Skip to main content

JPMorgan Finds A Better Use For Its Commodities Brainpower Than Selling Electricity

  • Author:
  • Updated:

UBS is selling its over-the-counter commodity derivatives portfolio to JPMorgan, prompting John Carney to say this:

Here's a good rule of thumb. When one bank buys a business from another bank, it's almost always a case of regulatory arbitrage. It's never really because of synergies or managerial talent or whatever other hokum the media relations churn out to their willing dupes in the press. It's just about one bank being better able to take advantage of the rules.

So even though the rationale for JPMorgan Chase buying the over-the-counter commodities derivatives business of UBS remains mysterious, you can safely surmise this is regulatory arbitrage. Most likely, it's got to do with capital requirements.

Umm maybe? I don't know, this question seems a little over-determined; the thing is that pretty much everyone thinks that (1) JPMorgan is pretty good at running an investment bank, the occasional hiccup aside, and that (2) UBS is pretty crap at doing so. So are US regulators relatively more comfortable with JPM managing this portfolio than Swiss regulators are with UBS doing so? Sure, probably, but probably so are the respective shareholders, and counterparties, and senior managements, and anyone else you might ask. Really moving any portfolio of anything from UBS to JPMorgan is probably Pareto optimal.

The light irony comes from - well here is Bloomberg's first sentence:

JPMorgan Chase & Co., which said last month it plans to get out of the business of owning and trading physical commodities from metals to oil, bought the over-the-counter commodity derivatives portfolio of UBS AG, which is exiting most of its raw-materials trading.

Hahaha you're buying all these commodity derivatives at the same time you're getting out of the physical commodities business, jerks. Banksters up to old tricks etc. But it's interesting to think about why a commodities business that is after all run by the inventor of synthetic CDOs might be perfectly happy to expand its derivatives business at the same time it's running away from physical commodities. Regulatory pressures, sure, but also a certain compatibility of mindset.

One of my favorite recent pieces of financial journalism is this article in Risk1 about how Goldman made a ton of money on the change from Libor to OIS discounting for derivatives. Basically:

  • Everyone used Libor curves to discount cash flows on swaps, etc.
  • Goldman realized, in around 2005, that it should be using the OIS curve.
  • So it did.
  • Which produced higher valuations for in-the-money positions, especially when the Libor/OIS basis widened starting in 2007.
  • Which allowed Goldman to pay more for in-the-money positions than their trading value, though less than their "actual" value.
  • So it did, aggressively buying ITM positions and offloading OTM ones to take advantage of the proper discounting.
  • Eventually the market figured this out and the market price of Goldman's positions moved to reflect their (higher, "correct") internal marks.
  • Leading to Goldman booking ~$1bn in profits on the trade.2

There's a lot more to it - disputes over collateral, CSA amendments, cross-currency effects, etc. - but that's the main joke. And much of the interest in the article comes from figuring out whether Goldman was laughing with or at various people, with Goldman and at least some of its clients adamant that Goldman brought its clients in on the joke.3 But there is this, from a swaps trader at another bank:

Between dealers, all bets are off. If they don’t get it, then that’s tough. If two dealers fundamentally are valuing the same instrument differently, then there are trades to be done. If someone thinks the price is X and the other thinks it is Y, then that is what creates markets. But when dealing with clients, you have to sit down with them and explain why prices are different, and why you are asking to novate or unwind trades.

"If they don't get it, then that's tough," more or less has to be your motto in at least some of your trades in complex instruments. You can quite sensibly - as this guy did - draw the line between interdealer trades (everyone's a big boy, if they don't get it it's tough) and client trades (you're responsible to them, if they don't get it you have to explain it), but someone's gotta be on the that's-tough side of the line. It makes no sense to invest a lot of time and money in being smart if you don't use it to take advantage of the stupid.

How'd that work out for JPMorgan in physical electricity? Quite well, was the answer, for a while, followed by quite terribly. JPMorgan came across a system designed by idiots and were like "ooh, idiots, sweet! Idiots are a great source of money!" And so they made money. "If they don't get it, then that's tough," they figured was a reasonable motto for dealing with the California Independent System Operator. After all, it was CAISO's own rules that they were exploiting. If anyone should be responsible for making those rules work, it was CAISO, not JPMorgan. JPMorgan's traders were just there to make money.

But it turns out that ruthlessly exploiting stupidities in CAISO's rules is not cool with regulators, or most people, and JPMorgan ultimately paid $410 million in fines for that exploitation. Because CAISO is not just an if-they-don't-get-it-that's-tough counterparty. CAISO is, y'know, the mechanism on which millions of people in California rely to get reliable and reasonably priced electricity. Messing with it is not okay, even if deep down you think CAISO deserved some messing with.

Or take this aluminum thing. I don't have the foggiest understanding of it! But a useful exercise is to read the perplexing Times story about how Goldman is making people wait a long time for their aluminum, and then read Izabella Kaminska's explanation of the aluminum-warehousing trade, and notice that they use none of the same words, pretty much, except maybe "aluminum." The banks are all "I will provide financing to customers who want to put on profitable but balance-sheet-intensive contango trades," and the aluminum users are all "MY BEER COSTS AN EXTRA FIVE BILLION DOLLARS," and those are just incommensurable. And then one day your contango-trade financing ends up on the front page of the Times for making beer $5 billion more expensive. And then you have to stop doing that.

The point is that when you make your living trading derivatives then a lot of your business is about noticing stupidities in market structure and taking advantage of them. And when you transfer that skill set to trading physical commodities you end up messing with tangible or tangible-ish things - beer cans, electric bills - and with markets that have a deeper logic than "we're going to let willing buyers and willing sellers try to scalp each other." If your skills are mainly of the scalping variety, it's probably best to stick to derivatives.

JPMorgan Buys UBS’s OTC Commodity Derivatives Portfolio [Bloomberg]
Morning Six-Pack: Brazil, Canada and the Holy See [NetNet / John Carney]
Goldman and the OIS gold rush: How fortunes were made from a discounting change [Risk, May 2013, paywalled]

1.Fact: if you click that link you probably will hit Risk's paywall. If you, say, Google "Goldman and the OIS gold rush" you probably won't. So.

2.By the way, take note, THIS IS HOW YOU TALK TO THE PRESS:

But how much Goldman itself earned is still shrouded in mystery – and the bank declined to provide any official comment for this article. The bank’s former traders also refuse to provide any numbers, but its rivals put it at around $1 billion. One of the Goldman traders, asked about that figure, only says he “wouldn’t want to understate the profitability of this enterprise”.

Emphasis added, as though it were necessary.


“We did a lot of education on this and presented papers not only to clients but also to regulators. We were offering better prices to unwind the trades than most on the Street and our clients knew why that was the case. This wasn’t something we were trying to hide. Clients could also take advantage of the differential in discounting between different dealers and recognise some decent value,” says the second former Goldman trader.

That is backed up by some swap end-users. “They gave us a presentation and explained the math behind it, so we knew fully what we were going into. And by the same token we were looking for opportunities too – if certain banks, like Goldman, were paying above market prices, then it made sense for us to do the trade with them. Others were significantly mispricing,” says another swaps trader at a different US insurer.