policy provided
that once a reimbursement grant agreement had been signed by a homeowner, a new agreement
would not be executed to show a change in the grant amount resulting from a duplication of
benefits recalculation in cases in which the original grant was less than the Program-recalculated
maximum reimbursement amount. For example, the initial Program-calculated maximum
reimbursement amount for homeowner A was $10,000. Therefore, the original grant amount for
homeowner A was $6,000 ($10,000 x 60 percent). If the maximum reimbursement amount for
homeowner A was recalculated to $8,000, City officials would not execute a new grant
agreement. The grant for homeowner A would continue to be processed at $6,000, although it
should have been reduced to $4,800 ($8,000 x 60 percent). Further, the policy memorandum
was to be included in the file for each grant processed in that manner.8
In cases in which the original grant was more than the Program-recalculated maximum
reimbursement amount, the grant would be processed at the newly calculated amount, and a new
agreement would be executed. For example, the initial Program-calculated maximum
reimbursement amount for homeowner B was $10,000. Therefore, the original grant amount for
homeowner B was $6,000 ($10,000 x 60 percent). If the maximum reimbursement amount for
homeowner B was recalculated to $5,000, City officials would execute a new grant agreement.
In this example, the grant for homeowner B would be processed at the newly recalculated
reimbursement amount of $3,000 ($5,000 x 60 percent) since the original grant of $6,000 was
more than the Program-recalculated maximum reimbursement amount of $5,000.
An analysis of a sample of 16 homeowners whose grant amount changed by $500 or more due to
duplication of benefits recalculations after the homeowners had already signed their
reimbursement grant agreements, showed that City officials overpaid for 5 grants by $32,107 and
underpaid for 11 grants by $101,398. Further, contrary to the City’s policy, one of the five
overpaid grants continued to be processed at the original grant amount of $54,533, although the


7
    City officials reported that 549 homeowners received a total of $5,783,160 in additional reimbursement assistance.
    However, this amount was reduced by the amount of the $1,793 error to $5,781,367.
8
    Documentation deficiencies related to grants processed in accordance with the January 29, 2015, memorandum
    are discussed in detail in finding 3 of this report.



                                                           8
original grant amount was more than the Program-recalculated maximum reimbursement amount
of $49,463. The Program-recalculated maximum reimbursement amount was reduced due to
$41,483 in flood insurance proceeds and $57 in allowable credits9 identified after the initial
duplication of benefits calculation had been performed. The details regarding the overpaid and
underpaid grants and the reimbursement percentages are shown in the tables below.
                                                   Overpaid grants
Column A         Column B          Column C          Column D             Column E          Column F        Column G           Column H
                                                                                                                             Percentage of
                                                                                                                                 revised
                                                                                                                               maximum
                                                                                           HUD OIG*                         reimbursement
                  Original            New                                  Original        recalculated                          amount
Number of         maximum           maximum            Difference            grant            grant          Difference         received
 overpaid      reimbursement     reimbursement        (column B -          (60% of       (60% of column     (column E -       (column E /
  grants           amount            amount            column C)          column B)             C)           column F)         column C)
    1                 $90,889           $49,463            $41,426             $54,533           $29,678         $24,855               110%
    2                  20,861             14,588             6,273              12,516              8,753           3,763                86%
    3                  64,921             62,630             2,291              38,953            37,578            1,375                62%
    4                  21,651             19,879             1,772              12,990            11,927            1,063                65%
    5                   9,849              8,096             1,753               5,909              4,858           1,051                73%
     Total           $208,171          $154,656            $53,515            $124,901           $92,794         $32,107
*OIG = Office of Inspector General

                                                   Underpaid grants
Column A         Column B          Column C           Column D            Column E          Column F        Column G            Column H
                                                                                                                              Percentage of
                                                                                                                                 revised
                                                                                            HUD OIG                             maximum
                  Original            New                                  Original        recalculated                      reimbursement
Number of         maximum           maximum            Difference            grant            grant          Difference     amount received
underpaid      reimbursement     reimbursement        (column B -          (60% of       (60% of column     (column E -        (column E /
  grants           amount            amount            column C)          column B)             C)           column F)          column C)
     1               $140,885          $208,726          ($67,841)             $84,531          $125,236        ($40,705)                40%
     2                  18,350           61,580           (43,230)              11,010             36,948        (25,938)                18%
     3                   1,591           19,932           (18,341)                 955             11,959        (11,004)                 5%
     4                137,898           155,898           (18,000)              82,739             93,539        (10,800)                53%
     5                  34,895           44,095             (9,200)             20,937             26,457         (5,520)                47%
     6                   5,643           10,643             (5,000)              3,386              6,386         (3,000)                32%
     7                  41,351           44,222             (2,871)             24,811             26,533         (1,722)                56%
     8                   1,114            2,414             (1,300)                669              1,449           (780)                28%
     9                   6,811            8,071             (1,260)              4,087              4,843           (756)                51%
    10                  38,781           39,831             (1,050)             23,269             23,899           (630)                58%
    11                  13,690           14,595               (905)              8,214              8,757           (543)                56%
       Total         $441,009          $610,007         ($168,998)            $264,608          $366,006      ($101,398)


Consistent with section 312(c) of the Stafford Act, both the coordination of benefits worksheet
disclosure and section 13(a)(iii) of the reimbursement grant agreement provided that a
homeowner might be required to repay the Program additional disaster recovery funds received
from other sources that were considered to be a duplication of benefits after the grant had been
awarded. However, City officials chose to continue to process the five overpaid grants at the
original amounts. Thus, the City should be held responsible for repaying the Program $32,107
from non-Federal funds.


9
    The Program gave homeowners credit for documented, eligible, and nonpermanent Sandy-related repair expenses,
    thus increasing the reimbursement assistance homeowners received.



                                                                      9
We attributed these deficiencies to City officials’ desire to reduce operational costs associated
with a second grant agreement signing. Further, City officials stated that additional assistance to
homeowners above the 60 percent reimbursement rate was allowable under the reimbursement
grant agreement.
Conclusion
The City’s action plan stated that the Program would provide assistance that covered a portion of
eligible reimbursable expenses. The Program defined a portion as 60 percent of eligible
reimbursable expenses. Yet City officials implemented policies that did not ensure that all
eligible homeowners received fair and equitable treatment. As of May 12, 2016, City officials
planned to disburse $10.2 million in additional CDBG-DR assistance to homeowners solely
because they received SBA loans. However, an updated analysis of the City’s reimbursement
review activity, provided on September 20, 2016, showed that City officials drew $4.5 million
and planned to draw $1.3 million in CDBG-DR funds for additional reimbursement assistance to
homeowners with SBA loans. As a result of City officials’ policy decision, the additional
assistance increased the reimbursement percentage from 60 to up to 100 percent of eligible
reimbursable expenses for homeowners with SBA loans. An analysis of the data showed that
39410 of the 549 homeowners with SBA loans received a total of 100 percent of their eligible
reimbursable expenses. City officials also provided 5 homeowners with $32,107 more and 11
others with $101,398 less than 60 percent of their eligible reimbursable expenses. This occurred
because City officials wanted to cut operational costs associated with conducting a second
reimbursement grant agreement signing to show grant amount changes resulting from
recalculated duplication of benefits. The City’s position that additional CDBG-DR assistance up
to 100 percent of eligible reimbursable expenses was allowable under the reimbursement grant
agreement was inconsistent with the HUD-approved action plan.
Recommendations
We recommend that HUD’s Acting Deputy Assistant Secretary for Grant Programs instruct City
officials to
           1A.      Submit an amended action plan for approval to ensure that it agrees with the
                    City’s policies regarding the use of $4,467,299 and planned use of $1,314,068 in
                    additional CDBG-DR assistance covering up to 100 percent of eligible
                    reimbursable expenses incurred by homeowners with SBA loans. If an amended
                    action plan is not submitted and approved, repay the Program from non-Federal
                    funds for additional reimbursements provided solely to homeowners with SBA
                    loans.
           1B.      Execute new grant agreements to show material changes in grant amounts
                    resulting from duplication of benefits recalculations.



