Deep underneath an industrial
park in Lenexa, Kansas — not far from
Shawnee Mission Lake and Arapaho
Park — the work goes on. In a vast underground
warehouse, historians and archivists
carefully unpack endless rows of boxes. The
cartons, sent from far-flung federal warehouses and
obscure outposts on Indian reservations, contain financial
records collected over the years by government agencies.

Some of the items date to the late 19th century, back to the days
of Indian chiefs Looking Glass, Red Cloud, and Gall. Once
indexed, the documents — mostly leases and bills — are coded and entered into a government computer system. To date, workers
in Lenexa have gone through some 160,000 boxes, processing
nearly 40 million coded pages.

Those pages, sometimes little more than bits of paper, are
pieces of a puzzle — a puzzle that the Department of the Interior
(DoI) dubs The Historical Accounting Project for Individual
Indian Money (IIM) Accounts. The purpose of the massive project
is to reconcile, for the first time, individual trust accounts
managed for decades and decades by the federal government on
behalf of Native Americans.

By the time the project is completed in 2011, the
accounting will have consumed $274 million over
eight years. That dwarfs audits of large publicly traded
companies, which tend to cost about $10 million.
It is, noted former Secretary of the Interior Gale
Norton, “an accounting project of unprecedented
proportions.”

It is also wildly controversial. Elouise Cobell, a
former treasurer of the Blackfeet Nation who has
sued the DoI to force this historical reckoning of the
individual Indian trusts (Cobell v. Kempthorne),
claims that more than $100 billion may be missing
from the individual accounts. But Ross Swimmer,
who heads up the Interior Department agency overseeing
the project (as special trustee at the Office of
the Special Trustee for American Indians, or OST),
believes the amount owed to individual Indians is
far less — more likely in the millions than billions.

The assertion has not gone over well in Indian
country. Some tribal leaders claim the Historical
Accounting Project has nothing to do with history
or accounting. By their lights, the Interior Department
is gaming the system, intent on producing a
low figure — one that will cow litigants into settling
on the cheap. “This accounting was intended to give
trustees a chance to assess their claims,” says Keith
Harper, class counsel for the Cobell litigants. “It’s not
supposed to be for limiting liability.”

Things turned ugly in March. That’s when the
White House offered to pay $7 billion over 10 years
to settle individual and tribal claims. The money
would also help pay for a DoI computer-security
upgrade, extinguish the Cobell class-action suit and
108 tribal suits, and prohibit any future trust-fund
litigation. “The offer was seen by all tribes as an
insult,” notes Michael Marchand, chairman of the
business council of the Confederated Tribes of the Colville
(Washington) Reservation. “It was a clear signal that the Administration
has no interest in settling anything.”

Takeover Strategy

Native Americans have seen this movie before. Administration
after Administration has pledged to resolve the trust issue — with
few tangible results. The difficulty in settling the ledger lies in
the peculiar nature of the accounts themselves.

Typically, those 320,000 or so accounts (from 1938 to 2000)
contain payments from oil companies or ranching concerns —
businesses that must negotiate with the government to gain access to the 10 million acres of individually owned
Indian allotments. (Tribes hold 46 million acres.)
Some documents related to the accounts go back
to the 1887 law that empowered the federal government
to seize Indian lands and then act as trustee
for individual Native Americans (see “Divide and
Conquer” at the end of this article). “There is nothing comparable
in U.S. history, where one sovereign took complete
control of the assets of another group,” says Melody
McCoy, an attorney at the Native American Rights
Fund. “This isn’t a pension fund. Tribes didn’t voluntarily
contribute to it.”

In 1989, at the prodding of the U.S. Congress,
the government hired Arthur Andersen to perform
a full historical accounting of both individual Indian
and tribal trust-fund accounts. Five years and
$21 million later, the firm reported that, due to
missing records and the lack of an audit trail in the
Bureau of Indian Affairs (BIA) systems, it could reconcile
only tribal accounts — and even then, only
those opened after 1972.

