The Indian Trust Fund Litigation: An Overview of Cobell v. Kempthorne
The Indian Trust Fund Litigation:
An Overview of Cobell v. Kempthorne
August 20, 2008
M. Maureen Murphy
American Law Division
The Indian Trust Fund Litigation:
An Overview of Cobell v. Kempthorne
On August 7, 2008, the U.S. District Court for the District of Columbia issued
an opinion awarding $455.6 million in restitution to the plaintiffs in Cobell v.
Kempthorne. The litigation has been before the courts since 1996. The dispute
involves the federal government’s alleged mismanagement of accounts held in trust
for individual Indians. Central to this dispute is the duty of the Department of the
Interior (DOI) to provide a historical accounting of the accounts. This duty has
proven very difficult to fulfill, however, for a variety of reasons.
In January 2008, the trial court reached the conclusion that DOI was unable to
produce the required accounting and scheduled an evidentiary hearing on an
appropriate remedy. That remedy, according to the August 7, 2008, ruling, is $455.6
million in restitution. The sum differed greatly from what the plaintiffs had sought
— approximately $47 billion. In its ruling, the court rejected the plaintiffs’
methodology of computing the amount owed to the trust beneficiaries and their
additional claims for “benefit to the government” for funds not credited to the
accounts of the beneficiaries. The court left for future determination the question of
how the award is to be distributed among the members of the plaintiff class.
The purpose of this report is to provide a brief background of the history leading
up to the litigation, a review of the issues that have proven so difficult for the
judiciary to resolve, and a brief description of the method used by the trial court to
determine the amount of the restitution. The report will be updated as warranted by
judicial decisions or legislative intervention.
The Latest Decision...............................................7
The Restitution Issue...........................................7
The “Benefit to the Government” Issue............................11
Application of Equitable Principles...............................13
The Indian Trust Fund Litigation:
An Overview of Cobell v. Kempthorne
On August 7, 2008, the U.S. District Court for the District of Columbia issued
an opinion awarding $455.6 million in restitution to the plaintiffs in Cobell v.1
Kempthorne. Individual Indian trust funds are the focal point of this case which has
been litigated since 1996.2 Individual Indian trust funds, as distinguished from tribal
trust funds, are monies which the federal government holds for the benefit of
individual Indians. The conflict in the case traces to the federal government’s trust
responsibility with respect to American Indians. The Supreme Court first formulated
the concept of the federal government as trustee for Indian tribes in 1831, likening3
the relationship to that of “a ward to its guardian.” In the capacity of trustee, the
United States holds title to much of Indian tribal land and land allotted to individual
Indians. Receipts from leases, timber sales, or mineral royalties are paid to the
federal government for disbursement to the appropriate Indian property owners.
Flowing from the federal trusteeship of Indian lands and mineral resources are
fiduciary responsibilities on the part of the United States to manage Indian monies4
and assets which have been derived from these lands and are held in trust.
The case is premised on statutory duties imposed upon the federal agencies
handling Indian monies as well as on the existence of property rights in funds and
assets held in trust for Indians. The courts have recognized broad powers of
Congress with respect to Indian affairs legislation and Indian property, but have also
recognized that Indian property may not be taken for a public purpose without just
compensation.5 This case is not a claim for just compensation; it is a claim for an
accounting by the trustee for receipts and disbursements representing the trust corpus
held for the benefit of individual Indians.
The Cobell litigation sprang out of the federal government’s trust responsibility
with respect to three groups of money accounts held in trust for individual Indian
1 Cobell v. Kempthorne, ___ F. Supp. 2d ___ (No. 96-1285 (JR)), 2008 WL 3155157
(D.D.C. August 7, 2008) (hereinafter, Cobell restitution award opinion).
2 Both the Department of Justice website [http://www.usdoj.gov/civil/cases/cobell/] and the
Cobell plaintiffs’ website [http://www.indiantrust.com/index.cfm] include documents and
opinions filed in the case.
3 Cherokee Nation v. Georgia, 30 U.S. 1 (1831).
4 United States v. Mitchell, 463 U.S. 206, 225 (1983) (quoting Navajo Tribe of Indians v.
United States, 624 F. 2d 981, 987 (Ct. Cl. 1980)).