10
     It was determined that 394 homeowners with SBA loans received a total of 100 percent of their maximum
     reimbursement amount after the first and second reimbursement approved amounts related to the previously
     discussed homeowner was corrected.



                                                          10
1C.   Repay the Program from non-Federal funds $32,107 in overpaid grants to
      homeowners whose grant amounts (1) were not revised to show recalculated
      duplication of benefits and (2) exceeded the Program’s 60 percent reimbursement
      rate.
1D.   Reimburse $101,398 in additional grants owed to the 11 homeowners whose grant
      amounts should have been materially increased as a result of recalculated
      duplication of benefits.




                                      11
Finding 2: City Officials Did Not Always Comply With HUD’s
Lead Safe Housing Rule Requirements
The Program files in the City’s electronic Case Management System (CMS) lacked
documentation showing compliance with HUD’s Lead Safe Housing Rule requirements11 and
contained lead test result discrepancies. CMS lacked documentation showing that lead-based
paint testing had been performed, identified hazards had been removed, and clearance had been
achieved. We attributed these deficiencies to design limitations in CMS, City officials’
expectation that homeowners would allow the Program to perform the lead hazard work
necessary to ensure compliance, and their wish to avoid recapturing grants from homeowners
who did not to allow the Program to complete the lead hazard work or provide evidence that
compliance had been otherwise achieved. As a result, City officials did not show that the
Program disbursed more than $1 million in CDBG-DR assistance for 47 lead-safe homes.
Program Files Lacked Documentation of Compliance With HUD’s Lead Safe Housing Rule
Requirements
Our review of 14 Program files found that the environmental analysis view in CMS did not
clearly document whether the CDBG-DR-assisted homes were exempt from HUD’s Lead Safe
Housing Rule requirements or how identified lead hazards would be addressed. HUD’s Office
of Community Planning and Development (CPD) Notice CPD-15-07 stated that a CDBG-DR-
assisted home may be exempt from HUD’s Lead Safe Housing Rule requirements in accordance
with 24 CFR (Code of Federal Regulations) 35.115 if the home

          Was built after January 1, 1978,
          Required repairs necessary to respond to an emergency,
          Underwent rehabilitation that did not disturb any painted surface,
          Was designated exclusively for the elderly or persons with disabilities but only if no
           children under the age of 6 were expected to reside in the home,
          Underwent a property inspection and no lead-based paint was found, and
          Underwent a property inspection and lead-based paint had been identified and removed
           and the property had achieved clearance.

Of the 14 files reviewed, only one CDBG-DR-assisted home was clearly exempt from HUD’s
lead requirements because the home was built in 1994. However, the home’s environmental
analysis view in CMS contained blank fields concerning lead. The environmental analysis view
of another assisted home built in 1920 showed that it had been tested for the presence of lead.
Yet, the reportedly negative test results were not documented in CMS. In addition, four Program
files did not contain evidence that identified lead hazards had been removed and the homes had
achieved clearance.
City officials explained that due to a design limitation in CMS, lead-related fields were visible,
although they were not used to determine reimbursement eligibility or postreimbursement


11
     HUD’s Lead Safe Housing Rule requirements apply to all pre-1978 housing units receiving CDBG funding to
     reduce the threat of childhood lead poisoning in federally assisted housing.



                                                         12
compliance. These lead-related fields were used to facilitate later construction. City officials
required the homeowners to allow the Program to perform the work necessary to address the lead
hazards during the repair option or provide a lead clearance report. Accordingly, without
evidence of documented negative lead test results for one homes and the removal of identified
lead hazards and lead clearance reports for four homes, City officials could not show that
$182,660 in CDBG-DR assistance was used for lead-safe homes.
Due to our audit inquiry, City officials identified and provided a list of an additional 41
homeowners who received reimbursement checks totaling $833,199, although the Program had
not performed lead-based paint testing, removed identified hazards, or achieved clearance for the
homes. City officials believed that during the Program’s repair option, which followed the
homeowners’ receipt of reimbursement assistance, the Program would perform the lead hazard
work necessary to ensure that CDBG-DR funds were used for lead-safe homes. However, the
homeowners opted out, and City officials wished to avoid recapturing grants when the
homeowners refused to take part in the Program’s repair option because they believed that grant
recaptures were not in the homeowners’ best interest. While City officials stated that they would
work with HUD to ensure that these homeowners complied with the lead requirements, their
only recourse may be to enforce the reimbursement grant agreement, which would require that
the homeowners allow the Program to complete the lead hazard work, otherwise achieve
clearance, or repay the grant.
Program Files Contained Discrepancies Concerning Lead Test Results
The NYC Houses: Feasibility Determination Report for 912 of 14 homes in our sample showed
that the homes had been tested for the presence of lead hazards and asbestos-containing materials
and the complete test results had been attached to the record in CMS. However, City officials
acknowledged that this statement was not always accurate since the template for the Feasibility
Determination Report automatically generated this statement. City officials stated that once they
discovered the inaccuracy of the autogenerated statement, they revised the report’s template.
Our file review uncovered one instance13 in which the Feasibility Determination Report showed
that the home had been tested for the presence of lead, but the test results were not documented
in the Program file. In another instance, one Program file contained a discrepancy concerning
the lead test results. The NYC Build it Back Repair Program Lead-Based Paint Risk Assessment
Report Summary, dated February 18, 2014, showed that the home did not test positive for lead-
based paint or a lead-based paint hazard. However, the March 12, 2014, lead test results showed
that there was a potential lead exposure hazard at the home and lead hazard reduction activities
were required. Since this discrepancy was not reconciled, City officials could not ensure that
$29,019 in CDBG-DR assistance was used for a lead-safe home.




12
   Of five remaining homes, the Feasibility Determination Reports for three stated that the homes were or would be
   tested for the presence of lead hazards. The report for the fourth home stated that the home would be tested, and
   the report for the fifth home stated the home was exempt from testing based on the year of construction.
13
   This example is the same one that was mentioned in a previous section.