The Andersen report did not exactly stun longtime
BIA watchers. The agency, part of Interior, has
long been criticized for its handling of Native American
trusts. The Government Accountability Office
has repeatedly slammed the bureau’s reporting systems.
In 1991, Interior itself characterized the entire
BIA as a material internal-control weakness.

In 1994, Congress passed the American Indian Trust Fund
Management Reform Act, mandating a tallying of the tribal and
individual accounts. But it took a federal court order six years
later to jump-start the project, and the judge on that case has
since been removed at the behest of the Justice Department.

Meanwhile, pressure is mounting on Indian litigants to settle.
Tribal leaders say if the Cobell lawsuit drags on much longer,
many elderly beneficiaries will die before seeing any money. The
board of the United Southern and Eastern Tribes also claims that
the DoI is offsetting some of the costs arising from the Cobell litigation by diverting funds earmarked for basic services on Indian
reservations. Sen. John McCain (R–Ariz.) introduced a bill
that would settle all individual accounts for $8 billion (with no
ban on future suits). The legislation, supported by many tribes,
was suddenly withdrawn from committee consideration in
August, after Interior indicated it wanted to help craft legislation
that would be acceptable to Indian litigants. But as John Dossett,
general counsel for the National Congress of American Indians,
points out, it’s hard for Native Americans to agree to an amount
if they’re not sure what they’re owed. “How do you know if you’re
getting a good deal or a bad one?” he asks.

Accountants have issued qualified opinions on audits of the
Indian trust funds every year since 1996, citing, among other
things, weaknesses in controls and processes. And while lax
internal controls are not unheard of at federal agencies, few
departments have been skewered like the BIA — and fewer serve
as trustees. Even Swimmer, a former principal chief of the
Cherokee Nation of Oklahoma, acknowledges that previous systems
“were not appropriate for the work that needed to be done
for the majority of the time of the trusts.”

Nevertheless, officials at the DoI say the historical accounting
is uncovering few mistakes. They claim that less than one
percent of all transactions reconciled to date have been in error.
Swimmer says he’s not surprised by the results. “It was a manual
system,” he notes, “but it worked.” He adds that the weak link
in the old system was the agency’s reliance on BIA field offices
to handle lease proceeds flowing into the system. “Accountants
found risk,” he grants, “but not losses.”

Critics contend that the department hasn’t found losses
because it isn’t looking for them. The Office of the Special Trustee,
for example, has yet to tackle most transactions dating from the
agency’s “paper era” — that is, before Interior moved to an electronic
ledger system in 1985. The agency also appears to be stacking
the deck in its favor when calculating error rates. It notes that
the absence of sufficient supporting documentation for a transaction
does not imply an error — and “it does not plan to include
such occurrences in the calculated national error rates.”

Fractured Nation

Classifying uncorroborated transactions as accurate does little to
mollify critics. In addition, the OST’s initial work zeroed in on the
types of trust accounts that are easiest to reconcile. These large,
often lump-sum payments (such as court awards) are made to native nations and later distributed to individual tribal members.
“They are the simplest [transactions], the least likely to be subject
to error or fraud,” claims Cobell. In short, “the low-hanging fruit.”

Reconciling land-based leases — for logging rights and the like —
could prove even thornier. Payments into those accounts, which
currently make up more than 75 percent of the individual Indian
accounts, often show up sporadically and in varying amounts. And
thanks to the handing down of land allotments to heirs, the
accounts often have a lot of beneficiaries (a process called fractionation). Currently, beneficial ownership of the more than 320,000
individual accounts is divided among 4 million interests, with some
allotments boasting a thousand beneficiaries or more. Deposits
into these accounts sometimes amount to just 3 or 4 cents.

