What I Learned from Losing $200 Million

January 5, 2016 7:21 PM

It was 2008, after the Lehman Brothers bankruptcy. Markets were in turmoil. Banks were failing left and right. I worked at a major investment bank, and while I didn’t think the disastrous deal I’d done would cause its collapse, my losses were quickly decimating its commodities profits for the year...

...after a black cab ride from Heathrow to our Canary Wharf office, I got the guys off the trading floor and into a windowless conference room and confessed: I’d tried everything, but the deal was still hemorrhaging cash. Even worse, it was sprouting new and thorny risks outside my area of expertise. In any case, the world was changing so quickly that my area of expertise was fast becoming obsolete.

The 2008 financial crisis taught me about the illusion of control, and how to give it up.

In mid-2008, as crude oil prices were soaring to nearly $150 per barrel, the government of Mexico (the world's sixth largest oil exporter), in an attempt to hedge a fall in oil prices, bought insurance in the form of put options. The options gave Mexico the right (but not the obligation) to sell oil at $70/barrel. If the price were to fall below $70, Mexico could exercise the options to receive a higher price on the oil than what it was selling for in the market. They paid $1.5 billion to hedge 330 million barrels of oil (100% of net exports for 2009).

The oil price fell to around $33/barrel by the end of the year.

The seller of the options (aka the option writer - Bob Henderson, on behalf of an unnamed bank) was effectively guaranteeing that Mexico would receive no less than $70/barrel for the oil no matter how low the price fell. To do so, he takes the other side of the trade.

The problem? There was no market for Maya crude on which the option could be structured, so he had to improvise using a standard derivative technique - he created a proxy for Maya via a "basket" composed of (roughly) West Texas Intermediate (the global benchmark for oil) and fuel oil. As the oil price fell, he needed to keep the option price steady (technical term: delta hedging) so he could pay Mexico if they exercised the option. How did he do it, as the value of the crude oil portion of the basket was plummeting? By selling fuel oil. How much did he need to sell? Enough to move markets.

There were many other decisions and guesses, some made alone, others with help from my team, and still others made by my boss. All were guesswork, none could I have anticipated in stress testing, and all involved abandoning my original strategy along with the illusion of control it gave me.

The Financial Times on Mexico's oil hedge before and after.
Press Release (Mexico): Federal Government 2009 Oil Hedge Results
Futures & Options World Awards
Black-Scholes: The mathematical equation that caused the banks to crash

posted by triggerfinger (29 comments total)

Bob Henderson studied physics, worked on Wall Street, and is now an independent writer focused on science and finance.

I don't know whether to cheer that this cowboy no longer has any direct power or wail that this steady hand is no longer at the tiller of our financial markets.
posted by infinitewindow at 7:56 PM on January 5, 2016 [2 favorites]

I was going to say that his loss is the people of Mexico's gain, and they need the money more.

But then I finished the article and he made it all back and then some within a couple months. The score is 1-1 I guess.
posted by miyabo at 8:07 PM on January 5, 2016 [3 favorites]

I think the place Black-Scholes breaks badly in practice is the assumption of the ability to continuously hedge, where continuous is both in reference to time and price, and specifically that liquidity constraints don't apply. But I've never worked for an options shop, so this is guesswork.
posted by PMdixon at 8:15 PM on January 5, 2016 [2 favorites]

I think the place Black-Scholes breaks badly in practice is the assumption of the ability to continuously hedge


Bla-bla-bla, now a can of soup costs $8.
posted by TheWhiteSkull at 8:22 PM on January 5, 2016 [1 favorite]

Bla-bla-bla, now a can of soup costs $8.

You obviously don't get your soup at the dollar store.
posted by ackptui at 8:37 PM on January 5, 2016 [1 favorite]

Like I wonder about a lot of former bankers, I wonder how much money he saved during his career (and this incident).
posted by limeonaire at 8:37 PM on January 5, 2016

Former options trader here. ...specifically that liquidity constraints don't apply... is fancy-talk for "position size is too big." This guy just took on WAY too much risk relative to the size of the market, and he got burned for it. The position was huge, not properly hedged, and the risk/reward was significantly skewed against him.

He probably justified it by saying that there was no way oil prices would ever move that far that fast, so these options are totally safe to write. I've seen plenty of traders use that same exact justification for all sorts of trades that will "totally never happen." Funny thing is, I've also seen plenty of those traders blow up because those things that are never supposed to happen, well, happen.

The really frustrating part of this is that the compensation structure at banks and trading firms absolutely incentivizes traders to take this sort of risk, and usually the traders get paid very well for a while, until one day it all goes wrong. The guy will lose his job, but typically nothing worse than that, and probably finds a new job soon enough.

