Debt Collection: Treasury Faces Challenges in Implementing Its
Cross-Servicing Initiative (Testimony, 06/08/2000, GAO/T-AIMD-00-213).
Pursuant to a congressional request, GAO discussed the Department of the
Treasury's progress in implementing the cross-service provision of the
Debt Collection Improvement Act (DCIA) of 1996, focusing on the: (1)
status of nontax delinquent debts that federal agencies have referred to
the Treasury for cross-servicing and the Treasury's actions to encourage
these referrals; (2) Treasury's cross-servicing process for collecting
referred debts; (3) Treasury's method for allocating debts to private
collection agencies (PCA) for collection; and (4) Treasury's estimated
cross-servicing costs and related fees earned on collections.
GAO noted that: (1) several agencies' reporting of debt balances and
related aging was not accurate, and the accuracy and completeness of
significant amounts reported as exclusions from cross-servicing were not
required to be and were not independently verified; (2) for various
reasons, many debts eligible for referral by certain agencies were
delayed in being referred or simply not referred even though the
Financial Management Service (FMS) took steps to encourage agencies to
refer such debt; (3) in addition, even when agencies referred debts, the
debts were not always valid and legally enforceable and thus not
eligible for cross-servicing; (4) DCIA authorize Treasury to designate
other government agencies as debt collection centers based on their
performance in collecting delinquent claims owed to the government; (5)
Tresury established standards for agencies that wanted to be a debt
collection center; (6) three agencies have applied to be governmentwide
debt collection centers, but were not found by Treasury to have the
needed capabilities; (7) only FMS is operating a governmentwide
cross-servicing debt collection center; (8) FMS developed a methodology
for distributing debts to PCAs for collection that FMS intended to be
performance based; (9) for each distribution, FMS placed all the debts
available into a pool and applied a systematic process to distribute the
debts to PCAs; (10) GAO's analysis of the debts found that the debts
within each distribution's pool were generally not of the same
composition; (11) GAO's analysis of FMS' distribution of debt accounts
to PCA form February 1998 through February 2000 showed that one PCA had
received a significantly higher percentage of the debts with smaller
balances; (12) concerns relating to FMS' distribution method have been
raised by some of the PCAs; (13) FMS has not covered its cross-servicing
costs through related fees collected and is not likely to in the near
future; and (14) based on FMS' own estimated cross-servicing costs and
using the current fee structure and FMS' fiscal year 1999 collection
experience, GAO determined that collection volume would need to rise
over sevenfold to put this operation on a full cost-recovery basis;
--------------------------- Indexing Terms -----------------------------
REPORTNUM: T-AIMD-00-213
TITLE: Debt Collection: Treasury Faces Challenges in Implementing
Its Cross-Servicing Initiative
DATE: 06/08/2000
SUBJECT: Financial management
Debt collection
Debt
Interagency relations
Government collections
IDENTIFIER: FMS, Payments, Claims, and Enhanced Reconciliation System
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Testimony. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO/T-AIMD-00-213
* For Release on Delivery
Expected at
10 a.m.
Thursday,
June 8, 2000
GAO/T-AIMD-00-213
debt collection
Treasury Faces Challenges in Implementing Its Cross-Servicing Initiative
Statement of Gary T. Engel
Associate Director, Governmentwide Accounting and Financial Management
Issues
Accounting and Information Management Division
Testimony
Before the Subcommittee on Government Management, Information and
Technology, Committee on Government Reform, House of Representatives
United States General Accounting Office
GAO
Mr. Chairman and Members of the Subcommittee:
Thank you for the opportunity to appear before your subcommittee today to
testify on the Department of the Treasury's progress in implementing the
cross-servicing provision of the Debt Collection Improvement Act (DCIA) of
1996. As you know, the Director of the Office of Management and Budget (OMB)
designated the implementation of this legislation, which your subcommittee
was highly instrumental in passing, one of the Priority Management
Objectives in the government's efforts to modernize and improve federal
financial management.
DCIA includes several tools to facilitate collection of defaulted
obligations to the federal government. Today, we are focusing on the
collection of nontax delinquent debt. Among the options available for
recovering these debts are (1) Treasury's consolidated federal payment
offset program, which the Financial Management Service (FMS) reported
collected over $2.6 billion in federal nontax debts and state child support
debts in fiscal year 1999, (2) wage garnishment, for which Treasury has
issued a final rule and is in the process of implementing, and (3) the
transfer of nontax debt over 180 days delinquent to Treasury for collection
action, known as "cross-servicing." For this hearing, you asked us to
address, the effectiveness of Treasury's use of the latter tool,
cross-servicing, through its FMS. FMS' success in implementing its
cross-servicing program, which focuses on debts that federal agencies have
been unable to collect, significantly depends on federal agencies accurately
and completely identifying their nontax delinquent debt that is eligible for
referral to the program and promptly referring such debt.
As you requested, I will discuss (1) the status of nontax delinquent debts
that agencies have referred to Treasury for cross-servicing and Treasury's
actions to encourage these referrals, (2) Treasury's cross-servicing process
for collecting referred debts, (3) Treasury's method for allocating debts to
private collection agencies (PCA) for collection, and (4) Treasury's
estimated cross-servicing costs and related fees earned on collections.
FMS has taken several steps to encourage agencies to refer eligible debt and
increase collections. However, the results thus far have been limited partly
due to much of the eligible debt not being promptly referred and the age of
the debts referred generally being significantly older than 180 days
delinquent. For example, our analysis of debts referred since the inception
of the program though May 1999 showed that almost one half of the dollar
value of the debts referred were over 4 years delinquent at the time of
referral. FMS reported that approximately $46.4 billion of debts were
delinquent over 180 days as of September 30, 1998. However, primarily due to
a significant amount of these debts being reported by the agencies as
excluded from cross-servicing requirements, through April 2000, FMS reported
only about $3.7 billion has been cumulatively referred to it since the
cross-servicing program began in September 1996. From the inception of the
program through April 2000, FMS reported that about $54 million has been
collected by its collectors and the PCAs on these referred debts.
We identified the following key issues related to the implementation of the
cross-servicing provisions of DCIA.