5 See, e.g., United States v. Sioux Nation, 448 U.S. 371 (1980).
beneficiaries. These accounts are commonly referred to as the Individual Indian
Money (IIM) accounts. They include (1) Land-based Accounts — established to
receive revenues derived from the approximately 11 million acres held in trust by the
U.S. for individual Indians;6 (2) Special Deposit Accounts (SDAs) — intended to be
temporary accounts to hold funds that could not be immediately credited to the
proper IIM account holder; (3) Judgment and Per Capita Accounts — established to
receive funds form tribal distributions of litigation settlements and tribal revenues.7
Congress has delegated to the Secretary of the Interior and the Secretary of the
Treasury its responsibilities as trustee with regard to the IIM accounts.8 The Bureau
of Indian Affairs (BIA) has general responsibility for trust land management and
Most transactions involving IIM accounts require BIA approval.10 One of BIA’s
most important duties in this regard is managing IIM funds derived from income-
producing activities on allotment land, including grazing leases, timber leases, timber
sales, oil and gas production, mineral production, and rights-of-way. The Office of
Trust Fund Management (OTFM) is responsible for BIA’s fiduciary duty to keep
accurate financial records of these activities. OTFM also shares the banking aspect
of DOI’s trust responsibility with the Treasury Department. OTFM and BIA officers
collect payments and deposit them into a local bank where there is a Treasury
General Account.11 The Treasury Department maintains a single “IIM account” for
all IIM funds, rather than individual accounts, while OTFM is responsible for
maintaining accounting records for the individual funds.12 Treasury also invests the
funds at the direction of the Department of the Interior (DOI).13
6 These accounts trace their beginnings to the federal government’s allotment program of
the late 19th and early 20th centuries. Under this program, the Secretary of the Department
of the Interior (DOI) was authorized to allot portions of reservation land to individual
Indians. Title would remain with the United States in trust for a number of years, after
which it would pass to the individual allottees free from all encumbrances. The allotment
policy resulted in large amounts of land passing into non-Indian ownership, and Congress
abandoned the policy in 1934, extending indefinitely the trust periods of allotments that had
not yet passed into fee ownership. Many of these properties remain in trust to this day. See
Felix F. Cohen, Handbook of Federal Indian Law § 1.04 (2005 ed.).
7 Id. at § 5.03[a], n. 138.
8 25 U.S.C. § 161a(a).
9 See, e.g., the federal timber management statutes, at 25 U.S.C. §§ 406-407, 466.
10 See Cobell v. Babbitt, 91 F.Supp.2d 1, 9 (D.D.C. 1999), afff’d, 240 F. 3d 1081 (D.C. Cir.
11 Id., at 9-10.
12 Cobell v. Norton, 240 F.3d 1081, 1089 (D.C. Cir. 2001). When OTFM issues a check to
an IIM trust beneficiary, the amount is deducted from the individual fund, even though the
money remains in the Treasury’s general account. As a result, the beneficiary loses any
interest that would accrue between issuance and cashing of the check, a time lapse that “may
be short in the private sector, [but] can be much longer in the IIM trust context because
OTFM often has incorrect addresses for the recipients.” Cobell v. Norton, 91 F. Supp.2d
13 Cobell v. Norton, 240 F. 3d 1081, 1088-1089.
The federal government — as holder of these accounts in trust for the Indian
beneficiaries — has fiduciary obligations to administer the trust lands and funds
arising from them for the benefit of the beneficiaries. The federal government has
stipulated, however, that it does not know the exact number of IIM trust accounts that
it is supposed to administer; nor does DOI know the correct balances for each IIM
account.14 DOI concedes that it is currently unable to provide an accurate accounting
for a majority of IIM trust beneficiaries.15 The Treasury Department also has
problems with trust fund management procedures. First, in conformity with federal
law, the Treasury Department allows the destruction of documents over six years and
seven months old, and makes no effort to ensure that documents related to accounting
for IIM accounts are preserved.16 In addition, there can be a time lapse between the
deposit of funds with the Treasury Department and the investment of those funds.17
There can also be a time lapse between the issuance of a check and when the payee
presents the check, resulting in lost interest.18
Congressional oversight committees became concerned with IIM
mismanagement in the late 1980s and began holding oversight hearings regarding the
IIM accounts in 1988. Four years later, the House Committee on Government
Operations produced a report highly critical of the Interior Department.19 In 1994,
Congress enacted the Indian Trust Fund Management Reform Act (the Reform
Act),20 recognizing the federal government’s pre-existing trust responsibilities and
further identifying some of the Interior Secretary’s trust fund responsibilities, such
as providing adequate accounting for trust fund balances; providing adequate controls
over receipts and disbursements; providing accurate and timely reconciliations;
preparing and supplying periodic statements of account performance and balances to
account holders; and establishing consistent, written policies and procedures for trust
fund management.21 Significantly, the original House bill (H.R. 1846) would have
made the accounting duty prospective only. When another similar bill was
introduced to take H.R. 1846’s place, that provision was left out. This new bill
became the Reform Act, and the courts interpreting it in the Cobell litigation have
determined that DOI owes a historical accounting duty going back to the act of June
14 Id., at 1089.
16 See Id., at 1092.
19 Misplaced Trust: The Bureau of Indian Affairs’ Mismanagement of the Indian Trust Fund,
H.Rept. 102-499 (1992). Largely in response to this report, BIA contracted with Arthur
Anderson & Co. to audit and reconcile a sampling of 17,000 IIM accounts, which the
company was unable to accomplish due to inadequate records.
20 P.L. 103-412 (25 U.S.C. 4001 et seq.).
21 25 U.S.C. § 162a(d).
