The Legal Services Corporation (LSC) is a federally
funded agency with 269 grantee offices around the country that have
provided over $6 billion of free legal aid to the eligible poor
since 1974. For fiscal year (FY) 2000, the LSC is requesting a $40
million increase, to bring its funding level to $340 million. This
represents about a 13 percent increase in agency funding at a time
when the U.S. General Accounting Office (GAO), the LSC's own
inspector general, and the press have uncovered serious problems
with the agency's case reporting statistics and performance
numbers. Audits of the LSC's 1997 caseload data for 11 grantee
offices--which reported handling 370,000 cases--determined that
only 198,000 were valid.
Until Congress receives accurate
information about the performance of the Legal Services
Corporation's grantees, it cannot hold the agency accountable for
its performance and its use of taxpayer dollars. Congress should
demand that all LSC programs supply timely and accurate data on
program performance; it also should require independent audits and
conduct investigative hearings. In addition, Members of Congress
should ask whether it is even appropriate for the federal
government to be funding this program. If it is not, they should
consider closing down the LSC by transferring its funding to the
Department of Justice, with a strict formula for block granting
those funds to the states based on the number of poor in each
jurisdiction. The responsibility for providing legal services to
the poor belongs more appropriately to state and local officials
and to private-sector institutions--those closest to the people in
need of assistance.
THE TROUBLE WITH THE LSC'S NUMBERS
The
LSC Factbook is a benchmark of LSC performance figures not only for
Congress, but for states and private funding sources as well. In
its 1998 Factbook, the most recent issue
available to Congress, the LSC reported that a total of 1,932,613
poor people were aided. (See the Appendix for the
1998 Factbook's 1997 case statistics for specific congressional
districts.)
After egregious errors in the 1998
Factbook numbers were reported in the press, however, the LSC was
compelled to admit that it had not served as many clients as it had
reported. An April 1999 LSC press release noted 400,000 fewer cases
closed in 1997 than were reported in the 1998 Factbook. The agency
also amended its 1998 projections, revising them downward. The LSC
soon will deliver its 1998 data to Congress in the 1999 Factbook.
Until additional audits are completed, no one can know with any
certainty what the agency has accomplished with the taxpayer
dollars that Congress has appropriated in the past.
The
LSC's representation of its open and closed cases is important,
because it is the only tangible information currently available to
Congress on the agency's overall performance. Until this year,
Congress has never seriously questioned the accuracy of LSC's
numbers.
As
Congress considers funding for the LSC within the Commerce,
Justice, State, the Judiciary, and Related Agencies appropriation
bill, it should seek credible, factual performance information that
justifies the Administration's substantial budget request,
especially in light of the LSC's previous misreporting of data.
The Search for Accurate Performance
Information
Congress relies increasingly on
performance measures, such as the number of clients served by the
Legal Services Corporation, to decide whether funding for the
agency's programs should be increased or decreased. Congress should
specify better performance measures that would include credible
data on the quality, and not simply the quantity, of the services
the LSC provides to the poor. Indeed, Members of
Congress--especially appropriators--look at the performance of
programs to determine whether a program is working efficiently and
achieving its goals. As Representative Harold Rogers (R-KY),
chairman of the House Appropriations Subcommittee on the
Departments of Commerce, Justice, and State, the Judiciary, and
Related Agencies, told the LSC in March 1999, "We want accurate
information.... We do make our judgments based on the volume of the
load that is represented to us."
In
December 1998, the Congressional Research Service (CRS) produced a
study that examined the extent to which the past two Congresses had
used the 1993 Government Performance and Results Act, a tool for
measuring the success or failure of government programs and holding
agencies accountable for their use of taxpayer funds. The study,
requested by House Government Reform and Oversight Committee
Chairman Dan Burton (R-IN), examined provisions in public laws
enacted during the 104th and 105th Congresses. According to the
CRS:
There are...indications that committees
are interested in using performance-related information in the
appropriations process and associated budget documentation. Over a
third of all the committee reports identified in [our study]
contained provisions linking performance measures and the budget
process. Such provisions either stated the intent of the committee
to consider the agency's progress in articulating outcome goals and
measures during the appropriations process, suggested that the
agency's budget submission include Results Act-related information
and measures, or referred to realignment of program and budget
structures in an agency's budget submission.... In addition, many
reports included language that noted that future funding for an
activity or program would be contingent upon establishing goals and
measures or upon future performance against established goals.