* Several agencies' reporting of debt balances and related aging was not
accurate, and the accuracy and completeness of significant amounts
reported as exclusions from cross-servicing were not required to be and
were not independently verified. For various reasons, many debts
eligible for referral by certain agencies were delayed in being
referred or simply not referred even though FMS took steps to encourage
agencies to refer such debt. In addition, even when agencies referred
debts, the debts were not always valid and legally enforceable and thus
not eligible for cross-servicing.
* DCIA authorized Treasury to designate other government agencies as debt
collection centers based on their performance in collecting delinquent
claims owed to the government. Treasury established standards for
agencies that wanted to be a debt collection center. The Departments of
Education and Health and Human Services were granted waivers by
Treasury to the cross-servicing provision of DCIA, which allows these
agencies to take collection action on certain classes of their own
debts. Three agencies have applied to be governmentwide debt collection
centers, but were not found by Treasury to have the needed
capabilities. Today, only FMS is operating a governmentwide
cross-servicing debt collection center. In operating its center, we
found that FMS had well-developed standard operating procedures (SOP),
however our testing showed that its staff did not always follow them or
properly use certain collection tools, such as skiptracing activities
to locate the debtor. For example, for 96 of 200 debts we statistically
selected and reviewed, other than the initial issuance of demand
letters, we found no evidence that FMS' collectors tried to contact the
debtors, as required by the SOP.
FMS recently changed many of the SOP's earlier requirements to perform
various collection techniques from "will" be performed to "may" or "should"
be performed. In addition, in 1999 FMS changed its SOP to reduce the 50-day
holding period to 30 days for performing cross-servicing procedures before
referring the debts to a PCA. These actions and our discussions with FMS
officials indicate that FMS is placing increased reliance on PCAs. However,
FMS has not performed an analysis to determine the potential effect such
reliance may have on net collections to the federal government. Such an
analysis may be warranted given that
(1) as debts are not actively worked by FMS and are awaiting referral to
PCAs, they continue to age and thus typically become more difficult to
collect and (2) the federal government pays a 25 percent fee on debt amounts
collected by the PCA that the government is not always able to recoup from
the debtor.
* FMS developed a methodology for distributing debts to PCAs for
collection that FMS intended to be performance based. For each
distribution, FMS placed all the debts available into a pool and
applied a systematic process to distribute the debts to the PCAs. Our
analysis of the debts found that the debts within each distribution's
pool were generally not of the same composition (i.e., not of the same
debt balance or age of delinquency). This factor contributed to the
distribution results experienced by FMS. Our analysis of FMS'
distribution of debt accounts to PCAs from February 1998 through
February 2000 showed that one PCA had received a significantly higher
percentage of the debts with smaller balances. Specifically, debts
distributed to this PCA had average balances of $11,436, while the
overall average balances of debt accounts distributed to PCAs were
$20,845. In addition, in many of the age of delinquency categories
(i.e., less than 180 days, 180 days to 1 year, etc.) this PCA had the
smallest average debt balances and had received a significant
percentage of the total number of debts distributed that were less than
1 year delinquent. Collection industry statistics as well as FMS'
collection experience to date have shown that collection rates are
generally higher on debts with smaller dollar balances and that are
less delinquent.
Concerns relating to FMS' distribution method have been raised by some of
the PCAs. During our interviews with the 11 PCAs, we found that the general
consensus among them when asked how the debts should be distributed was that
the distribution should take into consideration the characteristics of the
debts, such as age of delinquency, type of debt, agency referring the debt,
and debt balance. Many of the PCAs indicated that stratifying the available
debts by agreed-upon characteristics would result in each of the PCAs
receiving a proportionate mix of the debts and foster a more competitive
environment.
* FMS has not covered its cross-servicing costs through related fees
collected and is not likely to in the near future. Based on FMS' own
estimated cross-servicing costs and using the current fee structure and
FMS' fiscal year 1999 collection experience, we determined that
collection volume would need to rise over sevenfold to put this
operation on a full cost-recovery basis.
We performed our work primarily at FMS and its Birmingham Debt Management
Operations Center (BDMOC). We conducted interviews with FMS officials and
representatives of FMS' eleven PCAs and the American Collectors Association
and reviewed pertinent policies, procedures, databases, and reports related
to cross-servicing. We also statistically selected and performed detailed
testing on certain debts that had been referred for cross-servicing from
April 1, 1998 through May 31, 1999. In addition, we analyzed FMS'
methodology for distributing debts to PCAs and reviewed certain FMS
cross-servicing fee and estimated cost data for fiscal year 1999. We did not
independently verify the reliability of certain information provided to us
by FMS (e.g., estimated costs, debts eligible for cross-servicing, total
debts referred for cross-servicing, and information in FMS' debt referral
databases). We performed our work in accordance with generally accepted
government auditing standards from April 1999 through May 2000.
In the rest of my statement today, I will discuss the results of our work
and highlight challenges that FMS faces in implementing a viable
cross-servicing operation.
Referral of Federal Debts for Cross-Servicing
DCIA requires agencies to refer all eligible nontax debt that is over 180
days delinquent to FMS for cross-servicing. FMS reported that at September
30, 1998, federal agencies held $46.4 billion of debt over 180 days
delinquent. Based on information obtained from the 24 agencies covered by
the Chief Financial Officers Act of 1990 (CFO) and FMS' own estimates for
non-CFO Act agencies, FMS reported that about 85 percent, or $39.6 billion,
of the debt as of September 30, 1998, was not eligible for referral to the
cross-servicing program because of various exclusions, such as foreclosures
and bankruptcies.
Our analysis showed that the debts agencies refer to FMS are generally well
over 180 days delinquent. Further, we noted problems in the reporting of
delinquent debt balances and related aging by certain agencies. We also
identified problems with FMS reports on the status of delinquent debts
governmentwide.
FMS reported that as of April 2000, about $3.7 billion of the approximately
$6.4 billion of eligible debt had been referred for cross-servicing. Because
the eligible amount is as of a specific date and the amount reported as
referred is a cumulative amount covering about 3-1/2 years, these two
amounts are not comparable. In addition, we found that agency-referred debts
were not always valid and legally enforceable and thus not eligible for
cross-servicing.
These reporting problems, coupled with the lack of independent verification
of the completeness and accuracy of debt exclusion amounts, make the
reliability of these reported amounts questionable. Lack of reliable
identification and prompt referrals of eligible debts by the agencies to FMS
is likely to result in lost opportunities for the government to recover
amounts owed.
