Friday, May 11, 2012

JPM: Oops...

OK, so since I am a big Dimon and JPM fan, how can I not post something on this?   So JPM held an emergency conference call to announce that they have $2 billion in losses and people seem to be going crazy.  The Volcker rule crowd, of course, love this.  "Told ya so!".

People keep saying "huge" loss, and yes, $2 billion is big.  Not too long ago, a $1 billion loss would have blown up a whole bank.  LTCM was single digit billions in losses.    I think Barings blew up on a billion dollar loss. 

Since people keep saying $2 billion with no context, I figure I'll put that into context.

$2 Billion in Proper Context
JPM has total assets of $2,300 billion, shareholders equity of $190 billion and tangible common equity of $129 billion

In 2011, JPM had net revenues of $97 billion, pre-provision profits of $34 billion and net income of $19 billion.

Corporate / Private Equity Segment

                 ($mn)                                                             Total
                Principle                Securities                          net
                transactions           gains                                  revenues
2006        1,181                        -608                                          14
2007        4,552                           39                                     4,419
2008       -3,558                      1,637                                          69
2009        1,574                      1,139                                     6,634
2010        2,208                      2,898                                     7,422
2011        1,434                      1,600                                     4,143

So I just grabbed some figures from the annual reports/10K for the Corporate/Private Equity Segment, the segment that incurred the losses.  The total of principle transactions since 2006 were $6.2 billion, securities gains were $6.7 billion and total net revenues were $22.7 billion.  The columns won't sum as I left out other income and net interest income etc. for simplicity.

The above figures include gains in the private equity portfolio, gains from Mastercard (or Visa) IPO, losses on FNM/FRE preferreds and things like that. 

The point is that a $2 billion loss is not a large event even within this segment.  Gains and losses in this segment typically do NOT drive the stock price and earnings expectations of JPM as a whole. Sometimes they have big gains, and sometimes they take big losses. Nobody upgraded the stock when they booked a $1 billion bankruptcy settlement gain in the first quarter, for example, nor were they shocked when they booked a $2 billion litigation reserve last year.

$800 million loss in 2Q After Tax in the Corporate/Private Equity Segment
So the loss is now $2 billion (or $2.3 billion) and Dimon says that will be offset by a $1 billion gain on the sale of securities.  But from a modelling point of view, I think it's OK to look at the $2 billion as a $2 billion loss (and not net the $1 billion against it).   It's good from a tax point of view as you realize a gain to offset losses, but from a shareholders standpoint I think it just turns a $1 billion unrealized gain (presumably in accumulated other comprehensive income (AOCI) on the balance sheet into a realized gain that goes through the income statement).

But an $800 million or $1 billion loss in this segment for a quarter is not that big a deal when you look at it in perspective.

Levin Comment
Not so bank-friendly Senator Carl Levin said that this JPM trade was a bet on the direction of the general economy.  I suppose you can say any transaction is a bet on the economy; if you are long credit you are bullish and if you are short, you are bearish.   So how can you tell the difference between a bet on the economy and a hedge?  Levin said this was not a hedge (how does he know?).

What I heard Dimon say on the conference call and haven't heard or read much about is that this was actually a transaction to reverse previous hedges.

He said that the CIO (Chief Investment Office) invests the cash of JPM and enters into transactions to hedge exposures that the company has overall.  Obviously, JPM is sitting on a lot of credit risk so it's not unreasonable to put on some offsetting hedges at the corporate level with the extra cash sitting around. 

JPM had hedges on to hedge the tail risk of their credit exposure.  It was designed to make money in a "credit stressed environment", Dimon said.  As the markets and economies recovered, they decided to lighten up and reverse some of those hedges.  And obviously, what happened was that instead of unwinding these transactions, they put on offsetting positions instead. 

Why would they do that instead of unwind?   

One reason would be taxes:  If the hedges were profitable, unwinding may incur taxes.  But I don't know if that applies to derivatives and synthetic transactions.  This would apply to cash securities holdings (sometimes it's better to offset a profitable long with a short hedge than to sell it and realize taxes).

The other reason would be that the offsetting transactions were priced more attractively than unwinding existing positions.   Go long similar credit to the ones they are short but at a better price (and then live with the basis risk or correlation risk).

So would this be a bet on the economy?  Not really.  This can still be within the context of hedging the total credit exposure for JPM as a whole.  There are probably a lot of side-bets within these portfolios, but the overall goal is to manage the risk of the whole firm.

If every time you short something, you are considered "betting against the economy" and you unwind a hedge, you are "making a bullish bet", then how can you ever hedge?

This also reminds me of a post I made a while back about how portfolios can get very complicated by overlapping/superimposing all sorts of things to 'protect' or 'adjust' the portfolio to manage risk.

(Read the Rube Goldberg portfolio post here.)

An equity portfolio example would be shorting S&P 500 futures to hedge an equity portfolio and then deciding to buy S&P 100 (OEX) combos (synthetic futures; long calls, short puts of the same strike price) to reduce the hedge (because maybe it was trading below fair value).    The S&P 500 futures short will have worked well as a hedge and then buying OEX combos would reduce that hedge as you are going long another equity index.  But if there is a large tracking error between the two indices, then your hedge calculation might be off (I can't imagine it doing too much damage in this example unless someone was actually hedging specific S&P 500 index risk with the S&P 100 index (going short an index swap and then going long a different index to hedge that can lead to big losses if the correlation/tracking error assumption is wrong)).

