A Little Perspective, Please - Wall Street and Trading Risk

As some of you likely know, there has been a bit of a rhubarb over Credit Suisse's reported $120 million derivatives trading loss in South Korea (supposedly on reverse convertible notes - ooooh, how scary!). This has been discussed in both MSM (mainstream media, for those who don't know my use of the acronym) and the blogosphere. For an overview, let's take a look at Monday's story in the Financial Times:

Brady Dougan has been talking about the need for "disciplined risk-taking" since he took charge of Credit Suisse's investment banking unit in 2004. But perhaps only the second half of the message is getting through.

The bank's equities desk is reported to have lost about $120m in the third quarter from issuing reverse convertible notes (RCNs) on Korean stocks. It remains to be seen how this fits into the overall trading result, but it is a big number all the same.

RCNs are popular among Asian investors because they pay a generous coupon like a high-yield bond. The quid pro quo is that the issuer can force conversion into a stock if that stock falls by more than a certain amount over a designated period. This gives the issuer an effective put option on the shares in question. Credit Suisse's losses were apparently made in so-called "worst-of" RCNs, a particularly risky breed. These convert into the equity of the worst performing out of several stocks. Hedging these is hard but, for a sophisticated derivatives desk, not impossible.

The most likely possibility is that falling volatility reduced the mark-to-market value of those embedded puts. The best that might be said for Credit Suisse is that it chose not to hedge its exposure fully in this case - aiming to maximise its profit on the trades assuming the underlying stocks performed to expectations. The snag is that this goes against Mr Dougan's strategy of disciplined risk-taking. Some rivals say Credit Suisse should have responded to early warning signs of slackening volatility.

These are problems that are capable of being solved by stronger management. As it happens, Mr Dougan has just overhauled Credit Suisse's top equities brass after repeated wobbles in the equity trading unit. Having restored Credit Suisse's appetite for risk, it may be time for more discipline.

I was asked to comment on this story for National Public Radio on their daily show Marketplace. The text of the story is below. As you can see from the questions and the comments of the Morningstar analyst who spoke before me, it's plain to see what the interviewers are pulling for:

KAI RYSSDAL: Let's take a second to do a little financial etymology here. Derivatives is the word of the day. Financial instruments whose value is derived from something else. In the most basic sense they're bets. That oil, for example, will go up because we're going to have a cold winter. Or down 'cause it'll be warm. Either way, derivatives are risky. An easy way to lose money. There's word today Switzerland's second largest bank — Credit Suisse — lost a lot of it on a bad bet in South Korea. Marketplace's Amy Scott has more.

AMY SCOTT: The situation in North Korea has captured the attention of options traders, who have taken advantage of volatility in the South Korean stock market. According to Bloomberg News, Credit Suisse adopted an aggressive strategy to cash in on that volatility. When the market calmed down, the report says, the bank ultimately lost $120 million. Morningstar analyst Ganesh Rathnam says Credit Suisse was late to the game its rivals were playing.

GANESH RATHNAM: The big lesson is don't chase something just because everybody else is making money on it. If you chase a hot investment, more likely than not the later players are gonna get burned.

Credit Suisse may have been under pressure to boost profits in its equity trading division, which haven't kept pace with competitors. Analysts say that may have driven Credit Suisse traders to take bigger risks.

But Roger Ehrenberg with research firm Monitor 110 says this is what investment banks do. Sometimes risky bets pay off. Sometimes they don't. ROGER EHRENBERG: The person sitting on the trading desk that incurred that loss is certainly not feeling so great about life. But by the same token, is that within the range of expected outcomes? Sure. A Credit Suisse spokesperson wouldn't confirm the losses, and said the company won't comment until it reports third-quarter earnings in a few weeks.

Analysts we talked to don't expect banks to disclose specific losses like those in South Korea. But if the word does get out, one said he'd rather hear it from Credit Suisse than Bloomberg.

In New York, I'm Amy Scott for Marketplace.

Sure, makes for some nice headlines, but what does this mean? After reading a bunch of commentary on this loss, I am thoroughly convinced that virtually nobody writing about this story has any idea what the hell they are talking about. In fact, it infuriates me when people take up column inches writing stuff that has no basis in reality (but looks great in black and white). Do you know how many bulge bracket Wall Street trading businesses have suffered a $120 million loss on a particular desk in a quarter (and we don't even know this to be the case - the loss in question is on a particular strategy)? I tell you how many - EVERY SINGLE ONE. Why? Ok, here is the answer. I want to make sure you are listening. Ok? Here it goes.

Trading desks try to make money. How? By taking risk. What does this mean? I will sometimes win, I will sometimes lose. Ok, but how much? Well, let's say I am running an array of positions across a $25 billion balance sheet (but in fact the bulge bracket broker/dealers are much, much larger). I have a risk budget. Maybe it's derived by Value at Risk, maybe by Monte Carlo simulation. Whatever. Bottom line - I have an expected loss factor for a day, a week, a month. These loss factors take into account rising correlations across asset classes, exogenous shocks that cause 20%, 25% or even greater losses over short periods of time. When I am allocating capital across my trading desks I am conscious of the cross-correlations on my trading platform, and using my skill to tilt bets based upon my expertise and those of my traders. Given all this, trading businesses over the past three years have done extremely well. However, within that time frame, say, the second quarter of 2004, trading desks got absolutely murdered. After a record P&L in Q1, most desks got carried out in Q2. I know people who literally dropped $200 large - that is $200 million - in that quarter. But you know what - they still ended up a $1 billion+ on the year. And these outcomes are well within the expected probabilities based upon stress tests. Were people unhappy during Q2 2004? Sure. But those with cooler heads went on a tear and had a great 2004.

