What can we make of this Barclays FERC thing? Besides, like, ha ha ha Barclays you sure like manipulating things? Quick version is: the Federal Energy Regulatory Commission has proposed to fine Barclays $470mm for manipulating California electricity markets by uneconomically buying/selling spot physical electricity in order to manipulate up/down the settlement prices it received on its electricity swaps. Barclays disagrees, etc. You can find the FERC order if you look hard enough1 and I am not an energy-trading guy and I will tell you: I found it entirely incomprehensible! So there’s that.
One quick and obvious thing to make of all this is: remember how one problem with Libor is that, though it underlies zillions of dollars of real economic activity (mostly swaps and stuff, some loans), Libor itself has for years been a nearly purely made-up number, not based on any actual transactions, so you can magically manipulate Libor by just pointing and saying “I MANIPULATE THEE”? That is not true of, God, some sort of electricity price or something:
Physical power at index is physical electricity transacted at a price determined by a methodology that calculates the volume weighted average price (VWAP) of all contributing volumes and prices. The determined or calculated index price, relevant to this investigation, was derived from transactions taking place on ICE in the cash markets. Although physical at index can consist of a number of different time periods (known as “tenors”), the types of physical at index most relevant to this investigation are daily index, balance of the month (BOM) index, and monthly index. Daily index is physical electricity transacted at the VWAP of all next-day transactions for a given day on ICE. The VWAP derived from the transactions on ICE is known as the “ICE daily index price,” “daily index,” or simply the “index. …
That daily index is about as real as can be; it is not only based on transactions but is based on all transactions, weighted by volume, so you can’t manipulate it by pounding a single-point price (e.g. the close) and make your money elsewhere. If you want to manipulate it, your effect is in exact proportion to your actual economic interest. This is a good way to prevent manipulation.
Unless of course you increase your economic interest with derivatives. Barclays’ supposed manipulation was a three-legged thing. Leg 1 would be (say) long swaps, where they had a contract that just paid them money if electricity prices went up over the next month. Leg 2 is that they’d start the month with a short index physical position, basically selling electricity at the average price for the upcoming month. And then Leg 3 is that they’d close out the short index by buying spot electricity each day and pushing the price up. Ignoring the averaging/indexing dynamics, you get long swap + short spot + long spot = just long swap, so they wanted the price to go up. But if they did enough trading they would artificially push up the average price that determined the swap settlement, making money for themselves.2 In the limit case, if Barclays bought all of the electricity each day, they could make the average price arbitrarily high, leaving them roughly indifferent on its “dailies against index” physical trade but making them infinite money on the swaps.
I guess. Here is an amazing exchange between Daniel Brin, a junior Barclays trader, and his friend Crowell at Mirant:
There is much to love here3 but: I share the Mirant guy’s skepticism? Like I said the FERC order is not designed for comprehensibility, but here is a thing that FERC cites approvingly:
The Barclays traders knew their loss-generating physical trading was likely unlawful and specifically ignored the warning of Joseph Gold, Managing Director and Head of Commodities, Americas, who had made clear the practice was unacceptable:
Uneconomic trading activity was something which I tried to make sure was very clear to all the traders. It was something that, during training, I would frequently – that was one of the sessions I was frequently asked to do, the reason being that compliance liked my way of expressing it, which we called the golden rule. The golden rule was always, under no circumstances, lose money on a transaction for the intention of making money on another transaction ….
Umm! Okay! That is … I mean that is obviously a bad rule, no? It cuts out loads of legitimate4 and maybe-quasi-legitimate5 sorts of transactions, but let’s just focus on one, which is: hedging! Here is the FERC order:
Staff finds that Barclays built significant physical index positions in the alleged manipulation months that were in the opposite direction of its financial swap positions.
The horror, but, like, that is what financial institutions do! That is the main thing they do! If you had to describe Barclays in a sentence, you could do worse than “Barclays is a bunch of people who have offsetting derivative and spot positions in various financial quantities.”6 If it had only swap positions, or if it had swap and physical positions that went in the same direction, it would be … UBS! Kind of!
So I’m a little torn. One way to characterize Barclays’ activities was that they were long swap, got short physical, and then closed out that physical short to manipulate prices up, screwing with both the market’s integrity and the poor schmo who was short swap to them. Another way to characterize it is that they were long swap, went short physical to hedge, and then averaged out of the physical short as the swap moved toward expiry. In this view they’re not really screwing with market integrity, and the schmo who was short swap to them was, if hedged, doing the same thing on the opposite side, so prices should balance out.