                                                          13
Conclusion
City officials did not maintain adequate records in CMS to show compliance with HUD’s Lead
Safe Housing Rule requirements. CMS lacked documentation showing that lead-based paint
testing had been performed, identified hazards had been removed, and clearance had been
achieved. As a result, City officials could not provide assurance that more than $1 million in
CDBG-DR assistance was disbursed for 47 lead-safe homes. These deficiencies occurred due to
design limitations in CMS and City officials’ belief that it was not in the homeowners’ best
interest to recapture grants when homeowners declined the Program’s assistance or did not
otherwise achieve lead compliance.
Recommendations
We recommend that HUD’s Acting Deputy Assistant Secretary for Grant Programs coordinate
with the Office of Healthy Homes and Lead Hazard Control to provide technical assistance and
instruct City officials to

      2A.     Ensure that Program files clearly identify whether a home required lead-based
              paint testing. When such testing is performed, City officials should ensure that
              the testing results are documented, identified lead-based paint hazards are
              removed, and clearance is achieved.
      2B.     Document the negative lead test results in CMS for one home and the removal of
              identified lead hazards and lead clearance reports for four homes, thus ensuring
              that $182,660 in CDBG-DR assistance was disbursed for lead-safe homes. If the
              negative test results are not documented, City officials should repay the $182,660
              from non-Federal funds.
      2C.     Provide supporting documentation that lead-based paint testing was performed,
              identified hazards were removed, and clearance was achieved for the 41
              properties for which homeowners received $833,199 in CDBG-DR assistance. If
              supporting documentation is not provided, City officials should repay the
              $833,199 from non-Federal funds.
      2D.     Advise homeowners of their obligation under the terms of the reimbursement
              grant agreement to allow the Program to perform lead-based paint testing or
              hazard removal. Homeowners who refuse to allow the Program to complete lead
              hazard work or provide evidence that the property achieved clearance must repay
              the grant.
      2E.     Reconcile the discrepancy in the lead test results. If it is determined that the home
              tested positive for a lead-based paint hazard, City officials should provide
              supporting documentation showing that the hazard has been removed and the
              home has achieved clearance, thus ensuring that $29,019 in CDBG-DR assistance
              was disbursed for a lead-safe home. If the lead test results are not reconciled and
              the lead safety of the home is not documented, City officials should repay the
              $29,019 from non-Federal funds.




                                                14
Finding 3: City Officials Did Not Always Maintain Complete and
Accurate Program Files and Records
City officials did not maintain complete and accurate Program files and records to show
compliance with all requirements. Specifically, the City’s electronic Program files and records
(1) contained duplicative forms, (2) included nonstandard file names and unused standard
document folders and subfolders, (3) did not always contain executed grant agreements that were
countersigned before reimbursement assistance was provided, (4) required clarification regarding
SBA loan data and eligible Sandy-related expense offsets, and (5) contained discrepancies
regarding projected additional reimbursements and duplication of benefits recalculations. We
attributed these deficiencies to the lack of adequate monitoring and quality control reviews to
ensure that the records were accurate and documentation was maintained in accordance with
Federal requirements14 and the City’s record-keeping procedures. As a result, City officials did
not provide assurance that records were complete, reliable, and reconciled with the source
documentation in the Program files to facilitate the review of overall Program compliance.
Program Files Contained Duplicative Documentation and Forms
Duplicate forms were not moved to the CMS duplicate or changed document or excess materials
folders for 6 of 14 files reviewed as required by section 3.2.1, subsection 3, of the City’s record-
keeping procedures. While the City’s record-keeping procedures were refined due to significant
Program changes, City officials did not establish adequate monitoring controls to ensure that the
documentation was appropriately maintained. As a result, during our file review, the dates and
amounts of each duplicative form had to be compared to determine the applicable version. This
was a cumbersome process and added to the time needed to complete the file review. City
officials acknowledged that the Program files in CMS contained duplicative forms and that
documentation may not have been properly filed. Accordingly, they planned to implement a
document and data cleanup process before closing out the files to ensure that duplicative
documents were archived, most recent documents were identified and filed in the appropriate
subfolders, and all files complied with record-keeping requirements.
Program Files Contained Nonstandard File Names and Unused Standard Document
Folders and Subfolders
Our review of 14 files found that contrary to sections 3.1.1 and 3.2 of the City’s record-keeping
procedures, City officials did not consistently use standard file names. Further, standard
document folders in CMS, such as eligibility and its subfolders, proof of location, ownership,
primary residency, citizenship, and identity, were often unused. In several instances,
documentation that should have been maintained in clearly labeled folders was misfiled.
In its July 18, 2014, monitoring report, HUD expressed concern that City officials were not
maintaining Program files in a clear and organized manner to facilitate access to key
information, such as duplication of benefits determinations, HUD’s Office of Fair Housing and
Equal Opportunity-required race and ethnicity data, and records necessary to determine
compliance with CDBG-DR crosscutting requirements. In their August 15, 2014, response to

14
     Regulations at 24 CFR 570.506 require City officials to establish and maintain sufficient records to show
     compliance with the Program’s requirements.



                                                            15
HUD’s concern, City officials stated that they had developed a single final applicant folder,
which contained the homeowner’s complete file. However, only 4 of 14 files reviewed
contained the final applicant folder.
In addition, City officials developed a single end-to-end file, which contained supporting
documentation to show compliance with Program requirements. However, they did not
consistently maintain the end-to-end file in the eligibility and benefits certification document
folder as required. In several instances, the end-to-end file was found in other folders, such as
the end-to-end folder or the final applicant folder, or in the documents section in no particular
folder. These deficiencies occurred because City officials did not establish adequate monitoring
controls to ensure that the Program files were clearly and consistently organized in accordance
with the City’s record-keeping procedures. Monitoring consisted of a quality control sample of
approximately 10 percent of the end-to-end files.

Executed Grant Agreements Were Not Always Documented as Required and
Countersigned by City Officials After Assistance Had Been Provided
HUD encouraged City officials to include the signed executed grant agreements in the end-to-
end file; however, none of the 14 files reviewed contained the executed grant agreement in that
file. The executed grant agreements were filed in the grant agreement subfolder under the grant
award folder. Further, contrary to section 3.2.1, paragraph 6a, of the City’s record-keeping
procedures, the executed grant agreements for three of the files reviewed were initially missing
from the grant award folder. However, as a result of our inquiry, City officials entered the three
executed grant agreements into CMS.
Although the 14 homeowners in our sample signed their reimbursement grant agreements before
receiving assistance, City officials countersigned 11 of the 14 agreements an average of 115 days
after the CDBG-DR assistance had been provided. The details are in the table below.
                        Reimbursement grant agreements executed after issuance of reimbursement checks
          Date applicant(s)                                 Date City official           Elapsed days from the countersigning of
            signed grant       Date reimbursement          countersigned grant          the grant agreement to the issuance of the
Count        agreement             check issued                agreement                          reimbursement check
  1                12/16/2014            12/29/2014                      11/30/201515                                          336
  2                04/22/2015            04/28/2015                      02/19/ 201615                                         297
               08/04/2015 and                                                                                                  191
      3           08/06/201516           08/12/2015                       02/19/201615
      4            04/01/2015            04/07/2015                        08/06/2015                                          121
      5            03/12/2015            03/09/2015                        07/07/2015                                          120
      6            07/10/2014            07/21/2014                        10/06/2014                                           77
      7            06/28/2014            07/14/2014                        09/24/2014                                           72
      8            04/17/2015            04/27/2015                        05/22/2015                                           25
      9            03/11/2015            03/25/2015                         04/8/2015                                           14
     10            10/13/2014            10/14/2014                        10/28/2014                                           14
     11            12/12/2014            12/22/2014                        12/24/2014                                            2




15
   As a result of our inquiry, City officials entered the executed reimbursement grant agreement into the grant
   agreement document folder in CMS in compliance with section 3.2.1, subsection 6, of the Record Keeping
   Standard Operating Procedure.
16
   Multiple homeowners signed the grant agreement on different dates.