Age of Debts Referred
Agencies are not promptly referring debts as soon as they are eligible for
cross-servicing. Figure 1 shows our analysis of the debts referred since the
inception of the program through May 1999. This analysis shows that about
$1.1 billion (or about 46 percent) of the $2.4 billion of debt referred
during this period was over 4 years delinquent at the time it was referred
to FMS for cross-servicing.
Figure 1: Dollar Amount of Debt by Age of Delinquency
A critical factor in FMS' success as a debt collection center is that all
debts eligible for cross-servicing are completely and accurately identified.
In addition, once identified by the agencies, debts need to be promptly
referred for cross-servicing because, as industry statistics have shown, the
likelihood of recovering amounts owed decreases dramatically with the age of
delinquency of the debt. Thus, the old adage that "time is money" is very
relevant for this effort.
Using cross-servicing collection rates for delinquent debt obtained from a
1998 study conducted by a FMS contractor, we estimated collections on debts
totaling about $1.8 billion that were referred from the inception of the
program through May 1999. Our analysis showed that estimated collections on
these debts ranged from about $40 million to $75 million. Based on our
review of FMS' collections database, we determined that FMS collected about
$27 million on debts that were referred for cross-servicing during this same
time period.
Figure 2 represents a timeline of the standard process involved in the
referral of debts to FMS and subsequently, as applicable, to PCAs. In
effect, this figure represents the minimum timelines for referral and
collection efforts. Accordingly, it reflects the optimum scenario for
cross-serviced debt, not what is actually taking place, as reflected in part
by figure 1. For example, as noted in figure 1, many debts are much older
than 180 days delinquent when they are referred to FMS. Also, as discussed
later in this testimony, we identified delays throughout much of the
process.
Figure 2: Declining Recovery Rates by Age of Delinquency
Source: GAO's analysis of FMS' collection processes and testimony of the
American Collector's Association before the Subcommittee on Government
Management, Information and Technology of the Committee on Government Reform
and Oversight, September 8, 1995.
Exclusions From Cross-Servicing and Accuracy of Reporting
To help monitor the extent to which agencies are referring eligible debts to
FMS for cross-servicing, FMS developed and implemented the Debt Performance
Indicator (DPI) report for each CFO Act agency. According to an FMS
official, for fiscal year 1998, FMS obtained the information on debts
excluded from cross-servicing from DPI reports submitted by agencies and
discussions with agency officials. FMS used the agencies' Report on
Receivables, DPI reports, and FMS estimates to compile the Summary Analysis
of Delinquent Debt for the Federal Government, a government-wide report. FMS
reported that about 85 percent of the debt or $39.6 billion was not eligible
for referral to FMS because of various exclusions (see table 1).
Table 1: Nontax Debt Eligible for Referral for Cross-Servicing as of
September 30, 1998a,
Amount
Characteristics of debt (in billions
of dollars)
Debt over 180 days delinquent $46.4
Exclusions:
Cross-servicing waivers 16.6
In forbearance or in appeals 5.1
At DOJ 4.9
Foreign debt 3.6
In bankruptcy 3.1
In foreclosure 2.4
Department of Defense 1.3
Eligible for internal offset 1.0
Other 0.9
At third party 0.7
Total amount excluded $39.6
Amount eligible to refer for
cross-servicing $6.8
aAn explanation of the terms used in this table appears in appendix I.
Source: Summary Analysis of Delinquent Debt for the Federal Government, Debt
Portfolio Analysis for the 24 CFO Agencies, June 1999, and other FMS
reports.
The reliability of amounts reported as excluded from cross-servicing by the
agencies has not been independently verified. According to FMS officials,
agencies were not required to certify that all information provided to FMS
was complete and accurate. Further, these agencies' respective Office of
Inspector General (OIG) were not required to and did not audit most of the
information, including the exclusion amounts.
We identified problems with FMS' estimates of exclusions for non-CFO Act
agencies. In preparing the fiscal year 1998 Summary Analysis report, FMS
generally estimated the amount of debts that would be excluded from
cross-servicing for non-CFO agencies using the CFO Act agencies' aggregate
percentages by type of exclusion. FMS estimated exclusions for the non-CFO
Act agencies because these agencies were not required to report such
information. FMS estimated that of the $39.6 billion of excluded debt in
table 1, about $3.6 billion (approximately 9 percent) was attributed to the
non-CFO Act agencies. We found that FMS' estimation of exclusion amounts was
not reliable. For example, FMS estimated the amount of cross-servicing
waivers for non-CFO Act agencies to be about $1.5 billion. Since we found
that none of these agencies had applied to FMS for a waiver, the amounts
reported as exclusions for cross-servicing waivers were overstated.
On the other hand, FMS' exclusion estimate for bankruptcies for non-CFO Act
agencies is understated. Our review of the Report on Receivables from the
Federal Communications Commission (FCC), a non-CFO Act agency, found that
FCC had $2.3 billion in bankruptcies for debts delinquent over 180 days as
of September 30, 1998. As such, FCC's bankruptcy amounts alone are
considerably more than the $300 million bankruptcy amount FMS estimated for
all non-CFO Act agencies.
Compounding these concerns are questions concerning the accuracy of the
underlying agency reports that FMS uses as the basis for its reports. In
December 1999, the President's Council on Integrity and Efficiency (PCIE),
the Executive Council on Integrity and Efficiency (ECIE), and Treasury's OIG
issued a report titled, PCIE/ECIE Review of Non-Tax Delinquent Debt. Among
other findings, the report stated that of the 16 agencies reviewed, 5 had
inaccurate accounts receivable balances as of the end of fiscal year 1998,
and 3 agencies did not accurately age their accounts receivable. For
example,
* The Department of State OIG found that the accounting system at the
department did not produce a reliable accounts receivable aging
schedule.
* In reporting on its audit of the fiscal year 1998 Veterans Affairs (VA)
Consolidated Financial Statements, the VA OIG qualified its opinion on
material amounts of accounts and loans receivable due to the inadequacy
of supporting accounting records. Specifically, of the total net debt
and foreclosed property of $4.7 billion, the OIG qualified its opinion
on the accounts relating to the Housing Credit Assistance Program,
which comprised $3 billion of the total balance. Further, the Veterans
Health Administration's receivable balance of $440 million was
overstated by
$65 million. These inaccurate balances resulted because VA did not
consistently follow its accounting procedures and certain of its
internal controls were ineffective.