Obviously from the IMs the Barclays traders sound like they thought they were manipulating the price rather than hedging, which is a big mark against them, though one thing to keep in mind is that talking about market manipulation over recorded IMs makes you obviously an idiot, so we shouldn’t take their point of view too seriously. But doesn’t Mr. Crowell’s skepticism, for all that it’s expressed in a recorded IM in which he plans a day of reckless trading with his company’s money and calls his bosses “retards,”7 have some appeal? If your physical and swap positions roughly offset, then you lose on physical whatever you make on swap. If your swap position is much bigger, then you can leverage small physical losses into large swap gains, which is nice. But this means that someone else has an offsetting big swap position against you, and why wouldn’t they be trading opposite you in the physical market to “hedge” or “manipulate” or whatever the heck is going on here?8
My rough hypothesis on Libor is that Libor worked for so long, despite being mostly a made-up number, because dealers had offsetting swap positions and so offsetting desires to manipulate Libor up and down, so it remained in balance. When dealers all developed a joint interest in manipulating Libor down to make themselves look better, Libor (sloooowly) collapsed. The electricity markets that Barclays supposedly manipulated seem to have had at least two big safeguards that Libor lacked: one, they were based on actual transactions with an economic cost to manipulate, and two, there’s no obvious joint incentive for everyone to manipulate in the same direction. That doesn’t mean that Barclays’ traders couldn’t or didn’t manipulate prices, but I’m not sure that buying in one market, selling in another, and writing stupid IMs about it is compelling proof that they did.
Power market watchdog seeks Barclays fine [Reuters]
Order to show cause and notice of proposed penalty re Barclays Bank PLC et al. [FERC, Word document]
Earlier: RBS Traders Have Yet To Find Anything They Don’t Like To Manipulate
1. FERC: get it together! The FERC order is a Word document that you can find at this link, I think; I could only find it by going to this search page and searching for “Barclay.” I worry for our nuclear power plants.
2. This, from page 11 of the order, is relatively clear:
As a hypothetical example of the type of behavior engaged in by Barclays, assume Barclays established a long peak monthly financial swap position for 1000 MW/h at $52 per MW/h. The financial swap settles against the daily index each day of the month. Barclays then establishes a short 1000 MW/h position in physical monthly index. Barclays proceeds to liquidate its short index position by buying dailies each day of the month. Assume that on the first trading day of the month, Barclays’ cash trades have a VWAP of $60 per MW/h. Other market participants trade at a lower price than Barclays, and the daily index settles at $57 per MW/h. Assume that if Barclays had not traded on this day, the daily index settlement would have been $50 per MW/h.
In this hypothetical, Barclays will lose $3 per MW/h on its cash trading [$57 per MW/h (daily index settlement) – $60 per MW/h (VWAP of Barclays’ dailies)] but will make $7 per MW/h on its financial swap position [$57 per MW/h (daily index settlement) – $50 per MW/h (daily index settlement absent Barclays’ manipulation)]. When compared to the price at which Barclays established its financial swaps, its total profit would be $2 per MW/h [$57 per MW/h (daily index settlement received for financial swaps) – $3 per MW/h (loss in dailies against index trading) – $52 per MW/h (initial price paid for financial swaps)]. If Barclays had let the financial swap settle without trading dailies against index to move the settlement, it would have lost $2 per MW/h [$52 per MW/h (initial price paid for financial swaps) – $50 per MW/h (daily index settlement in the absence of Barclays’ manipulation)]. The total benefit to Barclays would be $4 per MW/h [$2 per MW/h (profit by subtracting settlement price from establishment price and cash trading losses) + $2 per MW/h (losses that it would have incurred absent manipulation)] or $64,000 [$4 per MW/h gain to financial swaps from manipulation × 16 peak hours/day × 1000 MW/h (financial swap position)].
3. FAVORITE PART, not pictured: whenever FERC quotes an IM chat, they include the automated “Notice: All instant messages sent to and from this buddy name will be logged by the IMAuditor and are subject to archival, monitoring, or review and/or disclosure to someone other than the recipient” that one or both chat participants gets on logging in. Which is a nice way of driving home the point that these people are all sort of idiots. Or! That they thought that what they were doing was legit?
4. Reuters calls the manipulation here a “loss leader gambit” but non-gambity loss leaders are pretty okay, no? If investment banks couldn’t give away valuable services free to try to win business down the line … well, analysts would be really happy, for one thing.
5. I remain creepily fascinated by the fact that Carl Icahn’s dealer undercharged him for the NFLX calls he bought but paid him zero for the puts he sold. Lose money on the calls to make money on the puts! That’s probably legitimate-ish.
6. If you added “… and then manipulate those quantities,” no one could really fault you.
8. And a FERC investigation that looks only at Barclays trading activity will see only “the price would have been lower if Barclays hadn’t been buying” but not “but it would have been higher if Barclays’ swap counterparty hadn’t been selling.”