Regulators (and New York Times) Discover Bank Use of “Customer” Trades to Place Bets

The very minute the Paul Volcker, who proposed the sound idea that government backstopped banks not engage in proprietary trading, said that trades done on behalf of customers were meant to be excluded from this proposal, anyone familiar with trading could see he’d just deep sixed his idea.

Proprietary trading existed LONG before banks decided to create separate “prop” desks to speculate with house money. And even in the era of prop desks, in the vast majority of cases when a prop trader puts on or exits a trade, who is the end buyer? A customer.

So the measure of whether a bank is making bets in its customer dealing book isn’t the composition of its counterparties, it’s how much in the way of bets it winds up carrying (traders, just like rug merchants, can shade the prices they buy and sell at to keep from accumulating too much inventory). It would have ben possible for regulators to devise rules, such as value at risk limits, requirements that dealers hedge or otherwise “flatten” their positions beyond a certain size level within a specified time period. Even with provisions like that in place, some desks will be exposed when markets turn chaotic, but it should be far fewer, with much less loss exposure.

Instead, as the New York Times recounts, life in big bank land continues more or less as it did before, including large losses when traders make bad bets:

But for all the talk of shutting down trading desks and reassigning employees to prepare for the Volcker Rule, proprietary-style trading will probably survive, if under a different name.

This year, for example, several large insurance companies approached Goldman Sachs, looking to bet that the markets would not stay quiet. Goldman gladly took the other side of the trades, but when the markets turned choppy in May, the firm was caught short and quickly lost $250 million.

Goldman, through a spokesman, declined to comment on its losses on that investment. But in a conference call with analysts last month, the bank’s chief financial officer, David Viniar, explained: “We didn’t hedge it fast enough. Things spiked really dramatically, really fast.”

Months before that trade, though, Goldman Sachs’s research department named a bet against volatility as one its top 10 trading strategies for 2010. Goldman followed its own advice and put its own money in play by failing to adequately hedge the trade with the client who wanted to bet on volatility, which would have given Goldman a neutral position. In this way, a client-oriented trade can effectively become a proprietary bet…..

“Goldman tends to have businesses that have a customer focus with a proprietary overlay,” said one hedge fund manager and Goldman alumnus who insisted on anonymity. “That overlay can effectively allow them to make directional bets by using the customer flow to get them there.”

But Goldman is hardly unique when it comes to walking the fine line between serving clients and taking positions.

Late last year, with clients eager to bet that coal prices would rise, JPMorgan took the other side of the trade and amassed contracts on hundreds of millions of dollars on coal — enough to dominate the European market.

Initially the trade went JPMorgan’s way and yielded profits, but in April the Morgan traders were caught off guard when European coal futures abruptly started rising. In fact, the wrong-sided bets erased all of the previous gains, and by the middle of June, it had turned into one of the commodities unit’s biggest losses — nearly $130 million.

Print Friendly, PDF & Email


  1. Bruce

    The prop desk trading ban idea was always peculiar to me. If implemented in it’s extreme (flat book), it would have removed a significant part of the market, in some assets almost the entire market. Without that liquidity, price swings would be enormously exacerbated, leading to higher costs etc. In addition, how is buying a syndicated loan from XYZ company and sitting on it any different than buying a bond from XYZ company and sitting on it. Exact same economic exposure. More fundamentally, banks exist to be long credit risk. That is their function. So I just never understood what the intention was.

    If regulators want to reduce bank failure risk, it’s very simple – require more and better quality capital, as well as assess capital charges on
    specific asset types more realistically. Of course, as we see with Basel 3, that’s a bit of conundrum for gvts that want banks to also “stimulate the economy”. Safer banks as a policy goal are in diametrical opposition to growing the economy. I think much of the confusion out there on “what to do” can be traced to this conflict.

  2. Tom Crowl

    Please help this non-economist understand something that seems basic to me…

    I ‘get’ the Adam Smith idea! You know, some guy has some extra cash… says, “Hey, I bet our town could use a mill!”

