diligence and insight but criticizes him
for the management lapses that ultimately cost him his company. The PNC deal
was one of several accounting scandals,
the most famous being the 2000 Gen
Re deal that sent five of its architects to
prison. Against the scandals of Enron
and WorldCom and the bursting of the
dot-com bubble, the maelstrom proved
Hank’s undoing.
But his specter remained. He was
the focus of an Eliot Spitzer investigation (trumped up, in Boyd’s telling), and
mutual acrimony led to a series of lawsuits between Hank and the company he
had built. Complicating things, because
of AIG’s labyrinthine structure, Hank remained a key ( 12 percent) stockholder.
With Hank gone, the empire that
only Hank could run fell to Martin Sulli-
van. Boyd portrays him as a cipher who,
even after his accession, couldn’t call his
disgraced boss and mentor anything less
adulatory than “Mr. Greenberg.”
Boyd may be unfair to Sullivan, though
he is not Sullivan’s only critic. A company
so complex requires a gentle handoff, one
that Boyd reports Greenberg had con-
templated before the scandals erupted.
Sullivan received none. Instead, his men-
tor became his legal enemy.
Greenberg left him with a mishmash
of companies and with systems that gave
scant clues on how the firm was oper-
ating. Preparing quarterly earnings,
according to Boyd, was “mayhem in its
purest sense.” Robert Willumstad (Sul-
livan’s brief successor) contrasted AIG
with his old company:
“Within five business days before
the quarter’s close, we could go to the
board at Citi and tell them within one
penny where earnings would be. I had
heard of companies where you wouldn’t
know until the night before what earn-
ings were—I had not thought AIG would
be” one of them.
Sullivan’s early days were consumed
with the legal detritus of Hank’s leadership. The company paid $1.5 billion in
While some AIG
units—notably its
American General
Finance Unit—foresaw
the mortgage crisis,
the information was
stuck in a silo.
fines to escape being labeled a criminal
enterprise. Still, under Sullivan, early
results dazzled. Operating income was
$15.2 billion in 2005, remarkable for a
company under such scrutiny. But with
Hank—the ultimate risk manager—gone,
minions tested the new order. And their
biggest bets all went the same way—
predicting that the mortgage market would
continue to be strong.
A Failure of Risk Management
AIG’s investment arm began ramping
up its securities lending program—
lending the company’s securities for short
periods, then investing the proceeds
in mortgage securities. The Financial
Products unit sold swaps that guaranteed that mortgage holders would pay off
their debts. And it wrote guarantees that,
should a collateralized debt obligation
fall in value, AIG would put up the cash
for the amount it fell—essentially fulfilling a deal’s margin call. Boyd writes that
few outside the Financial Products unit
knew about these guarantees, which implies the audit function at AIG may have
been weak in this area.
Other AIG units—notably its American General Finance unit—foresaw the
mortgage crisis. They curtailed their
writings, but there was no way for them
to relay their concerns up the corporate
chain of command. The information
was stuck in a silo—a classic risk management failing.
Instead, writes Boyd, “Greenberg had
set each unit up to be almost totally au-
tonomous. The only thing that bound it
all together was Greenberg and his com-
mand of every last detail.”
Financial Products, meanwhile, was
shunning the rigorous diligence that had
marked the company’s earlier trades. In
the early days, slam-dunk deals—like the
JPMorgan swap—would take weeks to get
signoff. The first mortgage-backed swaps
were rubber-stamped within hours.