                                                               16
City officials explained that due to a system issue, CMS did not notify designated users when a
grant agreement requiring a City official’s signature had been entered into CMS. Upon
discovery of the system issue, the City initiated corrective action, and as of February 2016, City
officials reported that more than 90 percent of grant agreements had been signed and entered into
CMS. Further, City officials held that the remaining grant agreements would be signed and
entered into CMS by the end of March 2016. City officials were developing a system
enhancement that would notify designated users daily when a grant agreement was ready to be
countersigned by a City official. Timely countersigning of a grant agreement is consistent with
prudent business practices.
Data Required Clarification
SBA loan and allowable activity credit data in CMS needed clarification to verify the
homeowners’ grant calculations. While City officials verified and documented SBA loan data
obtained through direct data feeds from SBA, the data contained undefined canceled codes and
unreconciled loan amounts that needed clarification and confirmation. As a result, we
questioned why City officials excluded three canceled SBA loans and a portion of a fourth loan
from the duplication of benefits calculations without documented explanations as required by
HUD Guidance on Duplication of Benefit Requirements and Provision of CDBG Disaster
Recovery (DR) Assistance, dated July 25, 2013.
Regarding the first SBA loan, City officials stated that a documented analysis of the canceled
loan was unnecessary because the $20,700 SBA loan was for personal property only. The file
did not contain an indication that the loan was for personal property and thus, it was not
considered duplicative assistance. Regarding the second and third SBA loans, City officials
stated that the loans were canceled by SBA as indicated by the code C12. This code reportedly
showed that there was an adverse change in the homeowner’s income that made the loan no
longer affordable, resulting in loan cancellation. However, the canceled code was not clearly
defined in the file or in the City’s Coordination of Benefits Standard Operating Procedure.
Regarding the fourth SBA loan, although SBA documentation, dated January 26, 2013, showed
that the homeowner received a $14,000 loan to be used for real estate repair and replacement,
City officials included only $5,300 of the loan amount in the duplication of benefits calculation.
They explained that a direct data feed received from SBA on October 10, 2014, showed that
$5,300 of the $14,000 SBA loan was for property damage. The remaining $8,700 was for
personal property. However, the files did not include documentation showing that $8,700 was
for personal property. Since the files did not contain documentation to support City officials’
explanations, direct confirmation was obtained from an SBA official.
City officials used documented eligible Sandy-related expenses to offset the amount of
potentially duplicative benefits, thus increasing the reimbursement assistance a homeowner was
eligible to receive. However, documentation in four files, namely the receipts, did not clearly
identify the expenses related to eligible recovery activities for which the homeowners were given
credit. As a result, the total allowable activity offsets or total verified allowable activity credits
shown in the coordination of benefits worksheet could not be easily verified. Due to our inquiry,
a City official provided four itemized listings of receipts detailing what was and was not
allowable. However, at the time of our file review, the documentation provided should have
been maintained in the Program files to clearly identify items eligible for offset credit.


                                                  17
Program Files and Records Contained Discrepancies
City officials provided an Excel spreadsheet showing 834 homeowners with SBA loans, to
whom they planned to disburse additional reimbursements of up to 40 percent of the
homeowners’ eligible reimbursement amounts. The projected value of the additional
reimbursements was approximately $10.5 million, but the documentation in the Program files did
not agree with the data provided. Using data analytics, we determined that eight homeowners
received additional reimbursements totaling $54,824. A review of the documentation in the
Program files found that additional reimbursements had not been provided since the homeowners
were not eligible for additional assistance. Further, City officials planned to provide additional
reimbursements to 38 homeowners for whom the projected amounts of the additional
reimbursements to be provided were equivalent to the remaining 40 percent of their eligible
reimbursement amounts. These amounts were more than both the SBA loans counted as
duplicative and all disbursed loans. However, the documentation in the Program files showed
that City officials determined that 11 homeowners were not eligible to receive $243,857 in
projected additional reimbursements. For 1 of the 11 homeowners, CMS documented that City
officials planned to provide additional reimbursement in the amount of $20,000, which was
$48,791 less than originally projected.
Since the data in the Program files did not reconcile with the data City officials provided, the
projected amount of additional reimbursements to be provided solely to homeowners with SBA
loans was reduced from approximately $10.5 million to approximately $10.2 million, and the
number of applicable homeowners with SBA loans was reduced from 834 to 817 to remove the
inaccurate data identified. We attributed this deficiency to the fact that City officials’ projection
of additional CDBG-DR assistance solely to homeowners with SBA loans may not have been
finalized and subject to a quality control review due to ongoing homeowner eligibility reviews.
This was particularly true since 53 homeowners with SBA loans were projected to receive more
than $1 million in additional reimbursements, which was more than $1.1 million more than the
remaining 40 percent of the homeowners’ eligible reimbursement amount of $408,184. This
practice was contrary to section 14.12 of the City’s policy manual, which states that the Program
may not reimburse a homeowner in an amount that exceeds 100 percent of the eligible
reimbursement amount as stated in finding 1.
As indicated in finding 1, in response to the draft report, City officials provided an Excel
spreadsheet containing an updated analysis of reimbursement review activity related to the 817
homeowners with SBA loans included in our initial review. They also provided an analysis of
other homeowners with SBA loans who were not a part of our review. Using data analytics, we
determined that one homeowner received $1,793 more than the maximum reimbursement
amount allowed. However, a review of the homeowner’s Program file showed that the excess
amount was reported in error. The first reimbursement approved amount was reported as $4,507
when it should have been reported as $5,378. The second reimbursement approved amount was
reported as $5,378 when it should have been reported as $3,585. Accordingly, the amount of




                                                  18
additional reimbursements disbursed to 549 homeowners with SBA loans was reduced by the
amount of the $1,79317 error.
Moreover, worksheets within the Excel spreadsheet contained conflicting data as detailed in the
table below.
   Conflicting data within the Excel spreadsheet containing the updated analysis of reimbursement review
                                 activity of 817 homeowners with SBA loans
          Worksheet name                  Number of SBA recipients        Second reimbursement approved
                                            approved for additional                     amount
                                                reimbursement
Summary table                                                       532                          $5,655,752
Summary table dataset (817)                                         695                          $7,297,373
Full Set (817)                                                      701                          $7,315,097

However, we were able to reconcile the conflicting data as detailed in the tables below.

  Reconciliation of conflicting data between worksheets containing the updated analysis of reimbursement
                              review activity of 817 homeowners with SBA loans
                                                 Summary table
                                                 Number of SBA recipients  Second reimbursement approved
                                                                                        amount
Approved                                                             532                         $5,655,752
Not approved                                                        28518                        $1,641,621
Totals                                                               817                         $7,297,373

                                         Summary table dataset (817)
                                          Number of SBA recipients            Second reimbursement approved
                                                                                         amount
Approved                                                               695                         $7,297,373
Not approved                                                           122                                 $0
Totals                                                                 817                         $7,297,373

                                         Summary table dataset (817)
                                          Number of SBA recipients            Second reimbursement approved
                                                                                         amount
Approved                                                               701                         $7,315,097
Subtract 6 SBA recipients for whom                                      (6)                          ($17,724)
duplicate second reimbursement
approved amounts were reported
Revised approved                                                       695                          $7,297,373
Not approved                                                           122                                  $0
Totals                                                                 817                          $7,297,373



17
   City officials reported that 549 homeowners received a total of $5,783,160 in additional reimbursement
   assistance. However, this amount was reduced by the amount of the $1,793 error to $5,781,367.
18
   While City officials reported that 285 of the 817 homeowners with SBA loans were not approved for additional
   reimbursements, the worksheet “Summary table dataset (817)” showed that 163 of such homeowners had a second
   reimbursement approved amount of $1,641,621. The remaining 122 homeowners did had no second
   reimbursement approved amounts.