The OST had planned to perform a broader transaction-by-transaction
reconciliation of individual Indian accounts, including
the badly fractionated land-based
accounts. But, facing time and funding
constraints, the OST has scaled back the
scope of the Historical Accounting Project.
Under the latest scheme, the agency ended
up performing statistical sampling on
land-based transactions valued at under
$100,000. With sampling, the OST takes a
small number of transactions and applies
the error rate to all accounts within specified
“strata.” Those accounts are then categorized
as reconciled.

It’s hard to assess the accuracy of the
approach. The consultants performing the
sampling work, a University of Chicago
offshoot called NORC, declined to be
interviewed for this article. Neal Hitzig, a
professor of accounting at Queens College
(New York), says statistical sampling is
“the way to go when the target population
is so large you can’t audit all the transactions.”
Certainly, a blow-by-blow corroboration
of every IIM transaction would be
prohibitively expensive. Most estimates
peg the cost of such a full-on reckoning at
$13 billion — equal to all the money that
has gone into the accounts since 1910.

But sampling has drawn heavy fire from
various quarters. Some question whether
the sample population is truly random or
whether it has been skewed to limit the
government’s liability. Others say a commercial
trustee would never rely on sampling
to verify the accuracy of account balances. In a report given
on behalf of the Cobell plaintiffs, Richard Fairchild, a former director
of trust policy at the OST, claimed that statistical sampling has “no place in a fiduciary accounting.” Indeed, some in Indian country
think the substitution of sampling for transaction verification
borders on racism. Says Harper, who heads law firm Kilpatrick
Stockton’s Native American rights practice: “The Interior Department
is saying that Indian beneficiaries don’t deserve the same
kind of accounting that any fiduciary would provide to any trustee.”

This Is Not Attest

Then again, accounting engagements rarely examine 120 years of
activity, involve 320,000 accounts, and cover 100 million transactions.
What’s more, an industry standard for fiduciary accounting
doesn’t exist. Auditors and trust managers
tend to follow the Uniform Principal and
Income Act of 1997. But that code was
developed by the insurance industry and
has not been adopted by every state. Says
Ira Herman, director of the estate and trust
practice at J.H. Cohn: “There is no such
thing as GAAP for fiduciary accounting.”

Even critics of the OST’s approach
acknowledge the unusual circumstances
surrounding the Historical Accounting
Project. “This is obviously not a traditional
audit,” says Geoffrey Rempel, a former
Price Waterhouse CPA working for the
Cobell litigants. But, Rempel asserts, “you
can still take the principles in GAS [governmental
accounting standards] and
GAAP and apply them here.”

In truth, it’s tough to tell what accounting
standards the OST is applying. The
DoI touts the phalanx of consulting firms
working on the project, including five
accounting firms. (All declined to be interviewed
for this story.) But the firms are not
performing an audit. Instead, the Historical
Accounting Project more closely
resembles an “agreed-upon-procedures”
engagement. Indeed, internal DoI documents
show that the agency has instructed
the accounting firms to refrain from
using the word audit or even attestation
when discussing their roles in the project.

Mostly, the firms are adhering to something called the Accounting Standards Manual. That lengthy, oft-revised book is extremely detailed. It is also the OST’s handiwork. Thus, the accounting firms are following OST guidelines — and are not offering conclusions about their findings. As Grant Thornton noted in a letter regarding its role as quality-control
agent: “[the firm’s] assessment of the quality of work performed
by…accounting teams performing the historical accounting
is based primarily on [the] Accounting Standards Manual.… This assessment is not an audit and, accordingly, we do not
express an opinion.”

The OST does note that the quality-control firm “tests for
compliance with the American Institute of Certified Public
Accountants standards.” Those guidelines, however, address the
professional conduct of consultants, and not audit standards.

Location, Location, Location

Still, the absence of a GAAP audit does not mean that the numbers
produced by the Historical Accounting Project are off. Nor
does it mean that lease money intended for Native Americans
found its way into the wrong pockets. “When I started working
with Cherokee Nation in 1972,” recalls Swimmer, “I met a lot of
BIA people. They took a lot of pride in what they did. And they
weren’t out to try and cheat anybody.”

