on Friday in
, American Banker's blog about ideas, trends, and other
developments in financial services.
Let's say I offer you a bet on the flip of a
. I'll pay you $2 for every $1 you bet if the coin lands on heads.
If it lands on tails, you lose your bet. But we'll do this only
once and you have to bet your entire net worth. Would you play?
This is a bet with very large positive statistical expectation
to you per dollar wagered, but it's so risky you'd be justified in
taking a pass. And you'd be crazy to take the bet if I paid you
only 75 cents for every $1 you bet. You'd be making a "sucker bet,"
one that offered insufficient odds.
It's easy to calculate the expectation of the above bets, harder
in games of imperfect information, like poker, and even harder in
business, where success can be framed as inevitable in retrospect,
and failure framed as impossible to have foreseen. But people who
gamble for a living are adept at thinking in probabilistic terms.
If they bet on a long shot, they expect to make multiples of their
bet should they win. A trader is just another type of professional
gambler. And for all their complexity, the models traders use are
trying to answer one conceptually simple question: is the trader
getting a fair or rich price to take on some risk? If not, he or
she should take a pass.
A lot of ink has been spilled on JPMorgan Chase's (
) $6.2 billion "London whale" trading loss since it occurred just
over a year ago, but as far as I can tell, no one has written about
its apparent cause: a large, crazy bet JPM made in 2011 that just
happened to pay off. And that means no one has considered the
JPM's Synthetic Credit Portfolio, housed in the bank's Chief
Investment Office ("CIO"), made $453 million in 2011, and this
trading gain made the CIO eager to employ its winning trading
strategies on a greater scale. How did it do this? According to the
that the US Senate's Permanent Subcommittee on Investigations
released March 15, in the fall of 2011, one CIO trader, Bruno Iksil
(later to be known as "the London whale") spent about $1 billion
buying credit protection that would expire worthless in four months
or less unless a corporate default occurred among 100 higher-risk
companies. Three weeks before the bet's expiration, American
Airlines declared bankruptcy, and the portfolio made $400 million.
It would have lost money for the year otherwise (see pages 53-56 in
So to recap, the CIO bet $1 billion and earned $400 million. In
expectation terms, that means unless the CIO believed there was at
least a 72% chance of default, or unless multiple defaults would
have led to meaningfully bigger profits, its sizable bet had
negative expectation. A sucker bet. Did Iksil think a default was
more than 72% likely? And even if he thought default was that
likely, shouldn't he still have feared the giant loss he'd incur if
he were wrong, like in the coin flip discussed above?
An internal memo from the Senate report's exhibits shows that
CIO head Ina Drew downplayed the riskiness of this bet when she
explained it to JPM Chairman and CEO Jamie Dimon (Exhibit 84a).
That buys him political cover. With JPM's 260,000 employees and
2011 earnings totaling $19.0 billion, Dimon shouldn't be held
responsible for the actions of a single employee. Or should he? Few
individual employees can affect JPM's earnings number on their own.
Arguably Dimon can't in the short run. But Iksil could; his
"winning" bet comprised around 1% of 2011 earnings. Which implies
Dimon should have been watching. But Drew, who was paid $14 million
in 2011, and whose job it was to watch Iksil, wasn't angry about
this trade. She was a fan of it (report, page 55). She pushed Iksil
to replicate it in 2012, and was quoted as saying that the CIO
"likes cheap options" (report, page 63). That ultimately drove the
CIO to lose $6.2 billion, or 17% of what JPM earned in 2012, in a
benign trading environment.
Why did JPM make enormous high-risk bets, and bad ones at that?
In poker, as in trading, luck plays a role. Professional poker
players know to separate skill from luck. A pro who misplays a big
hand but is saved by a miracle card might breathe a sigh of relief,
but he doesn't pat himself on the back, because he knows that if
all he has to rely on is luck, he'll be a long-term loser. Were
Drew and her team familiar with this concept? Is Dimon?
There have been rumblings that Dimon might resign should JPM
shareholders vote next week to strip him of the Chairman role. JPM
shareholders should take that risk seriously. JPM's stock price has
grown 28% since Dimon became CEO, whereas it fell 17% during prior
CEO William Harrison's six and a half year tenure. But JPM's
all-time high of $65.67 occurred on Harrison's watch, in March
2000, when the bank had $391 billion in assets, versus its current
$2.39 trillion. On Thursday, JPM closed at $50.97, 22% below the
high of 13 years ago. Wells Fargo (
) and US Bancorp (
), both large, high-performing banks, are far closer to their
respective all-time highs - highs which occurred in 2008.
The Chairman/CEO split issue is a sideshow. Dimon is
unquestionably among the best CEOs in financial services, if not
the best. But he hasn't been able to get JPM's stock price above
where it was in 2000. Is this a reason for why the firm was (or is)
To put it another way to those who are especially nervous about
Dimon's potential (although at some point inevitable) departure:
what big opportunities can only Dimon help JPM exploit, and what
big mistakes can only Dimon keep JPM from making? He's spent over
seven years improving JPM as CEO. After all that time and effort,
is he still the only person capable of running the bank
I have no positions in any stocks mentioned, and no plans to
initiate any positions within the next 72 hours. I wrote this
article myself, and it expresses my own opinions. I am not
receiving compensation for it. I have no business relationship with
any company whose stock is mentioned in this article.
What The Market Missed On Kandi