                                                       19
A review of 29 CMS Program files related to homeowners who were reportedly not approved for
additional reimbursements showed that in two instances, homeowners were approved for
additional reimbursements on September 15, 2016, and September 27, 2016. However, the
reimbursement checks had not been issued. Given the recent approvals, we questioned whether
the data City officials provided represented the actual population of homeowners who would
ultimately receive additional reimbursements.

In addition, City officials provided an Excel spreadsheet showing 122 homeowners whose grant
amounts were not revised to show changes resulting from duplication of benefits recalculations
after their reimbursement grant agreements had been signed. The documentation in the Program
files did not agree with data provided for 2 homeowners in our sample of 16. For one
homeowner, City officials’ data showed an original maximum reimbursement amount of
$140,885. However, the reimbursement grant agreement in CMS showed a maximum
reimbursement amount of $183,867, for a difference of $42,982. For the second homeowner,
City officials’ data showed a new maximum reimbursement amount of $61,580, but the revised
coordination of benefits worksheet in CMS showed a new maximum reimbursement amount of
$49,432, for a difference of $12,148.
Duplication of Benefits Recalculation Discrepancies Were Not Always Documented
City officials did not always document discrepancies resulting from duplication of benefits
recalculations that caused grant amounts to change after homeowners had signed their
reimbursement grant agreements. Of the 16 Program files reviewed relating to homeowners
whose grant amounts changed by $500, 6 did not contain the revised duplication of benefits
calculations. Further, seven Program files did not contain the memorandum, “Changes to
Reimbursement Amount Post-Grant Agreement Signing,” dated January 29, 2015. This
memorandum was required to document why a new agreement was not executed to show a
change in the grant amount resulting from a duplication of benefits recalculation after a
reimbursement grant agreement had been signed. This deficiency occurred because City
officials did not establish adequate monitoring controls to ensure that the revised duplication of
benefits calculation and the memorandum were always maintained in the Program files in
accordance with their policies. Accordingly, without the revised duplication of benefits
calculations or the required memorandum, there would be no record of the change in the grant
amount or why the grant was not reprocessed or a new grant agreement was not executed to
show the change.
Conclusion
City officials did not maintain complete, accurate, and reliable Program files and records in
accordance with the City’s record-keeping requirements. The source documentation in the
Program files did not reconcile with the data provided by City officials. In addition, based on
our sample, City officials countersigned reimbursement grant agreements on average
approximately 115 days after CDBG-DR assistance had been provided, which was inconsistent
with prudent business practices. These record-keeping deficiencies occurred due to the lack of
adequate monitoring and quality control reviews to ensure that the Program files and records
were appropriately maintained to facilitate the review of overall Program compliance.



                                                 20
Recommendations
We recommend that HUD’s Acting Deputy Assistant Secretary for Grant Programs instruct City
officials to

      3A.    Complete the planned document and data cleanup process in CMS before file
             closeout to ensure that duplicative documents are archived, the most recent
             documents are identified and filed in the appropriate subfolders, and all files are
             auditable and comply with the requirements and the City’s record-keeping
             procedures.
      3B.    Ensure that all financial reports are accurate and agree with supporting
             documentation in the Program files.
      3C.    Update the Coordination of Benefits Standard Operating Procedures to include
             definitions of SBA cancellation codes.




                                               21
Scope and Methodology
The review generally covered the period January 29, 2013, through September 30, 2015, and was
expanded as necessary. Audit work was performed from November 2015 through June 2016 at
the City’s Office of Management and Budget located at 255 Greenwich Street, New York, NY,
and at our office located at 26 Federal Plaza, New York, NY.
To accomplish our objective, we

          Reviewed relevant CDBG-DR Program requirements and applicable Federal regulations,
           including the Disaster Relief Appropriations Act of 2013, Federal Register notices, HUD
           CPD notices, and HUD guidance.

          Reviewed the City’s HUD-approved action plan and applicable amendments.

          Reviewed the City’s written policies and procedures, including several versions of the
           Program policy manual and the record-keeping standard operating procedure.

          Met with City officials to obtain an understanding of the Program’s operations, system of
           internal controls, and CMS functionality.

          Reviewed HUD’s monitoring reports issued during the period December 2013 to July
           2015.

          Reviewed quarterly performance reports for the period October 2013 to September 2015
           generated from the Disaster Reporting Grant Recovery (DRGR) System19 for the purpose
           of obtaining background information on the City’s activities and disbursement of CDBG-
           DR funds only. We did not assess the data.

          Reviewed the August 2013 executed grant agreement and June 2015 amendment between
           HUD and the City.

As of September 30, 2015, City officials assisted 4,326 homeowners who received
approximately $82.8 million in reimbursement assistance. Of that population, we selected a
statistical sample of 14 homeowners who received $427,105 during the period March 2014 to
September 2015, and reviewed the Program files in CMS for documentation of Program
compliance, including compliance with HUD’s Lead Safe Housing Rule requirements.

19
     The Disaster Reporting Grant Recovery System was created by HUD’s Office of Community Planning and
     Development for the CDBG Disaster Recovery Program and other special appropriations. Grantees use the
     DRGR system to access grant funds and report accomplishments in quarterly performance reports. HUD officials
     use the DRGR system to review grant-funded activities, prepare reports to Congress, and monitor program
     compliance.



                                                          22
We also selected a nonstatistical sample of 16 of 122 homeowners whose grant amounts were
adjusted by $500 or more due to duplication of benefits recalculations, to determine the number
of grants that were overpaid or underpaid.
City officials provided an Excel spreadsheet showing 834 homeowners with SBA loans who
were projected to receive approximately $10.5 million in additional reimbursements. From that
spreadsheet, we selected a nonstatistical sample of approximately 10 percent, or 84 homeowners,
to determine their national objective determination.
Of the 834 homeowners with SBA loans, an analysis of the data using the Audit Command
Language Program20 identified eight homeowners who received an equivalent of 100 percent of
their eligible reimbursement amount. A comparison analysis of the data provided by City
officials to the source documentation in the Program files in CMS noted discrepancies in the
City’s data and determined that the eight homeowners received an equivalent of 60 and not 100
percent of their eligible reimbursement amount as reported and they were not eligible for
additional reimbursement.
Using the Audit Command Language Program, we performed an additional analysis of the 834
homeowners with SBA loans. Our analysis showed that 151 of such homeowners were projected
to receive additional reimbursements totaling approximately $2.5 million. This amount
exceeded the approximately $1.4 million in the SBA loans considered duplicative by
approximately $1.1 million. Of the 151 homeowners projected to receive additional
reimbursements,

                66 were projected to receive additional reimbursements that were more than the
                 amount of the SBA loans considered duplicative but less than all disbursed SBA
                 loans,21
                47 were projected to receive additional reimbursements that were equivalent to
                 the amount of disbursed SBA loans, and
                38 were projected to receive additional reimbursements that were equivalent to 40
                 percent of the eligible reimbursement amount.
We selected a nonstatistical sample of approximately 25 percent of the 66, 47, and 38
homeowners, namely, 17, 12, and 10 sample items, respectively, to determine whether source
documentation in the Program files in CMS confirmed the accuracy and the reliability of the
data. A review of the source documentation regarding the sample of 10 of 38 homeowners found
inaccuracies in the data. As a result, we expanded our sample to include 100 percent of the 38
homeowners. Of the 38 homeowners, the Program files clearly showed that 10 homeowners
were either not eligible to receive additional reimbursements or had no reimbursement amounts
for additional assistance. Further, in one example, a homeowner was projected to receive an
additional reimbursement that was $48,791 [$68,791 (original projection) - $20,000 (new


20
   The Audit Command Language Program is a software used to perform data analysis and audit tests enabling its
   users to identify fraud patterns and data irregularities.
21
   SBA offers three loan types: real estate (property damage), mitigation, and personal property loans. However,
   the program included only real estate loans in the duplication of benefits calculation.