Someone managing an equity portfolio can quickly make their 'book' more complex.  To continue the above example of hedging with S&P 500 futures, maybe he wants to lighten up the hedge but decides the Nasdaq is cheaper, so enters a long Nasdaq 100 index swap.  The market rallies and it looks a little pricey so he wants to increase his hedge, but sees more risk in financials, so maybe shorts some Financials ETFs.  And then the market goes down, he wants to reduce the hedge but sees value in energy stocks so goes long a swap on some energy stock index.  But then he realizes the inherent exposure to crude oil prices so he shorts crude oil futures.  But then crude goes down so he wants to cover some of that hedge but sees that natural gas is cheaper so covers some of his energy shorts by going long natural gas futures.


If you call directional exposure delta, then I don't think it's unfair to say that these guys often spend a lot of their time trying to find 'cheap' delta to buy and 'expensive' delta to short (and maintaining their target delta).

So you can see how this can become quite complex pretty quickly even though the net positions might look like one thing offsets another.  You can say the stock shorts offset the stock longs, and the crude oil shorts are offset by natural gas longs etc...   But...  

And the scary thing about the above example is that on every transaction, it seems like the intent was to *reduce* the risk in the portfolio. But as we often find out, we are often increasing risk, not decreasing it.

Anyway, that's just an example.  Hopefully nobody runs such a complex book like that.  I know that similar things do happen all the time, but maybe not all in one book at once.

This is just an example of the sort of thing that may happen in a hedging book.  I obviously have no idea what goes on in JPM's CIO book but I imagine it's a simliar thing just with credit products (bonds, swaps, credit swaps etc...).

Economically Non-Event But...
From a strictly economic point of view, I don't worry about this at all.  Sure, it can get worse and the sharks are circling so I'm sure the prices are moving against them as the street tries to force JPM's hand.

By the way, I think one reason why Dimon held the conference call (one analyst said the loss is too small to hold an emergency conference call) was to warn the market (sharks) that JPM's hand will not be forced. 

More than once, he said they will not do anything stupid, meaning they are not going to go into a stop-loss liquidiation at any price or panic-sell or cover anything.  If they have to, they will hold onto the positions for the longer term even if it's volatile.  They have the balance sheet and liquidity to do so.

If they did nothing, rumors of multi-billion dollar losses would have spread, hedge funds and others would pile into opposing trades to try to hit JPM's stop-loss trigger or something and it could be much worse. 

And I am not too worried about JPM risk management overall.  JPM is a great shop but they, like anyone else, will make mistakes.  So they made a mistake.  They'll get over it.  I think this emperor talk is nonsense (emperor has no clothes).

It's interesting to see some TV pundits just dump Dimon on one thing like this (that is not even economically significant) saying, "I trusted him but I was wrong". 

Even though I think things will work out economically, it is a total nightmare politically.  This really gives a lot of ammo for the "break up the big banks" and Volcker rule proponents.  

This may be what the street is more worried about than the actual $2 or $3 billion loss that JPM can digest. 

I think this will turn out to be a non-event over time  (How can a 1% loss of net worth cause such a firestorm!?).  Of course, this isn't over so the losses can turn out to be much larger but since they are willing to hold the positions for a longer time, I get the sense that the pricing is out of whack so they'd rather sit out this 'volatility', bear it on the balance sheet and then get out over time at reasonable prices.   This is not Lehman, Bear Stearns, LTCM or anything like that.

I do still like Dimon and JPM, but as I said before, financial stocks are hard to own as they will go up and down a lot according to the headlines and these occasional surprises as they are leveraged black boxes.   (So if you own any, make sure you don't own so much it will make you throw up if Europe imploded, for example)

Oh, and I know there is a large population out there (majority, maybe) that will disagree with this post.  They will roll their eyes and call me a 'muppet'.  I know.  That's OK.   I don't do this to change people's minds or argue.


  1. This will turn out to be a non-event but the bigger issue is of the PR disaster. Jamie Dimon is the face of Wall Street in Washington and one of the most outspoken critics of financial regulation. This clearly does not help his case. But good in depth analysis, thank you.

  2. Your points are very well balanced and in context to real numbers , as opposed to all the noise that is bandied about by the media and politician .
    It is a shame that the reporting in the press has been so biased and one-sided .
    Thanks for objective explanations and writing

  3. One other thing ,

    as a practitioner of running long/short books with derivative exposure and managing 7 trading desks over 25+ years .... it is extremely difficult to hedge out everything , as the changing Deltas , Gammas , IV's , etc. , never move the way you want or model ...
    many times , the best thing to do is to let the positions wind down to their contract expiration

    FAS 157 is the mark-to-market accounting tool , and derivatives are marked all the time , while the cash (underlying) , doesn't have to be

  4. It is still uncertain how much JPM will loss from these positions. And there are so much rumors out there speically from the short sellers. The buying oppotunity may get even better as the panic sell continues.