FMS officials recognize the problems with the manner in which the exclusions
and eligible debt amounts for cross-servicing were identified. For example,
FMS officials stated that for fiscal year 1999 FMS is using the Revised
Report on Receivables to determine debts eligible for cross-servicing for
all agencies, including non-CFO Act agencies. The revised report includes
the various exclusion categories which agencies will be required to use to
report debt amounts excluded from cross-servicing. According to these
officials, FMS plans to require agencies to certify as to the accuracy and
completeness of the amounts that they report as excluded from
cross-servicing, however, agencies' respective OIGs do not currently
independently verify such amounts. Lack of such verification, along with the
problems noted in the PCIE/ECIE report regarding inaccurate balances and
aging of accounts receivable, raises concerns about the extent to which FMS
can rely on agencies' reporting of 180-day delinquent debt and exclusions of
debts from the cross-servicing program.
Factors Affecting Agencies' Debt Referrals
* Agencies focused their computer programming resources on Year 2000
problems - a decision with which we agree - rather than on
cross-servicing systems requirements, such as computer systems'
compatibility, so that debt information can be transmitted to FMS
electronically.
* Certain agencies had to perform detailed and time-consuming due
diligence reviews of the files to identify debts eligible for
cross-servicing because such information was not readily available.
* Some agencies delayed referring debts while waiting for FMS to decide
whether the agency's request to be designated a debt collection center
was approved. Such decisions were to be rendered within 120 days, but
processing time ranged from 10 months to 19 months.
Some specific cases cited in the PCIE/ECIE report are as follows.
* Treasury's OIG selected and reviewed 10 debt case files at Treasury's
departmental offices for timeliness of the referral. The average time
frame that lapsed after debts were eligible to be sent to FMS was 197
days. According to a Treasury OIG official, this delay involved
referrals to TOP and cross-servicing and occurred because Treasury
personnel did not properly age the receivable balances.
* The Department of State did not have a routine process to certify and
send its debts for cross-servicing after they became eligible. This
department was sending debts to FMS for TOP collection actions only
once a year, and its officials mistakenly thought that FMS would
transfer the department's delinquent debts from offset to
cross-servicing.
We also found that even when the agencies referred debts, the debts were not
always valid and legally enforceable and thus not eligible for
cross-servicing. Based on our analysis of a statistical sample of 200
delinquent debts referred to FMS, 22 delinquent debts were likely invalid or
legally unenforceable. Specifically, we found 14 debts that were
subsequently returned by FMS to the referring agency because the debts were
invalid or involved debtors in bankruptcy. At the completion of our detailed
testing, another eight debts had not yet been returned to the agency. Five
of these debts involved debtors in bankruptcy, and the other three debtors
were deceased.
FMS' Outreach Efforts
In an effort to encourage debt referrals, in the spring of 1999, FMS
requested written debt referral plans from 22 of the 24 CFO Act agencies.
The plans were of limited use because (1) FMS had no assurance that agencies
had properly identified all nontax debts that were eligible for
cross-servicing, (2) many of the plans did not include debt amounts or
timeframes for referral, and (3) FMS did not use the plans to closely
monitor actual agency referrals.
According to FMS officials, as of the completion of our fieldwork, 21 of the
22 CFO Act agencies had submitted debt referral plans. The Social Security
Administration (SSA) did not submit a plan and has not referred any debts to
FMS even though information provided by FMS indicates that SSA had about
$444 million of eligible debt for cross-servicing as of September 30, 1998.
Five of the 21 agencies reported that all eligible debts had been referred.
Ten of the remaining 16 agency referral plans did not contain details on the
specific debt amounts to be referred and/or time frames for cross-servicing
referrals. For example, one agency submitted a plan stating that all of its
components would refer debts for cross-servicing in July and August 1999,
but did not mention any specific dollar amounts. Information prepared by FMS
as of February 2000 indicated that this agency had referred only $109,000 of
the $11 million of debt that it had reported as eligible as of September 30,
1998.
According to an FMS official, because some of the agency plans were
incomplete, FMS did not closely monitor agencies' adherence to their
referral plans. In addition, as long as agencies were referring some debts,
FMS generally did not contact agencies about their plans. FMS officials also
stated that FMS did not have the authority to assess penalties or take other
formal actions against agencies that did not promptly refer their eligible
debts. In May 2000, FMS sent letters to 23 of the 24 CFO Act agencies
requesting debt referral milestones for fiscal years 2000 and 2001. In the
letter, FMS enclosed a debt referral schedule for agencies to complete
detailing the specific debt amounts to be referred and the related time
frames. In addition, FMS has recently been meeting with CFO Act agencies to
determine debt amounts eligible for cross-servicing. Further, a FMS official
stated that FMS plans to request written referral plans from non-CFO Act
agencies and meet with officials from the larger non-CFO Act agencies
regarding such plans.
FMS' Cross-Servicing Operations Management
Since then, FMS has established the Birmingham Debt Management Operations
Center (BDMOC) as its primary facility for handling governmentwide
cross-servicing operations. This facility was a former FMS payment center
that was being phased out as part of Treasury's consolidation of payment
operations. Except for efforts to collect on erroneous payments under the
former payment center operations, the staff had little prior experience in
debt collections. According to FMS officials, FMS trained BDMOC employees
before they assumed their debt-collection duties and periodically updated
their debt collection training, as needed, on topics such as debt collection
techniques and the servicing of particular debts.
Thus far, FMS has not engaged the services of federal agencies with ongoing
and experienced debt collection operations to assist in governmentwide
cross-servicing debt collection efforts. In December 1996 and October 1999,
FMS issued standards allowing agencies to apply to be part of a collection
network. The Departments of Education and Health and Human Services were
granted waivers by Treasury to the cross-servicing provision of DCIA, which
allows these agencies to take collection action on certain classes of their
own debts. Three agencies -VA, the Department of Health and Human Services,
and the Department of Agriculture's National Finance Center - submitted
applications for designation as governmentwide debt collection centers.