22 Cobell v. Norton, 240 F.3d 1081, 1102 (D.C. Cir. 2001).
a portion of the government’s specific obligations and created additional means to
ensure that the obligations would be carried out.”23
In 1996, a group of IIM account holders filed a class action suit to compel
performance of trust obligations, alleging that the Secretaries of the Interior and the
Treasury — as delegatees of the federal government’s trust responsibilities — had24
breached the fiduciary duties owed to plaintiffs by mismanaging the IIM accounts.
Two years later, the district court judge bifurcated the trial into two phases, with
Phase 1 to focus on reforming the management and accounting of the IIM trust funds,
and Phase 2 to address the historical accounting of those accounts.25 In 1999, United
States District Judge Royce C. Lamberth issued a ruling as to Phase 1, holding that
the Treasury and Interior Secretaries had breached their fiduciary duties to the IIM26
account holders. The transition to Phase 2 has proven increasingly difficult,
however because the defendants have been unable to submit — in forms acceptable
to the court — plans for reforming the account-management system and for providing
a historical accounting.27
Providing the far-reaching accounting required in the Cobell litigation is a
difficult task on a number of levels. The plaintiffs argue that an accurate accounting
is impossible, due to lack of records. Two issues, however, have proven particularly
difficult for DOI in performing this task. The first of these issues is the fractionation
of interests in many of the allotment lands. These interests have been fractionated
over the years as they have been divided among the heirs of the original allottees,
increasing exponentially with each generation and leading to incredibly small
interests that are difficult to track. DOI estimates that there are currently over 1.4
million fractional interests of 2% or less involving 58,000 tracts of land.28 No matter
23 Id., at 1100.
24 For a discussion of the history of the case, see Id., at 1092-1093 (D.C. Cir. 2001).
26 Cobell v. Babbitt, 91 F.Supp.2d 1 (D.D.C. 1999), aff’d, 240 F.3d 1081 (D.C. Cir. 2001).
27 Cobell v. Norton, 226 F.Supp.2d 1, 162 (D.D.C. 2002). Finding a workable method to
provide the historical accounting has proven extremely difficult. Adding to that difficulty
is the fact that the litigation has become increasingly acrimonious. See Can a Process Be
Developed to Settle Matters Relating to the Indian Trust Fund Lawsuit?: Oversight Hearingthst
Before the Committee on Resources, U.S. House of Representatives, 108 Cong. 1 Sess.
50 (2003) (statement of John Berry, Chairman, Quapaw Tribe). During the course of the
litigation, Judge Lamberth issued contempt orders against, among others, former Treasury
Secretary Robert Rubin, former DOI Secretary Bruce Babbitt, and former DOI Secretary
Gale Norton (the D.C. Circuit later overturned the contempt order against Secretary Norton).
28 See Can a Process Be Developed to Settle Matters Relating to the Indian Trust Fund
Lawsuit?: Oversight Hearing Before the Committee on Resources, U.S. House of
Representatives 9 (statement of James Cason, Associate Deputy Secretary, U.S. Department
of the Interior). For a description of the various accountings that DOI has attempted,
how small the revenue generated by the interest (DOI cites some interests as being
so small that they generate less than one cent per year), DOI is required to account
for them.29 While DOI has stated that it can perform a transaction-by-transaction
accounting of the judgment and per capita accounts and the SDAs, the problems
presented by the land-based accounts have proven very difficult to resolve. DOI has
argued that it should be able to use statistical sampling with respect to some of these
The second difficult question is how far back a historical accounting should
reach. At various points in the litigation, the different parties have argued for an
accounting of transactions as far back as 1887 (date of the Allotment Act), 1938 (date
of the Indian Reorganization Act), and 1994 (date of the Reform Act). Resolving this
problem would likely encompass a choice between what is fair and what is possible.
One could have very different answers to these two questions, mainly because, as the
litigation so far has shown, DOI and Treasury records relating to the IIM accounts are
at best incomplete.
In January 2003, DOI provided a new historical accounting plan to Judge
Lamberth that would cover all accounts open as of October 25, 1994, when the
Reform Act was enacted. After reviewing DOI’s plan, Judge Lamberth in September
2003 issued a controversial structural injunction giving the court broad oversight
authority to ensure that (1) DOI carries out the accounting (the court adopted what
is essentially a modified version of DOI’s historical accounting plan, but did not
allow DOI to use statistical sampling with respect to the land-based accounts); and
(2) DOI reforms its system for managing the IIM accounts. Judge Lamberth also
appointed a monitor to ensure compliance with the injunction order.30
One month later Congress passed an appropriations rider stating that “nothing
in [the Reform Act] or in any other statute, and no principle of common law, shall be
construed or applied to require the Department of the Interior to commence or
continue historical accounting activities with respect to the [IIM] Trust” until 2005
or when Congress more clearly delineates DOI’s accounting obligations under the
Reform Act.31 Congress took this action in direct response to Judge Lamberth’s
structural injunction order, stating that compliance could cost upwards of $6 billion
and that diverting that amount of resources could be “devastating to Indian
country.”32 Two subsequent appropriation bills have limited the funds available to
DOI for the historical accounting to $58 million for FY2005 and FY2006.33
including the historical accounting for the named plaintiffs in Cobell, see Id., at 12-13.