Even
the LSC recognizes the linkage between performance and funding. Its
president stated recently that
Case statistics play an essential role in
the budget request and performance plan submitted by LSC to the
United States Congress each year. Therefore, the reliability of
case statistics submitted by programs to LSC is vital to obtaining
continued Federal funding for Legal Services.... We believe this
type of information...holds great promise for securing increased
federal funding for legal services.
Congress appears to base its funding of
the Legal Services Corporation on the agency's reported caseloads.
The LSC then distributes the money to grantees based on a formula
that takes local poverty statistics into account.
ACCOUNTABILITY VS. SPIN CONTROL
As a
key funding source and manager of taxpayer dollars, Congress must
have access to trustworthy LSC data about caseloads, clients, and
agency operations. If the LSC cannot provide dependable information
on even the quantity of cases handled by its offices, there is
little hope that Congress can obtain more sophisticated and
meaningful quality performance information on this government
service.
The
demand for accountability pressures government officials to define
and then demonstrate their performance, particularly when programs
are seriously deficient. As one former LSC employee recently
lamented, for example:
For
more than 20 years I worked at a small Legal Services
Corporation-funded program and the work was immensely rewarding.
But in 1996, when I was the program's litigation director, I quit
due to the many practices and policies which, to my mind, had all
but destroyed the program's ability to render competent legal
assistance to people unable to afford counsel. One of these
practices was the counting of virtually every telephone call as a
"case" in order to build up numbers to report to LSC and other
funding sources. Consequently, hundreds if not thousands, of
reported cases were nothing more than referrals or other responses
given by paralegals or secretaries.
The
LSC chose to use case statistics as a measure of its performance.
On February 25, 1998, for example, LSC President John McKay
testified:
For
FY 1999, LSC seeks an appropriation of $340 million. We estimate
that this amount will enable local legal services programs funded
by LSC to resolve over 1.6 million cases involving critical legal
problems for eligible clients and their families.... Because of
limited resources, local programs are forced to turn away tens of
thousands of people with critical legal needs.
Although Congress did not know about the
caseload reporting problems, appropriators approved a $17 million
increase for the LSC on October 21, 1998, bringing its FY 1999
budget to $300 million. Congress provided the funding with an
expectation that the additional money would enable the LSC to serve
1.6 million clients in calendar year 1998. According to the LSC
president, this new funding
represents the strong bipartisan backing
that LSC has developed, and signals a renewed confidence that LSC
is carrying out the will of Congress and is a vital part of the
justice system. The increase will allow LSC-funded programs to
serve a greater number of poor and disadvantaged clients more
effectively in 1999.
Uncovering the LSC's Reporting
Problems
Concerns about the LSC's misrepresentation
of its actual caseloads began to build after the agency's inspector
general began in March 1999 to release the results of several
audits. Since the IG has a dual reporting responsibility--both to
the LSC Board and to Congress--the IG should have informed Congress
of the seriousness of the errors found in the data. For example, of
149,589 cases reported for 1997 by six grantee offices, two-thirds
were found to be invalid (see Table 1).
The
inspector general's findings include the following examples, among
others:
- The Legal Aid Society of San Diego
claimed it closed 33,096 cases for 1997, but the IG audit revealed
that only 10,787 of these cases were legitimate.
- Florida Rural Legal Services
admitted in August 1998 that 39,471 of the cases it reported were
invalid. This reduced the actual number of legitimately reportable
cases to 13,922 out of 53,393 reported.
- Legal Services of Miami claimed to
have closed 23,800 cases in 1997; only 7,607 were found to be
valid.