Treasury denied approval of these agencies primarily because it determined
that these entities did not have the needed capabilities. FMS officials
stated that they have not received any additional applications for
designation as governmentwide debt collection centers.
Oversight of FMS Debt Collectors and PCAs
FMS' strategy for debt collection is reflected in its Cross-Servicing
Implementation Guide, Standard Operating Procedures (SOP), and contracts
with PCAs. These documents indicate that when nontax debts are referred to
FMS, BDMOC collects debts using tools such as demand letters, phone calls,
and payment offset through TOP. Debts were to be referred to PCAs for
collection if FMS could not secure an acceptable agreement with the debtor
or locate the debtor. To expedite debt collection, the procedures stated
that debts may be referred to TOP and will be referred to PCAs within
stipulated time frames.
We statistically selected a sample of 200 debts from a population of 61,269
debts with balances greater than $100 referred to FMS from April 1998
through May 1999. We identified FMS' collection activity for these 200
debts, as well as by PCAs for the debts that were subsequently referred to
the PCAs.
Our tests of these 200 selected debts found that FMS collectors did not
always adhere to the cross-servicing SOP. For example, we found that FMS
collectors
* did not always attempt to contact debtors or perform skiptracing to
locate debtors who did not respond to demand letters,
* negotiated two repayment agreements that significantly exceeded
authorized pay back time frames, and
* did not always promptly refer all debts to TOP or PCAs.
Contacting Debtors
When the debtor fails to respond within 10 days after the demand letter is
sent, collectors were required to contact the debtor by telephone. If a
valid telephone number was not available or the demand letter had been
returned as undeliverable, the collector was then required to start
skiptracing activities to locate the debtor.
The debt history files indicated that FMS issued demand letters for all 200
debts we statistically selected and reviewed. However, for 168 of these 200
debts, there was no subsequent contact between the debtor and the collector,
and, as shown in figure 3, we found no evidence that FMS' collectors
subsequently tried to contact 96 (48 percent) of these debtors. For the
remaining 72 of the 168 debts, information in the debt history files
indicated that FMS collectors had no success in their attempts to contact
the debtor by telephone or locate the debtor by other collection activities,
such as skiptracing. FMS collectors attempted to phone the debtor for 31 of
the 72 debts. However, for certain of these debts, available documentation
suggests that these efforts were limited. Specifically, the debt history
files showed that for 18 of these 31 debts, the collector placed only one
phone call to the debtor with no subsequent follow-up. For the remaining 41
of the 72 debts, FMS collectors performed skiptracing.
Figure 3: Most Debtors Not Contacted by FMS Collectors Subsequent to
Issuance of Demand Letter
(Statistical sample of 200 delinquent debts referred
to FMS from April 1998 through May 1999)
In our sample of 200 delinquent debts, the debt history files indicated that
demand letters for 46 of the debts were returned as undeliverable. For 29 of
these 46 debts, we found no evidence that FMS collectors performed the
required skiptracing to locate the debtor.
FMS officials stated that its collectors might not have documented all
discussions they had with debtors. However, FMS' procedures required
collectors to record all debtor conversations and collection activity in the
debt history files. Moreover, the Comptroller General's Standards for
Internal Controls in the Federal Government states that all transactions and
other significant events need to be clearly documented and that the
documentation should be readily available for examination.
FMS officials also stated that its collectors might not have been able to
perform all the collection activities for each debt within the stipulated
time frames. In particular, according to these officials, FMS' collection
efforts were negatively affected when FMS received large batches of referred
debts from the Department of Housing and Urban Development (HUD) and the
Small Business Administration during our detailed testing period. Our review
of referred debts by month for the 14-month testing period showed that,
generally, as referrals increased, the percentage of debts in our sample
with no evidence of attempts to contact the debtor also increased.
FMS did not establish any written guidance to help its collectors determine
which debts to cross-service first during peak referral periods. In
addition, FMS did not perform any analysis to determine if it would be more
cost effective, especially during peak referral periods, to send all or
certain types of debts immediately to the PCAs. For example, FMS did not
review its history of debt collections to determine if its collectors have
had more success in collecting debts with certain characteristics (e.g., age
of delinquency, dollar value of debt, referring agency, commercial versus
consumer). During peak referral periods, FMS collectors could then focus on
such types of debts while forwarding the rest to the PCAs, thereby avoiding
further aging of debts for which no collection efforts are likely to be
taken.
Repayment Agreements
For 13 debts in our sample of 200 that involved repayment agreements, 2 had
terms of 75 months and 96 months, significantly exceeding the 36-month
preauthorized limit established by the referring agency. We found no
evidence in the debt history files that FMS collectors obtained approval to
exceed these limits from the referring agency or FMS management. According
to FMS officials, this lack of evidence was likely due to errors made by FMS
collectors.
Compromised Debts
According to FMS' cross-servicing SOP,
* collectors first had to attempt to obtain payment in full before they
offered a compromise,
* before offering a compromise, the collector was required to obtain the
debtor's Taxpayer Identification Number (TIN) so that the compromised
amounts could be reported to IRS,
* collectors were required to obtain current financial statements from
the debtor to determine the debtor's ability to pay and assess the
merits of a proposed compromise, and
* collectors were authorized to enter into a written compromise repayment
agreement not to exceed 3 months.
In our statistically selected sample of 78 compromised debts, our analysis
showed that the compromised amounts ranged from about $27 to $36,000 and
averaged $4,863. The compromised amount as a percentage of outstanding
balance averaged 39 percent. During our review of the 78 compromised debt
history files, we found that
* 75 ( 96 percent) files did not indicate why collectors compromised
debts or the basis used to determine how these debts met FMS' criteria
for compromising,
* 72 files had no evidence that the collector attempted to obtain a lump
sum payment in full or a repayment agreement for the full amount before
compromising the debt, and
* 74 files did not have documentation, such as financial statements or
any other type of financial analysis, to support the compromise
decision.
For example, one debt history file indicated that a debtor was allowed to
pay $62,000 to settle an agency's debt with an outstanding balance of about
$98,000. This agency had authorized FMS to compromise up to 10 percent of
the outstanding balance of the debt, but there was no authorization from the
agency or other support for compromising about $36,000, approximately 37
percent, of the outstanding debt. Further, there was no documentation to
show that the FMS collector had followed the cross-servicing SOP or had
analyzed the debtor's financial condition or ability to pay.