So why am I saying this? Because while I don't know Brady Dougan personally (though know plenty of people at CS), I think he and CS are getting a bad rap. A $120 million loss on a trading desk, unless the loss was the result of poor controls or a rogue trader, is neither a show stopper nor something that warrants intense eyebrow-raising and upset stomachs. THIS HAPPENS TO EVERY FIRM. Even the almighty Goldman Sachs drops a few hundred million (give or take a few hundred million) now and again. So people, please, if you are going to write about trading please get a clue. You're wasting all of our time. And get off Brady's back. He's doing ok.

 

Comments

[Roger]

Yaser, it's not an issue of luck, it's an issue of portfolio management. If a desk isn't taking a $120 million hit now and again, and has billions of excess capital as most of these firms do, then they aren't taking enough risk. The issue of luck in the Taleb-Fooled by Randomness sense is completely different.

The issue here portfolio construction, diversification, performance measurement over an appropriate time period and risk thresholds. Again, what we saw here with CS is an probabilistically expected outcome in the ordinary course. We all need to take a step back, take a deep breath and remember the business they are in.

[Yaser Anwar]

John W: You're right when you say the size of the bets spooks MSM, especially since derivatives are known to be Weapons of Mass Financial Destruction.

The nagging doubt you have- from my perspective: I think ever since the LTCM catastrophe and the increase in demand of automated quant strategies, banks and hedge funds who employ these tactics have to ensure it is in their circle of competence.

How do they do that? They employ Ph.Ds and math whizzes [read: James Simmons]. I forgot the name of the recent company in oxford, london NYT highlighted which employed Physics and Math whizzes to do all the modeling.

So you see it is imperative that they understand the pricing models and the potential catastrophes. Will all of them be able to do that with just hiring Ph.Ds and whiz kids? Probably not, but its a step in the right direction.

[Yaser Anwar]

RE:

If CSFP knew the risks they were taking, had good risk management tools at their disposal & their legacy excellence- does it all boil down to luck then?

Since they have every other tool from experience to tools...

If all else is equal and it comes to luck- is there a way to ensure you come out on top? or you just have to roll the dice and let the chips fall where they may..

[Roger]

John, as always, insightful comments. Concerning your question, I'll tell you that the risk management departments of the bulge bracket firms have seriously upped their games, investing millions in enhanced infrastructures and portfolio management systems. It is a different world than it was in the late 1990's, where everything was VaR, normal distributions assumed and correlation matrices woefully inadequate to take into account the interrelationship among markets and asset classes.

Concerning CS specifically, given the legacy excellence of CSFP in both trading and risk management I am highly confident that they knew exactly what they were doing and were acutely aware of their exposures. Did they make some bad trades, sure. But that is not the issue. Did they know the risks they were taking? Based upon my knowledge of bulge bracket risk management systems and their trading skills I am confident the answer is yes.

[John Wilson]

Roger, as you say the game is risk taking and you if never lose, you're not taking enough risk to make serious money.

Likewise, as you say, only reporting the losers in a portfolio is daft - there will usually be offsetting winners.

I guess what spooks the MSM journos and their readership is the size of the bets that can be placed by the desks - what other industry allows such big bets outside of the main board members. To those of us who've been around capital markets for a long time, we shrug at the numbers involved because we see them daily. Joe Public doesn't and hence is awed.

In the same way, IB compensation is always headline news because it's so out of kilter with main street. No matter that talentless celebs/bands make millions.

However, the nagging doubt for me is how well the banks understand the risks and pricing models. The pace at which new dervis are churned out usually outstrips the abilities of the systems and the risk personnel. Many rely on high complex maths models - can many senior execs truely claim to understand the models, the assumptions and the potential outcomes?

[Roger]

Yaser, yes, I believe this to be a bunch of garbage and that CS got caught because their loss appeared to be sustained in a small-ish market where people are aware of the players. Crappy luck. And yes, MSM (and the copycat part of the blogosphere) simply jump on the bandwagon and waggle their fingers and say "bad boy." This is such hypocritical junk.

Prop trading desks and hedge funds are cyclical businesses like construction, automotive and high-end retailers. There are periods where they rake it in, periods where they make a little and periods where they lose. How well these businesses are operated can't be viewed through the prism of a quarter, but over longer time frames. If CS loses $120 million as part of their overall strategy but ends up $1 billion for the year, are people still squawking?

If one accepts that such variations are within the ordinary course of business (which, I assure you, they are), then people should better be focusing on overall management of the business and NOT a short-term periodic loss. If Warren Buffett were viewed in this light he would be vilified on a regular basis. Solomon Brothers? US Air? The precipitous decline in the value of Coke? His short position against the dollar? Come on, people, WAKE UP. Don't hold prop traders and hedge funds up to different standards than other business and other types of investors. It is simply not fair and absolutely wrong.

[Yaser Anwar]

RE, since you know people who've dropped 200 mill in a quarter, would it be fair to assume that CS was just the unfortunate one to have the drain in the P&L leaked to Bloomberg?

Would it be fair to say that even though CS is not a hedge fund, any news that is in the "loss of 100mil cohort" will generate intense scrutiny due to the misfortunes of the HF industry of late?

It seems the MSM won't neglect any chance to throw stones at the industry. What say you?