- Of the 16,490 cases reported by the San
Francisco Neighborhood Legal Assistance Foundation, only 4,134 were valid; the program's
director submitted a revision when the increased scrutiny of
caseload data began.
- Legal Services of Northern Virginia
reported 9,115 cases; only 5,156 were deemed valid.
- The Houston officereported 13,695
cases, but in a preliminary report yet to be finalized and released
by the IG, only 9,995 potentially were valid.
WHY THE LSC'S PERFORMANCE NUMBERS WERE
FLAWED
The
reporting problems found in every program audited by the LSC's own
inspector general and the U.S. General Accounting Office since the
1997 case statistics were reported raise concerns about systemic
LSC performance deficiencies and reporting abilities. According to
the IG
and GAO
audits, LSC performance numbers included:
- Repeat reporting of old "open"
cases;
- Phantom or non-existent cases;
- Telephone contacts reported as cases when
eligibility was not determined and the applicant was not accepted
into the program;
- Inclusion of non-LSC-funded cases in
reports; and
- Double counting of cases.
Upon
learning of the inspector general's preliminary audits, newspapers
and editorial pages began to report on the LSC's problems. On April
8, 1999, for example, the Associated Press released a story
documenting the problems that the IG had begun to acknowledge in
March 1999. The news story indicated that some Members of Congress
were concerned that the LSC might have misrepresented the number of
cases it handled intentionally in order to secure additional
funding.
Following the AP story, several editorials in newspapers across the
country criticized the LSC for its errors.
On
March 3, 1999, during an annual appropriations oversight hearing,
Representative Tom Latham (R-IA) began to ask questions about the
LSC's veracity in reporting its caseload. The hearing was
significant not only because it was the first time that LSC's
numbers had been challenged by a Member of Congress, but also
because it established clearly the committee's interest in linking
the budget request to the agency's performance. In a follow-up
written response to Latham's questions, the LSC's IG admitted that
the agency's reported caseload figures are used for the annual
budget request submitted to Congress. This admission heightened
concerns in Congress and eventually precipitated a congressionally
requested GAO audit of LSC grantees.
GAO Confirms Serious Data Problems
On
May 3, 1999, five Members of Congress asked the U.S. General
Accounting Office to continue to conduct random audits of LSC
programs to collect additional facts before this year's allocation
of tax dollars to the LSC. Congress asked the GAO to provide
preliminary results of audits on five grantees by June 21. On June
25, 1999, the GAO reported that all five grantees audited had
problems accurately reporting the number of cases handled. This
reinforced the findings of the IG's own audits.
As
Table 2 shows, the grantees overreported closed cases, and four of
the five grantees overreported open cases. The operations of all
five grantees included cases in which the eligibility of clients
was not verifiable. In addition, the GAO reported that four of the
five offices reported closed cases in which no activity had
occurred during the past year, and five reported open cases where
no activity had occurred during the past year.
Clearly, the LSC's reported caseload
figures did not stand up to independent review and auditing.
THE LSC'S RESPONSES TO THE FINDINGS
Officials of the LSC, including the
agency's president, were aware of the grantee reporting errors
months before Congress was informed. In the summer of 1998, the
LSC's president was informed of the audit findings of case
reporting problems. In September 1998, the
inspector general informed some of his staff that "the numbers
provided to Congress were inaccurate." The LSC, however, did not
plan to release this information until March 2000, when its first
Performance Report under the 1993 Government Performance and
Results Act is due.
By
late 1998, the LSC should have viewed the audit findings as serious
enough to bring to the attention of Congress, especially since
Congress at the time was debating whether to increase LSC funding
for FY 1999 by $17 million. The IG, knowing that the LSC did not
plan to inform Congress of significant errors in its reported data,
should have informed Congress of what the auditors were finding.
Instead, the semiannual report issued by the inspector general on
September 30, 1998, reported "no significant problems, abuses or
discrepancies" in LSC programs.