29 DOI is also required to provide trust services to the owners of such interests, including
title records, lease management, and probate. See Id.
30 Cobell v. Norton, 283 F.Supp.2d 66, 225 (D.D.C. 2003).
31 P.L. 108-108 (117 Stat. 1263).
32 See H.Rept. 108-330, at 117 (October 28, 2003).
33 P.L 108-447 (118 Stat. 2809) and P.L. 109-54 (119 Stat. 499).
On December 10, 2004, the D.C. Circuit issued an opinion striking down almost
all of Judge Lamberth’s injunction.34 The court first held that, pursuant to Congress’s
directive contained in the aforementioned appropriations rider, DOI could not be
compelled to perform any historical accounting. The court noted, however, that the
directive would sunset on December 31, 2004, and the judges pointed out that they
could not “address the issues that would be relevant if the district court [after
December 31, 2004] reissued those provisions [compelling a historical
accounting].”35 The court next largely overturned Judge Lamberth’s injunction as to
DOI’s systemic reform. Looking to Supreme Court precedent, the D.C. Circuit held
that judicial review under the Administrative Procedure Act (APA) is limited to
specific agency actions, and that such review cannot be extended to “claims of broad
programmatic failure.”36 The court held that Judge Lamberth, in issuing his
injunction, had impermissibly wandered into this latter area, which is more properly
reserved for executive or legislative action. While the D.C. Circuit upheld Judge
Lamberth’s requirement that DOI submit a plan laying out how it will come into
compliance with its fiduciary obligations, the court found that the other elements of
Lamberth’s order (e.g., the appointment of a monitor, the listing of and compliance
with tribal laws) were not tied to specific findings of wrongdoing and suggested
greater, and unlawful, judicial intrusion into agency discretion.
On February 23, 2005, Judge Lamberth — noting that the deadline contained
in the appropriations rider had passed — issued another structural injunction with
respect to the historical accounting.37 Once again, he adopted a modified version of
DOI’s historical accounting plan, but prohibited the use of statistical sampling and
required an accounting going back to the Allotment Act of 1887. He refused to stay
the order pending appeal, citing the plaintiffs’ nine-year wait and “a delay directed
by Congress in a bizarre and futile attempt at legislating a settlement in this case.”38
On November 15, 2005,39 the D.C. Circuit vacated Judge Lamberth’s injunction and
historical accounting order and directed that, on remand, the district court, in
evaluating DOI’s plan for a statistical sampling to accomplish the accounting, should
not ignore the general language of the Reform Act and subsequent congressional
limitations on funding, suggesting that the Reform Act should not be seen as
mandating “the best available accounting without regard to cost.”40
On July 11, 2006, the court of appeals found the district court to have abused
its discretion in ordering DOI to disconnect from the Internet many of its computer
34 Cobell v. Norton, 392 F.3d 461 (D.C. Cir. 2004).
35 Id., at 468.
36 Id., at 472.
37 Cobell v. Norton, 337 F.Supp.2d 298 (D.D.C. 2005).
39 Cobell v. Norton, 428 F. 3d 1070 (D.C. Cir. 2005).
40 Id., at 1075.
systems,41 ordered the district court to vacate an order requiring DOI to include in any
communication to class members a warning that information provided about trust
matters might be unreliable, and removed Judge Lamberth from the case for bias42
in accusing DOI of present-day racism. While the appellate court recognized that
Judge Lamberth’s opinion did not reflect an animosity toward DOI, independent of
the developments in the case, it concluded that the language of the opinion combined
with the court’s actions might give rise to a public impression that justice could not
be done in the case.
On January 30, 2008,43 Judge James Robertson, assigned to the case in
December 2006, found DOI unable to remedy its breach of fiduciary duty and
ordered a hearing to develop a process for determining an appropriate remedy. He
rejected DOI’s historical accounting plan not because of its use of statistical sampling
methodology, but on the basis of finding its scope legally inadequate in terms of
years and accounts or funds to be covered.
The Latest Decision
The August 7, 2008,44 decision addresses both substantive issues — how to treat
plaintiffs’ claim for restitution for funds not credited or disbursed to individual Indian
beneficiaries and how to deal with their claim for “benefits to the government” with
respect to such funds. It also addresses jurisdictional issues — whether sovereign
immunity bars the court’s consideration of a remedy and whether a monetary remedy
is available under any federal statute. Critically, because an award of restitution
relies on the equitable power of the court, the opinion grapples with the question of
how to apply equitable principles to this unique case.