                                                         23
projection)] less than the original projection, which was less than the amount of the SBA loan
considered duplicative.
Due to inaccuracies in City officials’ data, we determined that 140 of 151 homeowners with SBA
loans were projected to receive $2.2 million in additional reimbursements, which exceeded the
$1.2 million in SBA loans considered duplicative by approximately $1 million. Additionally,
City officials’ projection of additional reimbursements to homeowners with SBA loans was
reduced from approximately $10.5 million to $10.2 million, and the number of impacted
homeowners with SBA loans was reduced from 834 to 817.22
In response to the draft report, City officials provided an Excel spreadsheet containing an
updated analysis of reimbursement review activity related to the 817 homeowners with SBA
loans included in our initial review. They also provided an analysis of other homeowners with
SBA loans who were not a part of our review. The spreadsheet showed that City officials
disbursed $5.8 million to 549 homeowners with SBA loans. Of that amount, they drew $4.5
million in CDBG-DR funds as of September 20, 2016. Using the Audit Command Language
Program, we analyzed the data contained in the spreadsheet and found a number of
discrepancies, which are discussed in detail in findings 1 and 3 of this report.
We selected a nonstatistical sample of approximately 10 percent, or 29 of 285 homeowners who
City officials reported were not approved for additional reimbursement to determine their second
reimbursement eligibility status in CMS.
Based on the audit work performed, we concluded that the Program files in CMS and the data
City officials provided were not sufficiently reliable. Further, while we selected several
nonstatistical samples to accomplish our objectives, the results from these samples related only
to the items sampled and could not be projected to the population.
We conducted the audit in accordance with generally accepted government auditing standards.
Those standards require that we plan and perform the audit to obtain sufficient, appropriate
evidence to provide a reasonable basis for our findings and conclusions based on our audit
objective(s). We believe that the evidence obtained provides a reasonable basis for our findings
and conclusions based on our audit objective.




22
     The 817 equals 834 minus 8 homeowners who received 60 and not 100 percent of their eligible reimbursement
     amount, minus 10 homeowners, all of whom were either not eligible to receive additional reimbursements or had
     no reimbursement amounts for additional assistance. One homeowner appeared in both categories; thus, to avoid
     double counting, this homeowner was added back (834 - 8 - 10 = 816 +1 =817).



                                                          24
Internal Controls
Internal control is a process adopted by those charged with governance and management,
designed to provide reasonable assurance about the achievement of the organization’s mission,
goals, and objectives with regard to

   Effectiveness and efficiency of operations,
   Reliability of financial reporting, and
   Compliance with applicable laws and regulations.
Internal controls comprise the plans, policies, methods, and procedures used to meet the
organization’s mission, goals, and objectives. Internal controls include the processes and
procedures for planning, organizing, directing, and controlling Program operations as well as the
systems for measuring, reporting, and monitoring Program performance.

Relevant Internal Controls
We determined that the following internal controls were relevant to our audit objective:

   Program operations - Policies and procedures that management has implemented to
    reasonably ensure that a Program meets its objectives.
   Compliance with laws and regulations - Policies and procedures that management has
    implemented to reasonably ensure that resource use is consistent with laws and regulations.
   Validity and reliability of data - Policies and procedures that management has implemented
    to reasonably ensure that valid and reliable date are obtained, maintained, and fairly
    disclosed in reports.
   Safeguarding resources - Policies and procedures that management has implemented to
    reasonably ensure that resources are safeguarded against waste, loss, and misuse.
We assessed the relevant controls identified above.
A deficiency in internal control exists when the design or operation of a control does not allow
management or employees, in the normal course of performing their assigned functions, the
reasonable opportunity to prevent, detect, or correct (1) impairments to effectiveness or
efficiency of operations, (2) misstatements in financial or performance information, or (3)
violations of laws and regulations on a timely basis.
Significant Deficiencies
Based on our review, we believe that the following items are significant deficiencies:

   City officials did not implement policies and procedures to ensure that CDBG-DR funds
    were always disbursed in accordance with the HUD-approved action plan and Federal
    requirements (findings 1 and 2).



                                                  25
   City officials did not implement adequate controls to ensure compliance with laws and
    regulations since they did not maintain sufficient documentation to show that HUD’s Lead
    Safe Housing Rule requirements and the City’s record-keeping standards were met (findings
    2 and 3).
   City officials did not implement adequate controls to ensure the validity and reliability of
    data in the Case Management System with regard to lead testing and records regarding
    projected additional reimbursements and duplication of benefits recalculations as the data
    were not always accurate (findings 2 and 3).
   City officials did not implement adequate controls to ensure that funds were always
    safeguarded against fraud, waste, and abuse as CDBG-DR funds were used for grants that
    should have been adjusted to show revised duplication of benefits calculations. Further, City
    officials did not pursue grant recaptures against homeowners who refused to allow the
    Program to perform lead hazard compliance work, thus ensuring that CDBG-DR funds were
    used for lead-safe homes (findings 1 and 2).




                                                  26
Appendixes

Appendix A


          Schedule of Questioned Costs and Funds To Be Put to Better Use
        Recommendation                                    Funds to be put
                           Ineligible 1/ Unsupported 2/    to better use 3/
            number
                1A               $4,467,299                               $1,314,068
                1C                   32,107
                1D                                                           101,398
                2B                                    $182,660
                2C                                      833,199
                2E                                       29,019

              Totals             $4,499,406         $1,044,878            $1,415,466



1/   Ineligible costs are costs charged to a HUD-financed or HUD-insured Program or activity
     that the auditor believes are not allowable by law; contract; or Federal, State, or local
     policies or regulations.
2/   Unsupported costs are those costs charged to a HUD-financed or HUD-insured Program
     or activity when we cannot determine eligibility at the time of the audit. Unsupported
     costs require a decision by HUD Program officials. This decision, in addition to
     obtaining supporting documentation, might involve a legal interpretation or clarification
     of departmental policies and procedures.
3/   Recommendations that funds be put to better use are estimates of amounts that could be
     used more efficiently if an Office of Inspector General (OIG) recommendation is
     implemented. These amounts include reductions in outlays, deobligation of funds,
     withdrawal of interest, costs not incurred by implementing recommended improvements,
     avoidance of unnecessary expenditures noted in preaward reviews, and any other savings
     that are specifically identified. We considered $1.3 million in additional CDBG-DR
     assistance yet to be to be drawn as funds that could be put to their intended use once City
     officials submits and HUD approves the amended action plan that would allow
     homeowners with SBA loans to receive additional reimbursements covering from 60 to
     100 percent of their eligible reimbursable expenses. Further, 11 homeowners are owed



                                              27
$101,398 in additional assistance due to changes in their grant amounts resulting from
duplication of benefits recalculations.