Our review of the 78 compromised debts also found the following.
* Thirty-two debts involved written or oral compromise repayment
agreements. Of these 32,
* 30 agreements exceeded the established 3-month repayment limit. The
terms of these agreements ranged up to 13 years, half of which exceeded
3 years, for an average of 57 months. One FMS collector compromised
$7,966, or about 50 percent, of an agency's debt with an outstanding
balance of $15,932 and entered into a compromise repayment agreement in
which the debtor is being allowed to pay $50 per month for 159 months.
* the debt history files for 19 debts did not contain (1) a signed
written compromise repayment agreement and (2) evidence that the FMS
collector attempted to follow up to obtain a signed written agreement
from the debtor.
* the debtors defaulted on 16 of the 32 agreements. Of these 16 debts, 7
had no evidence in the debt history files that the FMS collector
contacted the debtor to determine whether the debt collection strategy
could be modified.
* Four of the 16 debts with defaulted compromise repayment agreements
were not requested for referral to PCAs until more than 30 days after
default.
* Eleven of the 78 debts did not have a TIN listed in the debt history
files or any evidence that the FMS collector attempted to obtain a TIN
required for IRS reporting and needed for referral to TOP.
TOP Referrals
We found that FMS' collectors referred 13 of the 62 debts that were eligible
for referral to TOP within the stipulated time frames. On the other hand,
FMS did not refer 36 of the 62 debts that were eligible for referral to TOP
promptly at 20 days because of interface problems between internal computer
systems. In addition, FMS collectors referred another 7 debts to TOP between
5 and 141 days after the 20-day period and had not referred 6 other debts as
of the completion of our detailed testing. According to FMS officials, the
late referrals or lack of referrals for the latter 13 debts were likely due
to errors made by FMS collectors.
Referral to PCAs
We found that FMS' collectors did not promptly request referral of many
debts to PCAs. In our sample of 200 delinquent debts, 183 debts were
eligible for referral to PCAs at the time of our detailed testing. For 33
(about 18 percent) of the 183 debts eligible for referral to PCAs, FMS
collectors requested referral of these debts to PCAs between 3 and 383 days,
or an average of 62 days, after the 30-day (or 50-day) time period.
We also identified instances where there were significant delays between the
date the FMS collector requested a debt referral to a PCA and the date FMS
actually transferred the debt to a PCA. Of the selected 183 debts eligible
for referral, 178 had been referred to PCAs as of the date of our detailed
testing. Nineteen of these 178 debts were transferred from FMS to the PCAs
between 30 and 64 days after the FMS collectors' request. As a result, no
collection activities were taking place and these debts continued to
increase in age of delinquency. FMS officials could not provide an
explanation for the longer time frame.
PCA Collection Activities
PCAs, among other things, are required by their contract with FMS to send
demand letters within 5 working days of receipt of the debt from FMS,
attempt to locate debtors through skiptracing, including obtaining credit
bureau reports for debtors with debt balances of $500 or more, and attempt
to obtain full payment before compromising any debt. In addition, PCAs are
required by contract to record all collection activity occurring on debts in
their respective debt collection systems. PCA contract monitors employed by
FMS have access to the PCA debt collection systems and are required to
regularly review debt records in these systems to verify that demand letters
are issued, ensure that collection activity is appropriate, and evaluate
payment schedules.
PCAs sent demand letters for 158 of the 178 debts (or about 90 percent) on
time. For 17 debts, PCAs sent letters between 1 and 87 days late, averaging
15 days late. As of the date of completion of our detailed testing, no
demand letters had been sent for 3 of these selected debts. According to FMS
officials, delays in sending 13 of the 17 late demand letters were primarily
caused by one PCA that did not download its electronic debt files in a
timely manner.
In addition, PCAs did not always obtain credit bureau reports as part of
skiptracing. Of the 178 debts referred to PCAs, 152 debts had balances over
$500, and their debt history files indicated that the collector performed or
should have performed skiptracing. For 19 of these 152 debts, we did not
find evidence in the PCAs' debt collection systems that the collector
obtained the required credit bureau report. When we brought this to the
attention of FMS officials, they immediately acted by issuing a technical
bulletin to PCAs to remind them of the contractual requirement to obtain
credit bureau reports as part of skiptracing activity for debts with
balances of $500 or more.
Of the 178 delinquent debts referred to PCAs, we identified 10 debts that
involved compromise offers by PCAs. For 4 of these 10 debts, the PCA's debt
collection system had no indication that the PCA attempted to obtain either
a lump sum payment in full or a repayment agreement for the entire amount
before compromise. Further, for all 10 debts, the PCA's debt collection
system did not indicate that the collector requested financial statements or
other documents reflecting the debtor's financial condition or ability to
pay, and such documents, if they existed, were not provided to us by FMS.
Revised SOP
Based on these actions and discussions with FMS officials, FMS is placing
increased reliance on PCAs. However, FMS has not performed an analysis to
determine the potential effect such reliance may have on net collections to
the federal government. Such an analysis may be warranted given that (1) as
the debts are not actively worked by BDMOC and are awaiting referral to
PCAs, they continue to age and thus typically become more difficult to
collect and (2) the federal government pays a 25-percent fee on debt amounts
collected by the PCA that the government is not always able to recoup from
the debtor.
Distribution of Debts to PCAs
In preparing for these subsequent distributions to the PCAs, the debts were
first arrayed by the earliest to the latest date the electronic data were
entered into the PMAC system and then, for each date entered, by dollar
amount from highest to lowest. The PCAs were then arrayed starting with the
PCA with the highest distribution goal amount, followed by the remaining
PCAs in descending order. The system then began sequentially assigning the
debts to the PCAs until each PCA had received at least one debt. After the
first round, the system reordered the PCAs from the largest remaining goal
amount to the smallest. The PMAC system continued to sequentially assign
debts and reorder the PCAs after each debt was assigned until all eligible
debts had been distributed. A more detailed description of this process is
included in appendix II.
We obtained copies of pertinent data from the PMAC system and performed
various analyses of the debt account information, including distribution of
the debt accounts to the PCAs, age of delinquencies, and collection rates.