The Restitution Issue
Having recognized on January 30, 2008,45 that what had been the central effort
in the case over the years, the development of an accounting, had failed, Judge
Robertson held an evidentiary proceeding on the issue of remedy. On August 7,
41 Cobell v. Kempthorne, 455 F. 3d 301 (D.C. Cir. 2006).
42 Cobell v. Kempthorne, 455 F.3d 317 (D.C. Cir. 2006). Bias was seen in Judge Lamberth’s
language in Cobell v. Norton, 229 F.R.D. 5 (D.D.C. 2005), such as the following quoted by
the appellate court: “‘Alas, our ‘modern’ Interior department [sic.] has time and again
demonstrated that it is a dinosaur — the morally and culturally oblivious hand-me-down of
a disgracefully racist and imperialist government that should have been buried a century ago
as the last pathetic outpost of the indifference and anglocentrism we thought we had left
43 Cobell v. Kempthorne, 532 F. Supp.2d 37 (D.D.C. 2008).
44 Cobell restitution award decision.
45 Cobell v. Kempthorne, 532 F. Supp. 2d 37.
2008, he awarded an equitable remedy of $455.6 million in restitution.46 The figure
represents the amount to be restored to plaintiffs as receipts not credited to their
accounts. Claims for damages for funds which never were collected or for
mismanagement of assets are not included in the figure and were not before the court
in Cobell.47 To determine the amount of restitution, the court had to examine the
models the parties had set forth for determining the difference between what Treasury
had posted as receipts to the IIM account and what had been disbursed to individual
accounts or account holders. Information accumulated from the years of attempts at
arriving at a satisfactory and reliable means of reconstructing the records at DOI and
Treasury aided the court in evaluating the models offered by the parties.
Before calculating the amount of restitution, however, Judge Robertson
reviewed some principles drawn from previous decisions in the case. He noted that
although the court has broad equitable power to remedy trust nonfeasance, the fact
that the defendant is the federal government limits the court’s authority to fashion a
remedy.48 He gave at least two reasons for this limitation: (1) the court has to
examine its jurisdiction in the case carefully because the federal government’s
sovereign immunity renders it immune to suits to which it has not consented ( i.e.,
there must be a statute according jurisdiction for the particular issues being litigated),
and (2) although common law trust principles are applicable, their application must
be viewed and tempered in light of the unique character of the Indian trusts and their
trustee. According to the opinion, trust law would require that where the trustee has
not kept an account, the trust beneficiary should enjoy the benefit of evidentiary
presumptions. Those presumptions must be adjusted when applied “to a 121-year old
perpetual trust, managed by civil servants, with rapidly multiplying beneficiaries and
a variety of ever-changing assets.”49
The decision not to accord plaintiffs the full benefit of evidentiary presumptions
in their favor led to a major conclusion reached by the court — to apply a modified
burden of proof on the government’s statistical model for calculating data that it
could not produce. DOI had shown that only 77 percent of the monies collected for
the IIM system had been posted to IIM accounts. Accepting the government’s
calculation of receipts but using their own formula for calculating a disbursement
rate for each year, the plaintiffs claimed a shortfall of $3.6 billion over 122 years.
To this they proposed adding approximately $43.4 billion as “benefit to the
government” based on a formula which they had devised that assumed that whatever
was not disbursed to the account holders was available for general governmental
expenses, relieving the government of a need to issue and pay interest on Treasury
46 Cobell v. Kempthorne, ___ F. Supp. 2d ___ (No. 96-1285 (JR)), 2008 WL 3155157
(D.D.C. August 7, 2008) (hereinafter, Cobell restitution award decision).
47 Id., n. 1, at 5. It is possible such claims may be within the jurisdiction of the Court of
48 For a brief summary of how the court applied equitable principles, see infra, at 12-13.
49 Cobell restitution award decision, slip op., at 4.
The government produced evidence explaining some of the shortfall between
receipts and disbursements. A DOI expert, Michele Herman, testified that the
discrepancy reflected the fact that not all the funds received into Treasury’s IIM
account are intended to be credited to individual Indian trust fund accounts. Some,
such as tribal trust fund receipts or bid or lease deposits, are to be funneled on a pass-
through basis to other recipients.50 DOI did not have sufficient records to
substantiate the full amount of the discrepancy as pass-through funds but presented,
through another expert witness, a statistical model as a means of quantifying the
actual discrepancy. Had the trust law evidentiary presumption been employed in
plaintiff’s favor, it is likely that DOI’s evidentiary burden would have been greater.51
The plaintiffs estimated that $4 billion of the $14 billion posted to the IIM
account had never been never been credited to the accounts of or disbursed to
individual trust beneficiaries. The court rejected this amount both because of what
it perceived to be defects in the plaintiffs’ method of calculating the shortfall52 and
because no audit conducted during the course of the litigation “states or hints at the
disappearance of anything close to 30 percent of trust receipts.”53 The court also
noted that “[w]hatever problems have existed in the history of this trust, and however
serious the misfeasances and malfeasances of the trustees over 120 years, there has
never been evidence of such prodigious pilfering of assets from within the trust
Having admitted that 23% of IIM receipts had not been posted to IIM accounts
and faced with the plaintiffs’ charge that approximately $4 billion had not been
disbursed to account holders, the government was required to explain the shortfall
and put a figure on it. The explanation was essentially that the individual Indian trust
beneficiaries were not entitled to all of the funds that Treasury deposited in the IIM
account. Some were held there on a pass-through basis. Because of the inadequacies
of the recordkeeping and the inability to provide an accounting, the government was
unable to provide comprehensive data establishing the quantity of funds that should
have been credited to the plaintiffs’ accounts. Instead, it provided a complex
50 DOI’s expert also testified to various means by which bookeeping entries might not be
captured as disbursements to individual accounts. The court found the testimony to be
persuasive in “explaining the nature of the discrepancy between receipts and postings.” Id.,
slip op., at 28. The testimony, however, did not quantify percentages of “receipts ... for
stakeholders, for tribal trusts, and for third parties,” but acknowledged that to do so would
be “not impossible, but ...time consuming ... and has not been done.” Id.. slip op., at 28-29.