                                        28
Appendix B
             Auditee Comments and OIG’s Evaluation



Ref to OIG    Auditee Comments
Evaluation




                               29
             Auditee Comments and OIG’s Evaluation



Ref to OIG
              Auditee Comments
Evaluation




Comment 1


Comment 2


Comment 3




                               30
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 3


Comment 4




Comment 2




                               31
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 2




Comment 3




Comment 5




                               32
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 5




Comment 6




Comment 7




                               33
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 7




Comment 8




Comment 9


Comment 10




                               34
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 10




Comment 9




Comment 9




                               35
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 9




Comment 9




                               36
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 11


Comment 12




Comment 13




Comment 14




                               37
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 14




                               38
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 14




Comment 15


Comment 16


Comment 17




                               39
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 18




Comment 19




Comment 20




                               40
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 20




Comment 21




Comment 22




                               41
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 23




Comment 24




                               42
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 24


Comment 3




Comment 25




                               43
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 25




Comment 26




                               44
             Auditee Comments and OIG’s Evaluation




Ref to OIG    Auditee Comments
Evaluation




Comment 26




                               45
                         OIG Evaluation of Auditee Comments


Comment 1   We disagree with City officials’ statement that we questioned their decision to
            prioritize financial relief. We questioned City officials’ lack of documentation to
            show compliance with HUD’s Lead Safe Housing Rule requirements. The CMS
            Program files reviewed did not clearly document whether assisted homes were
            exempt from lead requirements. The files lacked documentation showing that
            lead-based paint testing had been performed, identified hazards had been
            removed, and clearance had been achieved. Further, City officials did not have a
            plan to ensure lead hazard compliance when homeowner did not allow the
            Program to complete the required lead hazard work.
Comment 2   City officials stated that they made the policy decision to provide additional
            reimbursement to SBA loan recipients in response to extensive public feedback
            regarding inequities identified in the treatment of SBA loans in comparison with
            other disaster recovery benefits. By providing additional assistance solely to SBA
            loan recipients, City officials attempted to correct fundamental complaints with
            Federal disaster assistance programs, namely that homeowners were unaware that
            SBA loans replaced CDBG-DR assistance, although they had to be repaid and
            CDBG-DR assistance could not be used to repay SBA loans. City officials
            believed that these complaints should be addressed by Federal partners involved
            in providing disaster assistance to residential property owners.
            City officials’ commitment to address what they considered to be unfairness in the
            treatment of SBA loans in comparison with other disaster recovery benefits is
            acknowledged. However, the implementation of their policy decision to provide
            additional reimbursement covering up to 100 percent of eligible expenses incurred
            by homeowners who received SBA loans was inconsistent with the HUD-
            approved action plan. The action plan stated that the Program would provide
            assistance that covered a portion of eligible reimbursable expenses. Further, this
            policy decision primarily benefited higher income homeowners who had the
            necessary income and credit to qualify, obtain, and repay their SBA loans.
Comment 3   City officials stated that the issues in this audit report must be considered to be
            preliminary because at the time of our review, much of the data provided did not
            show actual beneficiaries since no funds had been disbursed and drawn. With the
            exception of the data related to additional reimbursements approved for SBA loan
            recipients, all data provided during the review related to actual beneficiaries. We
            have revised the audit report to include the actual number of beneficiaries and the
            total amount of additional reimbursement paid as September 20, 2016, the date on
            which the updated data were provided. Other than the revised figures, the issues
            in the audit report remain unchanged.
            Our initial data request was made on January 29, 2016. In an email, dated
            February 11, 2016, City officials indicated that they were working on the list and
            expected to have it completed by February 16, 2016. The email also stated that


                                              46
            that the applicants on the list were not guaranteed to receive an additional
            payment until they were reviewed for eligibility, which would take another
            month. Based on the email, eligibility reviews should have been completed in
            March 2016. More than 2½ months later, City officials provided the data attached
            to an email, dated May 2, 2016. The email stated that the list was current as of
            April 29, 2016, and was subject to change based on homeowner eligibility
            reviews. However, by May, it was reasonable to believe that most of the
            eligibility reviews had been completed since they were supposed to be completed
            in March. Neither this nor later emails indicated that there would be substantial
            changes in the actual number of beneficiaries and the total amount of additional
            reimbursements issued. However, the actual number of beneficiaries is
            approximately 33 percent less than City officials’ initial projections, and the total
            amount of additional reimbursements issued is approximately 44 percent less than
            City officials’ initial projections. We question the accuracy and reliability of City
            officials’ data, given the substantial differences between the actual and projected
            amounts.
Comment 4   City officials maintained that at the time of our review, they were aware of the
            application-specific issues addressed in the audit report but had yet to resolve
            some of the issues because of the order in which they process applications. City
            officials believed that during the Program’s closeout phase, they would resolve
            the issues identified in this audit report. Therefore, their planned actions are
            responsive to our recommendations.
Comment 5   City officials stated that they revised the cap for second reimbursement payments
            from the maximum amount of SBA loans received for structural repairs to the
            total amount drawn from all SBA loans because the latter represented the
            homeowner’s true Sandy-related debt obligations. However, the policy of
            capping second reimbursements payments at the total amount drawn from all
            SBA loans resulted in higher benefit amounts for SBA loan recipients only, which
            was inconsistent with the HUD-approved action plan.
Comment 6   City officials stated that they were modifying the language in the action plan to
            ensure that the action plan would be consistent with the Program’s
            implementation regarding providing additional reimbursement of up to 100
            percent of reimbursable expenses to homeowners with SBA loans. Thus, City
            officials’ planned actions are responsive to our recommendations.
Comment 7   City officials stated that their closeout process designed to detect benefit changes
            was effective, given that only 10 of 122 homeowners, representing 8 percent of
            the total population, had a change in their benefit calculation after the initial grant
            amount had been calculated. However, our audit report identified 16 homeowners
            whose grant amount changed by $500 or more due to changes in their benefit
            calculation. While we chose to report on 16 homeowners only, we identified 25
            homeowners whose grant amount changed by more than $100. This represented
            approximately 20 percent of the total population.



                                               47
Comment 8     City officials’ development of a robust closeout process to detect and resolve
              changes to benefit amounts that may affect a homeowner’s eligible grant award is
              responsive to our recommendations.
Comment 9     City officials’ planned review of the reimbursement to determine whether grant
              recapture is necessary is responsive to our recommendation.
Comment 10 City officials stated that while four homeowners had grant awards that were less
           than the new maximum reimbursement amount, grant recaptures were not
           necessary because the grants fell within the 60 to 100 percent reimbursement
           range. However, these four homeowners were reimbursed in excess of the
           Program’s 60 percent reimbursement rate and received additional benefits that
           other homeowners were not given. Further, both the coordination of benefits
           worksheet disclosure and section 13(a)(iii) of the reimbursement grant agreement
           stated that a homeowner might be required to repay the Program additional
           disaster recovery funds received from other sources that were considered to be a
           duplication of benefits after the grant had been awarded.
Comment 11 City officials stated that one of the four homeowners was eligible for a second
           reimbursement payment and any overpayment would be removed from the second
           reimbursement payment. Second reimbursements to SBA loan recipients are not
           authorized by the HUD-approved action plan. Therefore, in the absence of an
           approved amended action plan, this amount should be repaid to the Program.
Comment 12 City officials believed that we used either incorrect fields or an incorrect
           methodology when calculating the homeowners’ eligible reimbursement amount
           in six of the cited cases. The remaining five cases had revisions that increased the
           homeowners’ maximum reimbursement amount. The data used to calculate the
           underpayments were obtained from a list City officials provided of 122
           homeowners whose grant amount changed due to duplication of benefits
           recalculations. We applied the Program’s 60 percent reimbursement rate to the
           original and new maximum reimbursement amounts and calculated the difference
           or the underpayment. We also reviewed the grant agreements, the original and
           revised coordination of benefits worksheets (if available), and the reimbursement
           review associated view in the CMS Program files. Accordingly, two
           discrepancies were noted between the data provided and the Program files. The
           discrepancies are discussed in detail in finding 3 of this report.
              City officials’ planned review of the applications during the closeout process to
              determine the appropriate action that should be taken is responsive to our
              recommendations.
Comment 13 City officials cited cost concerns and the administrative burden associated with
           executing a new grant agreement as reasons why they did not believe that they
           should be required to execute a new grant agreement when a homeowner’s actual
           grant payment was less than what was shown on the homeowner’s reimbursement
           grant agreement. However, by not executing new grant agreements, City officials