At a
recent public debate at The Heritage Foundation, LSC President John
McKay admitted that his inspector general had advised him of the
seriousness of the reporting problems in the summer of 1998:
Our
Inspector General is here in the audience, and I would hesitate to
speak for him, but it was very clear that, based on the strength of
oral advice, from him to me, beginning actually in the summer of
1998, that we [the LSC] had a problem concerning the accuracy of
the cases.
Why
this serious problem was not reported to Congress is the heart of
the issue. Indeed, in April 1999, the LSC revised downward its
public estimates of the numbers of clients served in 1997 using
taxpayer dollars, even though it was asking
all grantee offices to increase the types of cases they report for
the next Factbook.
The
LSC sent new guidance to all grantees to modify future methods for
reporting cases. One of the many changes
required would have the effect of helping each program to report
more cases. This subtle yet important change requires each LSC
program to report cases on which it has worked regardless of
funding sources. Since 40 percent of the funding for most LSC
grantees typically comes from non-federal sources--states, bar
associations, or other private or public sources--this change not
only will have the effect of inflating some of the future caseload
numbers, but also will make it difficult to compare data relating
to LSC's performance.
In
May 1999, following congressional inquiries and the April AP story,
the LSC sent out another letter advising all grantees that, among
other things, a GAO audit was underway concerning the data they
already had reported for 1997. It asked program directors
to affix their signatures attesting to the accuracy of their case
statistics for 1998 because these figures would be compiled for the
1999 Factbook (which, because the LSC Factbooks typically are
published in May of each year, is now overdue).
Unacceptable Excuses
Official denials of systemic reporting
problems by the LSC have involved the following claims:
LSC CLAIM #1:
The scope of the problem is overstated; only five grantees
overstated their cases, which is less than 3 percent of the LSC's
caseload.
FACT:
Each of the 11 LSC programs reviewed by an independent
auditor--either the LSC's inspector general or the GAO--showed
false case reporting problems. Specifically, the IG identified
problems at Northern Virginia, Houston, San Diego, Miami, Florida
Rural, and San Francisco. The LSC self-identified
similar problems at Alameda, Central Michigan, Los Angeles, and
Western Carolina. Of 25 randomly selected cases audited at Farm
Workers Legal Services of North Carolina, nearly all lacked data
critical to determining whether the clients helped were indeed
eligible for federal aid. The GAO reviewed five
additional programs, each one of which involved similar errors.
LSC CLAIM #2:
If anything, the LSC is underreporting its caseload.
FACT:
Because each new review of the LSC's 1997 data reveals broad
miscounting and overreporting, all numbers provided by the LSC are
now viewed with skepticism. The LSC has taken steps to change the
method for developing caseload estimates for Congress by asking
offices to increase the types of cases handled, even if they are
not funded by federal tax dollars. This will make annual
comparisons of LSC caseload data, as well as performance measures
for federally funded programs, nearly impossible.
LSC CLAIM #3:
There is no evidence of fraud.
FACT:
The LSC has not engaged in candid self-disclosure of problems with
its 1997 case statistics. The gap in time between when the IG and
the LSC leadership learned there was a problem and when Congress
was advised of that problem is unacceptable. Moreover, the LSC's
request for a $40 million increase in its FY 2000 budget--based on
the same overinflated estimates from the disputed 1997 case
statistics, and after LSC's president had been informed of the
problem--is itself nothing short of fraudulent.
LSC CLAIM #4:
The Inspector General Act prevents the IG from informing Congress
before his semiannual report is due.
FACT:
Nothing prevents the LSC or its IG from advising Congress of
discrepancies found in data used to award taxpayer funding. In
fact, compliance with the 1994 Government Auditing Standards
requires the IG to advise Congress and management whenever there is
a need for timely reports. These standards encourage
interim or oral reports to stimulate, not stymie, information
flowing to policymakers.
LSC CLAIM #5:
The LSC itself uncovered the problem through self-initiated audits,
brought it to Congress's attention, and took steps to correct
it.
