How did JPMorgan lose $2-billion in six weeks?

http://www.theglobeandmail.com/globe-investor/investment-ideas/streetwise/how-did-jpmorgan-lose-2-billion-in-six-weeks/article2429647/?

Losing $2-billion (U.S.) in six weeks is quite the feat for JPMorgan Chase & Co., (JPM-N40.740.100.25%) the venerable U.S. bank that was, until Thursday, widely respected for sound risk management.

Now not only must shareholders stomach the gaping loss, they must also live with a management team that admits it screwed up.

Unlike past trading scandals, this wound wasn’t inflicted by a rogue trader. JP Morgan, as a bank, is at fault, and chief executive officer Jamie Dimon was contrite about his team’s massive mistake.

But how exactly did the bank get itself into this mess? And what trade blows up this quickly? JP Morgan isn’t saying much – likely because the more it discloses, the more its rivals will have to use against it. But keen market observers are trying to figure it out, and the speculative theories narrowed down over night.

To understand the trades, you have to start by looking back a few weeks. In early April, Bloomberg News reported that Bruno Iksil, a trader in JP Morgan’s chief investment office, had placed massive bets on the health of certain companies. His positions were so big that rival hedge funds said they were starting to distort the market.

Mr. Iksil works for JPMorgan’s chief investment office, which is led by Achilles Macris. This division, based in London, is supposed to hedge the bank’s risks, but many people say that its ‘hedges’ are in fact risk trades designed to duck the proposed Volcker Rule that bans prop trading.

When news of this division’s massive positions broke, JP Morgan didn’t say much. Then out of nowhere it hastily put together a conference call on Thursday to disclose that the CIO lost $2-billion in about six weeks.

The first explanation for the losses relates to what is known as the basis trade, a hedging strategy for corporate debt. Assume you own a 5-year bond. If the issuer goes bankrupt, you never get your principal back. To protect yourself, you can purchase a credit default swap (CDS) that mimics an insurance contract. Just like your car insurance, you pay a premium to the swap writer, and in return, if you car gets stolen, or in this case the company goes bankrupt, you get paid a lump sum.

But for this hedge to work, the corporate bond spread must match the CDS spread. Think of this as the interest rate paid to you matching the insurance premium that you have to pay. If they line up, your next exposure is nil.

Historically, this held up. If anything, investors had to pay just a smidgeon more in premiums than they received in interest, and this small difference was known as positive basis. This excellent 2009 post from Zero Hedge has a much more detailed explanation.

Every once in a while ‘negative basis’ would appear for random reasons, creating risk-free profits because interest payments received amounted to more than the insurance premium doled out. When this happened, big name hedge funds and prop trading desks would pile in and exploit the anomaly. To them, it was free money.

The problem, though, is that this trade is predicated on bond spreads remaining equal to CDS spreads. As Felix Salmon points out, the market is currently very screwy, and many CDS spreads are blowing out while the bond spreads hold tight. UBS’s CDS spread is 85 basis points higher than its corporate bond spread, and Deutsche Bank’s spread difference is 83 basis points.

Mr. Salmon believes this dislocation is what got JPMorgan in trouble. The way he sees it, the bank was selling insurance protection, betting that CDS spreads would go down rather than up. Ultimately, they went up, and the corresponding spreads didn’t.

Think of it this way. If JPMorgan was using this trade as a hedging strategy, it means the company was probably short corporate bonds. (By going short, the banks benefits when the bonds do badly, while in the CDS market it suffers because has a higher change of paying out.) But because JP Morgan was selling insurance protection, the lower CDS value from wider spreads wasn’t offset by any bond spread movements.

JP Morgan wouldn’t be the first U.S. financial institution to get dinged by the basis trade. Merrill Lynch posted a $16-billion loss in the fourth quarter of 2008 for the same reason.

The second theory is much more complex, and you can dig into it in more detail here. But the main idea is that JP Morgan may have tried to hedge a position by placing various trades on CDX Series 9, an index of liquid credit default swaps. As the market rallied over the past few months, it forced JP Morgan to increase its position with this index. This resulted in massive mispricing because the bank made the index more expensive that the sum of its constituents.

“This is akin to looking at the 500 names in the S&P 500 – weighting them and seeing that the S&P 500 index should trade at 1,200 but it is trading at 1,400…” Zero Hedge noted.

If this played out, hedge funds would have tried to short the index because they knew it was too expensive. After they did, the index should have fallen right away, yet JPMorgan kept supporting it because it needed the hedge. The longer this went on, the more questions the hedge funds asked, and eventually Mr. Iksil was discovered.

After building this position, JPMorgan could have gotten in trouble because the market stopped rising. The problem is that it couldn’t unwind its position because its rivals have figured out what it’s up to.

That’s why Jamie Dimon may have said that more losses could come. If he’s stuck with these trades, hedge funds are going to do everything they can to make sure JPMorgan can’t get out of them.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s