Why JP Morgan’s CIO found it so easy to make money

By Felix Salmon
May 16, 2012
proof that JP Morgan was -- is -- using its Chief Investment Office to gamble with taxpayer-backstopped funds?

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

You want proof that JP Morgan was — is — using its Chief Investment Office to gamble with taxpayer-backstopped funds?

The CIO unit also had a lower cost of capital than other parts of the bank, an artificial advantage that gave it an incentive to take more risk and behave in a less disciplined way, people familiar with the unit said.

“It was very large, but was never very transparent, and it wasn’t clear that they had an appropriate funding cost,” said the source with direct knowledge of the CIO.

In any unit of any bank, one of the key drivers of profit and loss is the internal cost of funds. If you’re paying 1% for your funds and earning 3%, then you can claim profits of the difference, 2%. But if your cost of funds is increased to 2%, then your profit is halved at a stroke. For someone like Ina Drew, who was charged with turning hedges into profitable trades, the easiest way to do that would always have been to simply get Jamie Dimon to decrease the CIO’s cost of funds.

And at JP Morgan, just like at any other bank, the cheapest cost of funds is always deposits. JP Morgan has hundreds of billions of dollars in excess deposits just because it’s too big to fail, and has an implicit government backstop. It’s bonkers that it should then be able to take the resulting ultra-low cost of funds, and turn it into eight-figure bonuses to people like Drew, all for taking that money and playing on derivatives indices in London.

As John Macaskill points out, the CIO, by its own faulty measurements, had for the past two quarters more money at risk than JP Morgan’s entire investment bank — and that was with a more lenient risk measurement and with a lower cost of capital. In reality, the CIO’s risk levels were vastly greater than those at the investment bank, as we discovered after the blow-up.

If JP Morgan wants the CIO to be taking that kind of risk, it has to significantly increase the CIO’s internal cost of funds. The CIO is at heart a hedge fund (it’s designed to put on hedges), and JP Morgan should extend it billions on the same kind of terms that it would extend money to top prime-brokerage clients. The CIO’s secret weapon, all these years, has been its artificially low cost of funds. If that number were more realistic, maybe JP Morgan wouldn’t have ended up parking such an insanely enormous amount of money there.

Comments
9 comments so far

As I was saying a few weeks ago when we were skirmishing about it in the comments, these people do not actually deserve the salaries they are paid.

Their pay isn’t actually being set by a functioning market, and instead is just favors for favors. Eight figures for turning a profit with a 1% cost of funds?

90% of what the financial sector does is actively detrimental to the overall economy and it needs to be much more heavily taxed, and/or regulated. It is rent seeking perfected to a science. Encouraging the sloshing of money as much as possible so as much as possible can be skimmed out.

All for mansions on Long Island and other frivolous goods.

And I am someone who likes markets! Imagine what the communists think!

Posted by QCIC | Report as abusive

I think you’re conflating two issues. One is the bank’s internal accounting – how they decide, given that at the end of the year they’ve made $X profit, where it came from so they can allocated a large chunk of it to bonuses. That should be an issue purely for management and shareholders.

The other is how much risk they should be allowed to take at some level of reserves, given the taxpayer backstop for deposits. That’s where the Volcker rule matters.

I fail to see why we (the taxpayers) should care about the bank’s internal allocation of its total risk, and corresponding allocation of cost of funds.

Posted by FosterBoondog | Report as abusive

The simplest way to get banks back under control would be for the FDIC to remove deposit insurance protection from any bank that engages in proprietary trading. That would wipe out most of the cheap money used for this reckless speculation and do it very effectively.

Posted by Chris08 | Report as abusive

The simplest way to get banks back under control would be for the FDIC to remove deposit insurance protection from any bank that engages in proprietary trading. That would wipe out most of the cheap money used for this reckless speculation and do it very effectively.

Posted by Chris08 | Report as abusive

@Chris08 – The failures of Lehman and Bear Stearns did not involve the FDIC guarantees. Prop trading creates systemic risks that are not necessarily related to traditional bank runs by depositors. The shadow “banking” markets created by repo and derivatives requiring collateral posting are subject to runs without having explicit gov’t backing for anyone. That’s what TBTF is all about. Even without explicit guarantees, there’s the implicit guarantee that the Treasury and/or Fed will provide backstops to prevent financial meltdown, and therefore a bank’s counterparties are more willing than they otherwise would be to lend large sums to opaque organizations.

Posted by FosterBoondog | Report as abusive

Maybe we need to borrow for a while from the Singapore method of bank regulation of the 80s – “everything is forbidden except that which is specifically approved”.

Posted by MrRFox | Report as abusive

JPM made some complex trades. Don’t let the complexity hide the simplicity:

You CANNOT UNDER ANY CIRCUMSTANCE PROFIT FROM A HEDGE.

Hedges are insurance they help limit your losses. Hedges… ALL HEDGES…. COST MONEY.

If you want to read how stupid this coverage is copy any story into MS word and replace “hedge” with “insurance” every time it appears. If you do that the stupidty just stares you in the face.

A stand alone division designed to hedge risk would alwasys be a loss center for the bank. In many cases the hedges would be “profitable” but never moresoe than the underlying asset lost. If you have more insurance than underlying assets you don’t have a “hedge” you have a short.

Dozens of people have said this dozens of times but it’s just not getting the attention of the financial media… probably because a very well respected Jamie Dimond dosen’t have the balls to admit that it was a directional bet.

Posted by y2kurtus | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/