Without a good paper trail, it’s hard to say for sure if Native
Americans have received all monies due them. DoI officials say the
effort in Lenexa is providing the necessary documentation to corroborate
transactions. In a recent court filing, the Interior Department
noted that “[w]ith few exceptions, the Indian trust records
needed for the historical accounting exist, are being located with
success, and are being used to reconcile IIM account transactions.”

Critics charge that the standards used by the OST to confirm
transactions fall far below what you’d find in the private sector.
DoI memos do reveal that the department advised accountants
working on the project to use “alternative procedures,” as well as
their own judgment, to verify transactions in the absence of moresolid
documentation. “When you peel off the veneer,” insists Rempel,
“the procedures they’re using could validate anything.”

A canceled check, for instance, is apparently not required to
prove that a payment was made by a lessee. This surprises J.H.
Cohn’s Herman: “As a fiduciary, you have the absolute responsibility
of knowing every single receipt and every single disbursement,
and you want to know every document that’s associated with it.”

The OST could have difficulty finding every document. Some
records-warehouses on reservations were little more than glorified
sheds. In 1999, Treasury reported that 162 boxes relating to
the Indian trust-fund litigation had been accidentally destroyed.
Apparently, the boxes contained government forms reflecting
disbursements made by federal agencies from 1900 to 1958.

The revelation did not exactly calm fears in Indian country
about the veracity of the Historical Accounting Project. “If the
normal bar for trust accounting is 100, this project started out at
50,” claims Harper. “Now it’s at 2. Ultimately, it will be down to
0.” He pauses. “This whole thing is an exercise in nothing.”

On September 10, the DoI said it would not comply with a
federal judge’s order to turn over electronic trust data to the
Cobell plaintiffs.

In Lenexa, the trucks keep rolling in. The work continues.

John Goff is a senior editor at CFO.

Divide and Conquer

While the federal government has been managing tribal
lands since the 1820s, it didn’t take over as trustee for individual
Indians until 1887. That’s when the U.S. Congress passed
the General Allotment Act, aka the Dawes Act. That legislation, ostensibly an attempt to assimilate Native Americans, gave each member of a recognized tribe up to 160 acres of land, to be held in trust by the government. The government then sold off the surplus acreage. By the time the allotment period was halted in 1934, Washington had auctioned off roughly 90 million surplus acres — nearly two-thirds of Indian country — to non-Indians.

The remaining 56 million acres (10 million for individual
Indians and 46 million for tribes) still make up one of the largest
land trusts in the world. Oil leases and grazing rights, among
other things, generated about $800 million in proceeds in fiscal
2006. The money went to 250 tribes and roughly 4 million
individual Indians.

Not surprisingly, the Administration would like to get out
of the Indian-trust business, handing that responsibility back
to tribes. Some tribes, such as the Confederated Salish and
Kootenai Tribe in Montana, already manage their leases.

But not all tribes want to see the government abandon
its role as trustee. They point out that on the Great Plains, for
instance, leases come mostly from grazing rights. Negotiating
a single lease with dozens of lessors can get extremely complicated.
Says John Dossett, general counsel for the National Congress
of American Indians: “The allotment holders rely on the
Bureau of Indian Affairs to pull together a whole lot of allotments
into a single grazing range.”

What’s more, some tribal members believe poorer
tribes — which is to say most of them — do not currently have
the resources or trust expertise to manage the leases. They believe the United States still has an obligation to look out for
the interests of those tribes, many of which gave up their land
at gunpoint. “Millions of acres of land were transferred to the
U.S. for settlement, and in return the U.S. promised to provide
health care, education, and economic development,” says
Michael Marchand, chairman of the Business Council of the
Confederated Tribes of the Colville (Washington) Reservation.
“This is what established the trust relationship. Now the U.S.
wants to [renege] on these agreements.” — J.G.