Our analysis of FMS' distribution of debt accounts from the inception of the
program (February 1998) through February 2000, which is partially summarized
in table 2, showed that one PCA had received a significantly higher
percentage of the debts with smaller balances. Overall, the average balances
of the debts distributed to this particular PCA were 45 percent lower than
the average balances of all debts distributed during this time frame.
Specifically, debts distributed to this PCA had average balances of $11,436,
while the overall average balances of debt accounts distributed to PCAs were
$20,845.
Table 2: Analysis of Debt Distributions
Dollar amount of
Number of debts distributed debts distributed
PCA as a percent of total debts as a percent of Average debt
distributed total dollars balance
distributed
1 4% 6% $31,034
2 6% 10% $31,944
3 6% 6% $23,041
4 29% 16% $11,436
5 8% 8% $22,275
6 7% 9% $29,620
7 9% 9% $19,722
8 11% 13% $24,922
9 7% 7% $18,603
10 7% 9% $25,025
11 6% 7% $26,217
100% 100% $20,845
Source: GAO's analysis of the PMAC database for debts distributed to PCAs
from February 1998 through February 2000.
This trend also existed for many of the age of delinquency categories: less
than 180 days, 180 days to 1 year, 1 to 2 years, 2 to 4 years, 4 to 6 years,
6 to 11 years, and greater than 11 years. We found that in the first four
delinquency categories noted above, the PCA mentioned above had the smallest
average debt balances. This PCA also had the next to the smallest average
debt balances in the 4 to 6 years and 6 to 11 years delinquency categories.
For example, we found that 27 percent of this PCA's debts were less than 1
year delinquent, with an average balance of $5,593. In addition, we found
that 32 percent of the total number of debts that were less than 1 year
delinquent were distributed to the one PCA. These debts also represented 19
percent of the total dollars of debt in this delinquency category. The PCA
to receive the next highest percentage of the debts less than 1 year
delinquent was distributed 11 percent of the total number of debts
representing 13 percent of the total dollars, with an average debt balance
of $10,694.
On the other hand, we found that for one agency's debts for which no
collections had been made through February 2000, 35 percent of the total
number of accounts for this same agency were distributed to two PCAs,
representing 55 percent of the total amounts received by these PCAs, with a
combined average balance of $2.7 million. The one PCA mentioned above, who
is not one of these two, received 17 percent of the total number of accounts
of this same agency's debts representing 14 percent of the total amounts
received by the PCA with an average balance of
$1.4 million. We noted that several agencies had referred debts with large
average dollar balances ranging from $54,000 to $1.7 million for which no
amounts have been collected.
Further, we analyzed collections on closed debt accounts with payments
categorized by age of delinquency. Collection industry statistics have shown
that collection rates are generally higher on debts with smaller dollar
balances and that are less delinquent. While the PCA mentioned above had
collected the most in total dollars, it ranked highest in collections as a
percentage of the total amounts referred only in the 1 to 2 years and 4 to 6
years delinquency categories. In the other five delinquency categories,
other PCAs had higher collection percentages. Our analysis also showed that
generally three out of the four PCAs with the highest collection percentages
in each delinquency category had average debt balances that were below the
overall average balances for that category. For example, the three PCAs with
the highest collection percentages in the less than 180 days delinquent
category had average account balances of $265, $810, and $857. The overall
average account balances for this delinquency category was $2,003. Thus,
FMS' collection experience appears to be consistent with that of the
collection industry statistics noted above.
Concerns relating to the distribution method have been raised by some of the
PCAs. During our interviews with the 11 PCAs, we found that when asked how
the debts should be distributed, the general consensus among them was that
the distribution should consider the characteristics of the debts, such as
age of delinquency, type of debt (consumer or commercial), agency referring
the debt, and debt balance. Many of the PCAs indicated that stratifying the
available debts by agreed-upon characteristics would result in each of the
PCAs receiving a proportionate mix of the debts and foster a more
competitive environment.
An important consideration to help ensure that each PCA receives a
proportionate mix of debts is that the population of debts to be distributed
is homogeneous, i.e., of the same characteristic, such as age of
delinquency, balance, type, or originating agency. For each distribution,
FMS placed all the debts available into one pool. Our analysis of the debts
found that the debts within each distribution's pool were generally not of
the same composition, i.e., not of the same average balance or age of
delinquency. This factor contributed to the distribution results experienced
by FMS which are discussed above.
FMS compiled information on the distribution and collection of the debts
referred to the PCAs in response to various congressional and other
requests. During our fieldwork, FMS had not yet analyzed these data. Given
the above noted PCAs' feelings about the distribution method and the results
thus far of FMS' distributions to the PCAs, it may be necessary for FMS to
periodically analyze the distribution and collection data to determine
whether adjustment is needed to the distribution model to assure that a
proportionate mix of debts is being distributed to the PCAs and competition
among the PCAs is more fully promoted.
FMS' Cross-Servicing Fees and Related Costs of Operations
FMS hired a contractor to assist with the development of its estimated
costs, a model for conducting break-even analyses, and fee setting. The FMS
contractor indicated that fees for cross-servicing would have to increase
substantially over current levels for FMS to achieve full cost recovery.
We determined, using the current fee structure and the fiscal year 1999
collection experience, that FMS would have to increase annual collections by
over sevenfold, or collect approximately $173.5 million, to cover its fiscal
year 1999 estimated costs of $11.0 million. The estimated
$173.5 million in collections would include approximately $141.6 million to
be returned to the referring agency for collected debts, $20.9 million in
fees paid to PCAs, and $11 million in cross-servicing fees paid to FMS.
The amount of collections needed to reach a break-even basis varies
substantially depending on who collects the debt, FMS or the PCAs. If FMS
did all the collections at its 18 percent rate, debt collections of about
$72 million would be needed to cover the $11 million estimated costs.
Conversely, if FMS relied on the PCAs exclusively for collections, its 3
percent fee on these collections would require that the PCAs bring in
collections of about $471 million to cover these costs.
Projected higher future costs will require even more collections to break
even. FMS' fiscal year 2000 cross-servicing cost estimate that we were
provided is about $12.9 million or about 17 percent greater than the fiscal
year 1999 estimate. According to FMS officials, FMS has not projected
cross-servicing fee revenues and costs beyond fiscal year 2000. Although the
officials stated that FMS is currently considering increasing
cross-servicing fees, they have acknowledged that the cross-servicing
program will not be fully reimbursable in the foreseeable future. Thus, the
cross-servicing program will likely have to be funded primarily through
appropriations at least in the near term.