    So he builds a mill… and if things work out the town has a mill, some jobs, some new income and maybe some extra flour to sell elsewhere… and the guy who put up the cash ends up a little richer…

    That all sounds pretty good! And it IS! Very reasonable. It’s often worked out very well.

    And if a different guy in the town has some extra money and decides to build a go-cart track… and loses his shirt… we understand that while that may be sad… in the long-run it will hopefully at least provoke smarter decisions about how to invest… the loss provokes some analysis of why.

    This is also good. Though it does involve some pain.

    Now I do understand that in markets a lot more complex than this example… ‘speculators’ serve a function in providing liquidity, price discovery and can actually add to stability in commodity prices to insure against potentially drastic swings in the price of necessities (and I’d emphasize ‘necessities’ here)… which could distort production of those vitals.

    But returning to my analogy… it seems the current situation is more akin to a town with say a Million Dollar$ looking for ‘investment’…

    And $1,000 is being used to build a mill… another $1,000 is being used to build a go-cart track…

    And the other $998,000 is being used to place bets with each other about any number of scarcely related factors… like which one has taller employees or whether the town will have a flood next year… (in other words zero-sum ‘bets’… NO jobs, no productive result, no growth, no ‘velocity-of-money’… etc.

    Now I don’t know the ratio in our real world…
    But what percentage of ‘investment’ out there is in ‘zero-sum’ BETS?

    What percentage of those ‘bets’ are ultimately attributable to FED ‘created-credit’ that has been passed directly to our ‘casino operators’ who largely now just place bets with each other? And pay each other (and politicians) very, very well for the privilege of observing their cleverness?

    What percentage of a market can be ‘zero-sum’ without destroying the very purposes of the market in the first place?

    This is a serious question? If it’s a stupid one… okay… I admit it… just help me understand this. It seems fundamental.

  3. RichFam

    Nonsense, unless you make dealer work on an agency only basis you will never solve this problem. Unless you’re point is to make all dealers work riskless in which case financial markets will begin to shut down (maybe that’s the point).

  4. Siggy

    What firms like Goldman bring to the market is some degree of liquidity. Firms like Goldman have very few clients, they have customers. Shuting down their prop desk will pretty much cause them to seriously consider going back to being a partnership.

    Now as a market maker, if you bring enough liquidity to the market you become the information focus of the market and that puts you in the position of being able to place counter wagers against customers whose knowledge and vision of the market is one dimensional.

    Now the bank’s use of customer orders to stand in stead of their own position is a form of collusive fraud and should be prosecuted. The current probability of prosecution is extremely low in that we can’t even get around to identifying the inappropriate accounting done by Lehman with its Repo105 bookings.

    The lights are on but nobody’s home.

  5. MichaelC

    This is hilarious:

    “Goldman tends to have businesses that have a customer focus with a proprietary overlay,” said one hedge fund manager and Goldman alumnus who insisted on anonymity. “That overlay can effectively allow them to make directional bets by using the customer flow to get them there.”

    As long as the NYT and the WSJ, etc keep publishing this drivel, as if it actually means something, the bogus illusion that prop trading is not prop trading will endure.

    Did JPM take the loss on coal on behalf of its customers? Did GS take the bad vol position to facilitate customer activity? If they did they should be shut down for being incompetent boobs who don’t know how to cover customer bets. Customer facilitation should earn them a spread, not cause them to face massive losses.

    The NYT article doesn’t ask that question. Rather, they wink and seem to be in awe that those clever guys will get away with it by calling it something else. It’d be nice if the final paragraph said, come on guys who are you kidding? It’d be better if the answer to the question was, Not us.

    I guess that’s your job Yves.

    There’s nothing to do but to keep shouting bullshit every time they try to make the prop trading is just customer facilitation argument. Luckily that’s pretty easy when they fuck up in such nice size.

Comments are closed.