51 The plaintiffs criticized the expert’s testimony as inadequately demonstrated. The court,
however, saw the method as a suitable means of dealing the immensity of the task, noting
that the accounts involved 115 million transactions since 1986. Id., n. 8, slip op., at 19.
52 The basic deficiencies cited by the court are: (1) including as collections sums not
destined for the IIM accounts and (2) choosing to accept as accurate historical receipts data
while rejecting historical disbursement data. The court characterized the later as “self-
serving.” Id., slip op. at 17.
53 Id., slip op., at 16.
statistical model developed by the government’s expert witness, Dr. Frederick
Scheuren of the National Opinion Research Center at the University of Chicago.
The government’s model employed a “multiple imputation” technique to
account for the multiplicity of variables imputed to the large range of missing data
with respect to the accounts. It involved several complex steps, which are described
in the court’s opinion.55 It was able to produce various figures to represent what
should be the present balance in the individual Indian trust fund accounts and specify
a “confidence rate”56 with a margin for error identified for each figure given. The
court described the result of the statistical analysis as follows:
The resulting histogram ... shows that the calculated mean would be $583.6
million, which is $159.9 million more than the current stated balance of the
trust.... Because there is so much uncertainty in the data, however, the histogram
displays a wide variance.... The 95 percent confidence interval is wide enough
that it encompasses a zero difference between the calculated and stated balance
of the trust, meaning that the current, stated balance could very well be exactly
correct.... that variance cuts both ways: [a]t the 95 percent confidence interval,
the histogram encompasses $365.7 million ... The 99 percent confidence interval57
would accommodate a finding of a shortfall of up to $455.6 million.
Judge Robertson found the model offered by the government to be “plausible,”
“based on the sound and principled use of available data,” “methodologically fair and
carefully employed,” even if it “may not have adequately accounted for just how58
unreliable the underlying data was.” He noted that the government’s statistical
model conformed with what had been discovered about the trust funds: (1) it
admitted that the stated balance was less than what it should be; (2) it found a
variance around the mean which coincides “with the one thing that 12 years of this
litigation has truly settled, which is that there is still much that is unknown about the
trust”59; (3) its confidence interval included the current stated value of the trust60; and
(4) its $455.6 million figure “is best understood as an admission that the range of
uncertainty about the aggregate IIM trust [data] includes the possibility that the return61
of $455.6 million is necessary to correct the trust’s balance.”Although the plaintiffs
argued against the use of the model and identified “minor omissions” in it, the court
55 Cobell restitution award decision, slip op., at 28-33
56 A “confidence rate” is indicative of the margin for error of each of the figures; a 95
percent confidence rate, for example, includes a five percent margin for error.
57 Id., slip op., at 34 (citations omitted).
58 Id., slip op., at 35-37
59 Id., slip op., at 35.
60 The value of this, according to Judge Robertson, is that it “reflects the reality that, in the
absence of some kind of equitable evidentiary presumption in favor of the plaintiffs, one
permissible conclusion from the record would be that the government has not withheld any
funds from the plaintiffs’ accounts.” Id., at 35.