Mr. Chairman, this concludes my prepared statement. I would be pleased to
respond to any questions you or other Members of the Subcommittee may have.
Contacts and Acknowledgments
Appendix I
Explanation of Terms/Data in Table 1
Cross-servicing waivers-Treasury granted the Departments of Education and
Health and Human Services waivers from cross-servicing for certain classes
of their own debts. According to FMS officials, the waiver, which is valid
for a 3-year period, allows the agencies to perform collection activity on
those debts subject to the waivers. As of October 1999, agencies can no
longer apply for waivers, but rather must apply for exemption from
cross-servicing for specific classes of debts.
Debts in forbearance or in appeals-Debts that are subject to forbearance or
that are in appeals generally are not "legally enforceable." Forbearance
action taken by a creditor generally extends the time for payment of a debt
or postpones, for a time, the enforcement of legal action on the debt. The
government cannot pursue collection against a debtor if the debt is not
legally enforceable.
At DOJ-Debts that are referred to DOJ for litigation or collection are
excluded for referral to Treasury for cross-servicing by DCIA.
Foreign debt-Debt that is owed by foreign governments is excluded for
referral to Treasury. Treasury stated that, for the most part, collecting
these delinquent debts is infeasible primarily due to foreign diplomacy
considerations and affairs of state.
Debts in bankruptcy-The automatic stay mandated by 11 U.S.C. Section 362
generally prevents the government from pursuing collection action against
debtors in bankruptcy.
Debts in foreclosure-Debts in foreclosure are governed by state laws. In
some states, to maintain the right to foreclose, a creditor must foreclose
the collateral securing the debt before seeking other collection remedies.
DCIA excludes debts that are in foreclosure for referral to Treasury for
collection action.
Department of Defense-According to an FMS official, certain contractor debt
held by the Department of Defense (DOD) and reported as debt over 180 days
delinquent as of September 30, 1998, was subsequently reclassified from
eligible to ineligible debt. Specifically, in August 1999 FMS and DOD agreed
to reclassify $1.3 billion of such debt to ineligible debt due to ongoing
litigation.
Debts eligible for internal offset-Debts that will be collected under
agency-initiated offset, if such offset is sufficient to collect the claim
within 3 years after the date the debt or claim is first delinquent, are
excluded for referral for cross-servicing by DCIA.
Debts at third party-Debts being serviced and/or collected in accordance
with applicable statutes and/or regulations by third parties, such as
private lenders or guaranty agencies, are exempt from cross-servicing by
Treasury regulations.
Appendix II
FMS Distribution Methodology
The PMAC system begins the distribution process by creating a list of debt
accounts for distribution that are ordered first by the earliest to latest
date the debt account was entered into the PMAC system. Then, for each date
entered, by dollar amount from highest to lowest. The PMAC system then
performs a series of calculations to determine a dollar-limiting amount that
represents the highest dollar amount of an individual debt account included
in the distribution list. All individual debts that exceed the
dollar-limiting amount are excluded from that distribution. According to FMS
officials and its contractor, the dollar-limiting amount was established to
help ensure that no debt accounts would be assigned that are larger than
every PCA's distribution goal amount. According to FMS, the list of debt
accounts are ordered by date entered into the PMAC system to help ensure
that no debt account will remain unassigned for an extended period of time.
Next, the PMAC system calculates the goal distribution percentages for each
PCA. The goal distribution percentages are based on the performance
evaluation results of the prior 4 month performance period and are used to
determine the dollar amount of debts each PCA will be allocated. The PMAC
system then orders the PCAs by listing the PCA with the largest distribution
goal first followed by the remaining PCAs in descending order according to
goal distribution amounts. The PMAC system assigns the debt accounts
sequentially from the debt account listing to PCAs so that the PCA with the
largest remaining distribution goal amount will receive the debt with the
largest balance and so forth. A debt can be assigned to a PCA if the debt
amount will not cause the PCA to exceed its remaining distribution goal
amount within a preestablished tolerance amount and the debt account has not
been previously assigned to the PCA in a prior distribution. If the debt
account cannot be assigned to the first PCA on the list, the system proceeds
to the next PCA. This process continues until each PCA has been assigned at
least one debt account.
After each PCA has been assigned at least one debt account, the PCAs are
reordered from the largest remaining distribution goal amount to the least.
The PMAC system continues to sequentially assign the debt accounts and
reorder the PCAs after each debt account is assigned until all eligible
debts that can be assigned have been assigned. Debt accounts that are not
assigned during the distribution process are included with the next
distribution.
Appendix III
FMS' Cross-Servicing Fee Rates
In its standard written agreement with all referring agencies, FMS requires
agencies to pay FMS cross-servicing fees for nontax debt collections on
debts referred to FMS. The agreement states that FMS is entitled to a
cross-servicing fee for all nontax debt collections received after it
initiates collection action, which is defined as the issuance of a demand
letter and/or an attempt to contact the debtor. FMS fees are based on a
percentage of the initial referred debt amount that is collected. FMS'
cross-servicing fees effective during the period of our fieldwork are listed
below in table 3.
Table 3: Cross-Servicing Fee Rates
Fee rates
Type of cross-servicing collection
(%)
Debts referred to FMS for cross-servicing and
collected by FMS collectors 18
Post-judgement debts referred to FMS and subsequently
collected by the Department of Justice (DOJ) 18
Debts referred to FMS for cross-servicing and
subsequently collected by PCAs 3
Debts referred to FMS for cross-servicing and
subsequently collected by TOP 3
Debts referred to FMS for cross-servicing and
subsequently collected by DOJ (excluding 3
post-judgement enforcement)
Debts referred to FMS and sent directly to and
collected by PCAs with no collection activity 1
performed by FMS, referred to as "pass-throughs"
(901803)
Orders by Internet
For information on how to access GAO reports on the Internet, send an e-mail
message with "info" in the body to:
[email protected]
or visit GAO's World Wide Web home page at:
http://www.gao.gov
Web site: http://www.gao.gov/fraudnet/fraudnet.htm
E-mail: [email protected]
1-800-424-5454 (automated answering system)
*** End of document. ***