FACT:
The timeline shows that the LSC and its own inspector general knew
of the emerging case reporting problems and did nothing to inform
Congress. In late 1998, Congress voted an increase in federal
funding for the LSC by relying on information the LSC knew to be
false, and on its exaggerated claims of its performance. As late as
March 1999, the LSC was still using these unreliable data in its
requests for additional funds from Congress.
LSC CLAIM #6:
Case numbers and performance information have no bearing on funding
levels. Specific allocations are based on the eligible populations
living in each service area, not on the number of cases handled or
referred. Therefore, there is no incentive to inflate
numbers.
FACT:
The LSC, Congress, and even LSC grantees use performance or
caseload numbers to influence funding from federal and non-federal
sources. The agency's own five-year strategic plan for 1997-2002
established as an annual goal "[to] seek to provide high-quality
legal services to the greatest number of eligible clients that our
appropriation will support." According to the LSC's
president:
Case statistics play an essential role in
the budget request and performance plan submitted by LSC
to...Congress each year. Therefore, the reliability of case
statistics submitted by programs to LSC is vital to obtaining
continued Federal funding for Legal Services.... [T]his type of
information...holds great promise for securing increased Federal
funding....
Moreover, inaccurate case numbers from the
LSC can:
- Attract
additional non-federal funding (in 1997, over $200 million in
revenue came from non-federal funding);
- Skew evaluations
for various competitive grants;
- Discourage
competition from more cost-effective providers of legal services;
and
- Mislead Congress
and the public into believing that the LSC is performing better
than is actually the case.
WHAT CONGRESS SHOULD DO
No
one denies that the less privileged in society benefit
significantly from free legal assistance. However, the LSC services
only about 5 percent of the eligible poor. The lives of thousands
of people have been improved by the efforts of pro bono attorneys
and the ad hoc network of organizations and people, such as private
foundations, churches, and synagogues, that have stepped up to
assist the poor when they are in need. Unfortunately, however, the
federal program to help the poor with legal assistance--the Legal
Services Corporation--has shown itself to be deceptive in measuring
its performance and impervious to efforts to institute
accountability.
What Members of Congress Should Ask in
Oversight Hearings
- What did the Legal Service Corporation, its board, and its
inspector general know, and when did they know it?
- Why wasn't Congress informed as remedial steps were
taken?
- Whose decision was it not to report data and performance
problems to Congress immediately, or to consider waiting until
March 2000 to inform Congress?
- For any given LSC grantee's reported caseload, how much is
attributable to cases opened in 1996, 1995, 1994, 1993, 1992, or
earlier?
- What is the status of the Houston audit, the findings of which
were given to the inspector general last August but still have not
been released?
- Who received the 1998 Factbook from the LSC and needs to be
informed of its inaccuracies?
|
For
this reason, the first question Congress should ask is whether the
federal government should be running this program at all. If it is
decided that providing legal services to the poor is not
appropriately a federal function, Congress should consider
devolving this responsibility to the states, local governments, and
private-sector institutions and putting the LSC on a clear path
toward eventual shutdown.
To
accomplish this, Congress should transfer funding for legal
services for the poor to the Department of Justice, with a strict
formula for block-granting funds to the states based on the number
of poor in each jurisdiction. Block grants not only would eliminate
federal overhead, but also would permit states to institute their
own accounting standards for grantees and allow them to conduct
their own audits. Recent strides in welfare caseload reduction at
the state level--driven in large part by the autonomy of the states
to design appropriate welfare-to-work transitioning programs--have
emboldened Washington to return to the states the responsibility
for other federal programs once thought too large or cumbersome for
states to handle. Washington should acknowledge that state and
local leaders who know best how to serve the legitimate and
critical legal needs of the poor are in a better position to design
the most efficient system and provide quality services to those in
need.
Alternatively, if it is decided that the
Legal Services Corporation does represent a proper federal
function, Congress should establish the criteria for evaluating the
LSC's performance, including case statistics. In the short term,
Congress should:
- Demand that the LSC issue its 1999
Factbook with 1998 caseload figures as soon as possible.