61 Id., slip op., at 35 (emphasis in original).
found that they “were unable to prove any bias that would render it fundamentally
Not only did Judge Robertson find the model offered by the plaintiffs to be
defective, he also criticized their failure to offer specific evidence to discredit the
defendant’s model. According to his opinion, the plaintiffs’ model could not be
considered, among other reasons, because it was inconsistent in accepting the
government’s estimate of receipts, but not of disbursements and reflective of “a
super-strong interpretation of the presumption against the breaching trustee that
cannot be equitably applied to the trusts at issue here.”63
Accordingly, Judge Robertson accepted the government’s model and provided
a certain evidentiary advantage to the plaintiffs by selecting the “maximally
conservative”64 estimate derived from the government’s statistical model rather than
the 95 percent level, which would be acceptable in a business context, or the 97.5
percent level that the government’s expert witness had suggested. He summed up
his reasons for this as follows:
Although it creates a highly conservative estimate — the most conservative
possible — it is far less draconian than the presumption that ordinary trust law
might apply in the ordinary trust case....After 12 years of contentious litigation,
the government can say only that, with 99 percent confidence, it believes that no
more than $455,600,000 is missing from the stated balance of the IIM trust. This
statement has the character of an admission — by responsible civil servants —
that there are limits to what can be confidently stated with respect to the IIM
system, and that a history of accounting nonfeasance makes a substantial error
plausible. It also represents the product of zealous and determined advocacy on
the part of the plaintiffs’ counsel, who have fought to vindicate their clients’
belief that the government’s inattention to its trust duties has had not only
historical significance, but real economic significance in the day-to-day lives of65
The “Benefit to the Government” Issue
The court completely rejected the plaintiffs’ request for compensation for
“benefit to the government,” which the plaintiffs added to their computation of a $3.5
billion difference between what had been received and disbursed. They had arrived
at that figure by using the government’s original figures for receipts and applying a
formula to devise a disbursement rate for each year. They also used a model to
compute what they alleged was the “benefit to the government.” This consisted of
deducting the disbursements they calculated for each year from the government’s
62 Id., slip op., at 38-39 (emphasis in original).
63 Id., slip op., at 68.
64 Id., slip op., at 69-70. According to Judge Robertson, the 99 percent confidence level
addresses the fact that there is more uncertainty in the data than the government’s expert
witness acknowledged, and such a figure “is a legally sound way of crediting obscurities and
doubts to the plaintiffs.” Id., at 70.
65 Id., slip op., at 70-71.
calculation of receipts and adding interest at the 10-year bond rate compounded
annually. The result was that they estimated that $43.3 billion of “benefit to the
government” should be added to the $3.5 billion for a total of $46.8 billion.66 The
court found the plaintiffs’ model inaccurate and not substantiated by any evidence
that the government had used any of these funds for general expenses.67 On the other
hand, the court found sufficient rebuttal evidence from a government witness for the
conclusion that no “benefit to the government” restitution was in order. 68
The court addressed two jurisdictional issues: sovereign immunity — whether
there was a statute that authorized the court to hear the suit against the United States
— and subject matter jurisdiction — whether there was authority for the court to
grant a monetary award. It resolved these issues principally on the basis of the
Administrative Procedure Act (APA).69 Under the doctrine of sovereign immunity,
a suit may not be brought against the United States unless a statute authorizes it. The
court found no statute which authorizes a federal district court to assess damages
against the United States in an action not based on torts and in an amount over
$10,000. It, therefore, rejected any claim that the plaintiffs may have as
compensation for the government’s failure to produce an accounting.70
However, the court did find that the APA could be invoked to restore to the
plaintiffs sums not credited to or disbursed from their accounts. The APA authorizes
an “action in a court of the United States seeking relief other than money damages.”71
The court reviewed Supreme Court jurisprudence on the question of whether
restitution of the sort contemplated comes within the meaning of “money damages.”
It concluded that although the Supreme Court has come close to holding that any
award of money constitutes “money damages,”72 it has recognized that a monetary
award is not always to be considered “money damages.”73 This is true particularly
when “the non-payment of money ... [is] the very unlawful act of which the agency
is accused,” rather than “compensation for another harm.”74 The court noted that the
66 Id., slip op., at 40-42.
67 Id., slip op., at 51 -53.
68 One government witness in rebuttal testified to the fact that Treasury borrowing decisions
do not involve consideration of funds outside of the Treasury General Account. Id., at 41.
69 5 U.S.C. § 702.
70 The court noted that non-tort claims for damages in excess of $10,000 are within the
exclusive jurisdiction of the Court of Federal Claims. See, 28 U.S.C. § § 1346(a)(2) and
71 5 U.S.C. § 702.
72 See e.g., Great-West Life & Annuity Ins. Co. v. Knudson, 534 U.S. 204, 210 (2002),
73 Bowen v. Massachusetts, 487 U.S. 879 (1988).
74 Cobell restitution award decision, slip op., at 44-45, citing Dep’t of the Army v. Blue Fox,
Inc., 525 U.S. 255, 262 (1999) (which read Bowen as holding “ that Congress employed this
facts of the case fit within this exception: the unlawful act at issue is a violation of
a principle of trust law requiring a trustee “to collect trust revenues and allocate them
to the beneficiary and not to himself.” For that reason, the court ruled that it had
jurisdiction to order disgorgement of trust funds not properly allocated and paid to
the beneficiaries.75 It also found that a monetary award as compensation for the
government’s breach of trust in failing to provide an accounting would be “money
damages” and, thus, excluded from its jurisdiction under the APA. The court also
concluded that because no “benefit to the government” had been identified, any
award for lost interest on funds not credited to plaintiffs would be “compensation”
and, thus, excluded from its jurisdiction under the APA as “money damages.”76
Application of Equitable Principles
Because the court was applying principles based on equity and because of the
uniqueness of the case, the court sought a means of drawing upon equitable
principles without doing an injustice to either party. The result was that it weighed
the government’s evidence less strictly than it would have had the case been a run-of-
the mill breach of trust case in state court. The court proceeded from the premise
that, if the plaintiffs were to benefit from evidentiary presumptions flowing from a
breach of trust, “the government — really, the taxpayers — [would be] ... held liable
for every transaction not proven by a method that is not legally required and would
be irrational to pursue.”77
Noting that there was no hard and fast federal trust law to guide the decision
other than various precedents indicating that the federal courts could turn to state
trust law and general trust law treatises for guidance, Judge Robertson reviewed the
history of courts of equity. He determined that a cardinal characteristic of a court
sitting in equity is flexibility to fashion a just remedy. He declared that he sought
to “seek fairness to both parties.”78 He determined that three elements in this case
called for modification of the evidentiary presumption that would have favored the
plaintiffs’ calculation of how much had not been credited to the beneficiaries: (1)
earlier the plaintiffs had “disclaimed” and the Court of Appeals had rejected “a ‘gold-
language to distinguish between specific relief and compensatory, or substitute, relief”).