                                                48
              deprived homeowners of valuable financial resources that they could have used to
              assist them in their recovery efforts. Further, under the reimbursement program,
              homeowners were eligible to receive 60 percent of their reimbursable expenses.
              Yet, due to City officials’ policy decision, they received much less than they were
              eligible to receive and should be compensated.
              City officials also stated that the reimbursement grant agreement accounted for
              grant changes within the 40 percent range, which made up the unreimbursed
              amount. However, the Program defined the reimbursement rate as 60 percent of
              eligible reimbursable expenses. By providing homeowners with as much as 110
              percent of their eligible reimbursable expenses, while providing others with as
              little as 5 percent of their eligible reimbursable expenses, City officials did not
              reimburse all eligible homeowners equitably.
Comment 14 City officials believed that the Program’s policy to reimburse applicants before
           lead testing or abatement was completed complied with the regulations because
           the Program provided assistance to homeowners after they agreed to allow the
           Program to address all lead hazards during construction. Further, City officials
           asserted that although some homeowners withdrew from the Program before
           required construction activities began, City officials would address all incomplete
           projects during its closeout process. At the time of our review, the CMS Program
           files did not clearly document City officials’ plan to address compliance with
           HUD’s Lead Safe Housing Rule requirements postreimbursement during later
           construction. However, City officials’ planned action is responsive to our
           recommendations.
Comment 15 City officials stated that they reviewed the first identified application and found
           that CMS correctly showed the results of the Program’s risk assessment and that
           the risk assessment was stored in CMS’ “Inactive Environmental Analysis
           Entity,” a location all relevant CMS users are trained to use when a homeowner’s
           pathway changes. This issue was first communicated to City officials in our
           February 12, 2016, tentative observations. At that time, City officials provided a
           general response concerning the Program’s lead compliance. However, based on
           City officials’ current response, we were able to locate the lead testing report
           showing the negative test results. Accordingly, we have removed all references to
           this issue from the audit report and have reduced questioned costs by $45,940, the
           dollar value associated with this issue.
Comment 16 City officials’ continual attempts to ensure that the applicant completes lead-
           based paint testing is responsive to our recommendations.
Comment 17 City officials stated that two of the four homes had completed construction with
           successful lead-based paint abatement. However, we were unable to locate the
           lead clearance reports in the CMS Program files. Further, City officials stated
           that the third home was in the construction phase and the fourth home was in the
           preconstruction phase and the lead abatement work for each home was included in
           the respective projects’ scope of work. When construction on these two homes is


                                                49
              complete, City officials must document the lead clearance reports for these and all
              homes that have completed lead-based paint abatement.
Comment 18 City officials stated that recommending repayment for the 41 homeowners who
           withdrew from the Program before lead hazard activities were completed was
           premature since the City had not completed its comprehensive closeout process.
           Recommendation 2C of this report recommends repayment only if supporting
           documentation is not provided to show that lead-based paint testing was
           performed, identified hazards were removed, and clearance was achieved. City
           officials’ acknowledgement that grant recapture will be required when applicants
           do not achieve clearance during closeout is responsive to our recommendations.
Comment 19 City officials stated that only one risk assessment was performed on February 8,
           2014. Our audit report stated that there was a discrepancy in the lead test results.
           The performance of two risk assessments was not discussed. The risk assessment
           was performed on February 18, 2014, as indicated on The NYC Build it Back
           Repair Program Lead-Based Paint Risk Assessment Summary Report and the
           actual lead testing report, which was dated March 12, 2014. While the risk
           assessment stated that the home did not test positive for lead-based paint or
           hazard, the lead testing report stated that there was a potential lead exposure
           hazard at the home and lead hazard reduction activities were required. City
           officials asserted that this was boilerplate language and that the amount of lead
           dust detected was below the hazard threshold. City officials will have the
           opportunity to work with HUD officials to resolve this issue during the audit
           resolution process.
Comment 20 City officials’ comprehensive document and data cleanup process is responsive to
           our recommendations.
Comment 21 We have revised the audit report based on City officials’ assertion that
           countersigned grant agreements are not legally required and when these
           agreements are countersigned has no bearing on their enforcement. However,
           countersigning grant agreements as much as 336 days after reimbursement
           assistance had been provided is inconsistent with prudent business practices.
Comment 22 City officials’ planned action to include definitions of SBA cancellation codes in
           futures versions of their standard operating procedures is responsive to our
           finding. Although the draft audit report did not specifically recommend that City
           officials include this information in their procedures, based on City officials’
           comments, we have added this recommendation (3C) to the audit report. Defining
           SBA cancellation codes is important because the reason why an SBA loan was
           canceled affected whether the SBA loan amount was included or excluded from a
           homeowner’s duplications of benefits calculation. While City officials’ standard
           operating procedures provided instructions on how to process applications based
           on the cancellation codes, the procedures did not explain why the loans were to be
           processed in a certain manner.



                                               50
Comment 23 City officials stated that the offsets were itemized in the Coordination of Benefits
           section of each applicant file in CMS. However, we could not trace the offsets to
           the corresponding receipts related to four applicants because the receipts did not
           clearly identify the items eligible for offset credit. This matter was first
           communicated to City officials in our February 12, 2016, tentative observations.
           In response to our observations, City officials provided four itemized listings of
           receipts to facilitate our review.
Comment 24 City officials stated that we could have obtained additional detail concerning an
           expense by clicking on the individual line item since the additional detail was
           captured in CMS’ “Notes” section. During a meeting with City officials held on
           February 4, 2016, we informed them that we did not have access to the “Notes”
           section in CMS and such access was not provided during our review.
Comment 25 City officials stated that none of the six applications referenced in the audit report
           required an updated coordination of benefits worksheet and this issue resulted
           from our misunderstanding of the Program’s procedures. However, the data used
           in our analysis were obtained from a list City officials provided. The list
           consisted of 122 homeowners whose grant amount changed due to duplication of
           benefits recalculations after the homeowners had signed their grant agreements.
           From that list, we selected a sample of 16 homeowners whose grant amounts
           changed by $500 or more. A review of the CMS Program files showed that 6 of
           16 files did not contain a revised coordination of benefits worksheet. City
           officials will have the opportunity to work with HUD officials to resolve this
           issue during the audit resolution process.
Comment 26 City officials acknowledged that the “Changes to Reimbursement Amount Post-
           Grant Agreement Signing” memorandum was required in one of the seven files
           cited in the audit report. Further, City officials stated that they had corrected the
           deficiencies by uploading the missing memorandum to the file. For the remaining
           six files, City officials provided various reasons why the memorandum was not
           required. The data used in our analysis were obtained from a list City officials
           provided of 122 homeowners whose grant amount changed due to duplication of
           benefits recalculations after the homeowners had signed their grant agreements.
           From that list, we selected a sample of 16 homeowners whose grant amounts
           changed by $500 or more. A review of the CMS Program files showed that 7 of
           16 files did not contain the required memorandum. City officials will have the
           opportunity to work with HUD officials to resolve this issue during the audit
           resolution process.




                                                51