Congress and the public need to review how the LSC has spent
taxpayer money--and what, if anything, it has accomplished--before
deciding how much (if anything) should be appropriated for FY 2000.
The 1999 Factbook may not be published until late July, giving
little time for appropriators to study changes from the disputed
numbers in the 1998 Factbook.
- Verify the accuracy of information
Congress receives from the LSC in the future.
To ensure that federal tax dollars are not wasted and that
those most in need are being helped, and to hold accountable those
LSC officials who are responsible for providing inaccurate
information to Congress, Congress should:
- Require
an annual independent audit of LSC case statistics, either by the
GAO or by an outside contractor, to obtain a verifiable and
accurate accounting of LSC performance. Audits should begin by
verifying 1997 and 1998 data, since only 11 of the 269 grantee
offices have been audited for their 1997 caseloads to date.
- Preventthe LSC from
administratively changing the definition of "reportable" cases to
avoid accurate assessments of performance. The LSC's new
administrative guidance to grantees in November 1998 will do just
that. Congress should specify how it wishes the LSC to track
federal funds and performance data.
- Apply the
Federal False Statements Act to the LSC and its grantees to prevent
future misrepresentation of facts during the appropriations
process. This act would allow penalties for misreporting data about
caseloads or clients served with federal taxpayer dollars.
- Reduce FY 2000 funding.
Congress should reduce the LSC's annual appropriation, or make
its funding contingent on the release of accurate data, to offset
the overfunding provided in the past from LSC's provision of
inaccurate data. Since federal funding is premised on delivery of
services to a certain number of poor people and LSC's data have
been questioned, federal funding should not be increased to the
requested $340 million. Congress must send a strong message that
deception will not be tolerated or rewarded with larger
budgets.
- Conduct new oversight hearings to
determine what LSC officials knew and when they knew it.
The information that comes to light in these hearings could
set the stage for future legislative changes to ensure that such
misreporting does not happen again. Such hearings would
establish a benchmark for LSC performance and would demonstrate
that Congress is serious about performance data and the accuracy of
the information upon which it bases appropriations.
- Highlight the need for reform in each
state.
As further scrutiny of the LSC and its 269 grantees' caseload
data continues, Members of Congress can encourage local oversight
efforts by state legislatures and local media. In Virginia, for
example, heightened press attention to the reporting errors in one
LSC program triggered new reporting and oversight by the state
legislature.
CONCLUSION
In
1993, Congress passed and the President signed the Government
Performance and Results Act with bipartisan support and the Clinton
Administration's stamp of approval. The act codified Washington's
desire to hold federal programs accountable for their performance
and use of taxpayer dollars.
This
law is useless, however, unless Congress can rely on the
information provided by federal agencies. Until Congress receives
accurate information about the performance of Legal Services
Corporation grantees, it will continue to be unable to hold the LSC
accountable. Congress should demand that all LSC programs supply
timely and accurate data on program performance, and it should
require independent audits and hold new investigative hearings to
determine the reliability of information supplied by the LSC.
In
the LSC's case especially, Members of Congress must be tenacious in
seeking and obtaining the facts before spending more taxpayer
dollars. The LSC's functions are carried out better and more
appropriately by the states, localities, or private organizations.
Until Congress can eliminate funding for this agency, however,
enhanced congressional oversight is needed. With better information
about the LSC's performance, Congress can assess the
cost-effectiveness of the agency's delivery of services compared
with other options to improve legal assistance to the poor.
Virginia
L. Thomas is a Senior Fellow in Government Studies and Ryan
H. Rogers is a former Research Assistant in Government Studies at
The Heritage Foundation.
APPENDICES
Legal Services Corporation Grantee
Caseloads
A Column Has Been
Included for 1998 Numbers When They Become Available
Source: Legal Services Corporation, Office of
Inspector General, and Office of Information Management
Click
on the appropriate link below to see the corresponding chart.
Endnotes