75 Cobell, restitution award decision, slip op., at 45, citing Restatement (Third) of Trusts §
76 Id., slip op., at 57. The court also found that 25 U.S.C. § 4012 is inapplicable. By its
terms, it requires “payments to an individual Indian in full satisfaction of any claim of such
individual for interest on amounts deposited or invested on behalf of such individual before
October 25, 1994, retroactive to the date that the Secretary began investing individual Indian
monies.” 25 U.S.C. § 4-12. It was found inapplicable because: (1) it does not cover funds
not invested and (2) it does not waive the court’s $10,000 jurisdictional limit under § 28
U.S.C. § 1491(a)(1).
77 Cobell, restitution award decision, slip op., at 64. Judge Robertson declared that in the
January 2007 opinion, the court had not held that an accounting was impossible but that “a
meaningful accounting was irrationally expensive.” Id., slip op., at 64 (citation omitted).
78 Id., slip op., at 65.
plated’ form of accounting” as “ unduly burdensome”79; (2) such a presumption
would not be fair in light of volume of transactions, duration, geographic scope, and
unique nature of the trust at issue; and (3) using the presumption would produce a
result that would be “manifestly inaccurate” because the plaintiffs are assuming that
all of the funds deposited into the Treasury account were funds for individual Indian
Weighing those factors, the court determined that the evidence that the
government had produced was sufficient to meet the requirements of equity in that
its “statistical model allows for the use of the best available data while preserving
measures of uncertainty that can be scored to the plaintiffs’ benefit in the final
anal ys i s .”81
The determination of the amount of restitution to be awarded does not end the
case. Not only must the court consider options for distributing the award, there is
also the possibility of an appeal of the decision. The half-billion dollar award
contrasts with the $7 billion that the Bush Administration was willing to commit to
in 2005 for a legislative resolution of tribal and individual Indian trust fund issues.82
It is also almost $27 billion less than what had been recommended by a plaintiffs’
group.83 The plaintiffs may be weighing various options: an appeal of the awards;
an action in the Court of Federal Claims for compensation for the failure to provide
an accounting and for mismanagement of the funds; or legislation to compensate
them and resolve all issues. It might be noted that if plaintiffs choose to bring an
appeal of the amount of the award, the government may counter by challenging the
rulings resulting in the court’s finding authority to order a monetary award without
explicit authority in the 1994 Reform Act.
79 Id., slip op., at 63-64, citing Tr. of Appellate Oral Argument 40:1-45:25 (September 16,
2005), [Dkt. 3519, Exhibit 1]. The court also noted that in the present case there has been
a finding that an accounting would be “irrationally expensive.” Id., at 64.
80 Id., slip op., at 65. To credit plaintiffs with all funds deposited to the Treasury IIM
account would produce a windfall, according to the court, and would be contrary to equity.
Id., citing Bollinger & Boyd Barge Serv., Inc. v. The Motor Vessel, Captain Claude Bass,th
81 Cobell restitution award decision, slip op., at 66. As authority for this conclusion, the
court referred to the earlier appellate court ruling which had accepted statistical modeling
as a means of analyzing the data; case law on use of the best data; and the trust law principle
of resolving doubts in favor of the beneficiary when there has been a breach in the duty to
provide an accounting. Id., at 66.
82 See CRS Report RS22343 Indian Trust Fund Litigation: Legislation to Resolve
Accounting Claims in Cobell v. Norton, at 6.
83 Id., at 4. This was in the proposal of the Trust Reform and Cobell Settlement Workgroup,
organized by National Congress of American Indians President Tex Hall and Inter-Tribal
Monitoring Association Monitoring Association Chairman Jim Gray, in its “Principles for
Legislation” (June 20, 2005). It is available on the Cobell plaintiff’s website.
[ h t t p : / / www.i ndi a n t r us t . c o m/ i nde x.c f m? Fus e Ac t i on= Pr e s s Re l e a s e s .V i e wDe t a i l &Pr e s s Re