On the other hand, those of us who never ever write options like that, (I was always the guy buying those puts, because get it right one time and you're retiring the next week,) typically earned less.
posted by Guernsey Halleck at 8:49 PM on January 5, 2016 [23 favorites]

You obviously don't get your soup at the dollar store.


Yeah, I don't know what they mean by "Creamed Corm," but I'm not going to find out.
posted by TheWhiteSkull at 9:04 PM on January 5, 2016 [6 favorites]

Black-Scholes in and of itself isn't a bad thing. It has lots of limitations and relies on a lot of assumptions (such as normal distributions, no trading costs, dividends or taxes, etc). All of these are known. The problem is that people become WAY too over-reliant on these models and look at the results they give as gospel, rather than as more of an approximation. And I agree with Guernsey Halleck that traders are wrongly incentivized. Combine this with regulatory issues in the system and that's where all the problems come from.
posted by triggerfinger at 9:11 PM on January 5, 2016 [1 favorite]

There's a guy who basically built a career around telling people that distributions have tails. You'd expect that everyone dealing with money would know this, particularly bankers and insurers, but no, apparently not.
posted by Joe in Australia at 9:40 PM on January 5, 2016 [4 favorites]

The problem is that people become WAY too over-reliant on these models and look at the results they give as gospel…

Statisticians, like artists, have the bad habit of falling in love with their models. — George E. P. Box

All models are wrong, but some are useful. — ibid.
posted by esprit de l'escalier at 11:57 PM on January 5, 2016 [5 favorites]

I think it's dangerously misleading when people keep trying to bin the blame for this kind of stuff on Black-Scholes - it's like blaming the Navier-Stokes equations when you fly your plane into a hillside and funnily enough the stuff you're trying to fly through is no longer a fluid. It shifts the blame away from really bad decisions that were made entirely outside of any kind of mathematical framework.

The Ian Stewart article is particularly bad because it gives the impression that options desks blowing up were the cause of the credit crunch, when it was exactly the opposite. The credit crunch was caused by lots of mechanisms causing people to buy mortgages that turned out to be worth much less than they were sold for - options desk blowups were collateral damage (and I say this having sat next to one which exploded in 2009).

If you really want to blame an equation or misguided faith in an equation for the credit crunch, the Gaussian Copula would be a better bet, although I still think it's a distraction from the real issues.
posted by doop at 12:43 AM on January 6, 2016 [5 favorites]

I said earlier that I’d insured a fraction of Mexico’s exports. What I didn’t say was that that fraction was two-thirds

This is a great line. I always think of this when people use that phrase, as in "you can buy this car for a fraction of its sticker price", I mean, what if the fraction is 2/1?
posted by chavenet at 3:53 AM on January 6, 2016 [12 favorites]

The most important point that he implicitly makes is that you can hugely profit if you and your bosses and your financial backers and your central bankers can hold it together through market dislocations.
posted by MattD at 4:26 AM on January 6, 2016 [3 favorites]

I don't know whether to cheer that this cowboy no longer has any direct power or wail that this steady hand is no longer at the tiller of our financial markets.

He provided an honest service to his employer and his employer's clients.

He's out.

The crooks that caused all this are not.
posted by ocschwar at 5:13 AM on January 6, 2016

There's a guy who basically built a career around telling people that distributions have tails.

Hi Boss. You know all that work I've been doing on derivatives for the last 10 years? How you pay me $5 million a year to come up with option pricing models, and your clients pay the company $500 million for this service? Yeah, I can't do it anymore, because it turns out there are these things called black swans. Sorry!
posted by miyabo at 5:22 AM on January 6, 2016 [1 favorite]

I play fantasy baseball. If I want to know the most minute detail about a player's record versus left-handed pitching in the seventh inning of a tie game with two outs and two runners on base, I can find it within seconds.

On the other hand, if any of us want to merely know the identities of the individuals that make epic blunders like this that can take down the financial institutions that are managing our retirement savings, we have to wait until nearly a decade later when they decide that, having made money gambling with other peoples' livelihoods, it's time to cash in on the stories about gambling with other peoples' livelihoods.
posted by tonycpsu at 5:38 AM on January 6, 2016 [9 favorites]

I've seen plenty of traders use that same exact justification for all sorts of trades that will "totally never happen."

When attending a class on capital markets, the instructor said something I will never forget about the mindset you have to go into when making a trade between a buyer and seller. He said, "one of these guys is wrong." That is, the buyer is assuming that the instrument will rise in value, and the seller is assuming it will fall in value. Both cannot be correct! (This is simplified, but bear with me)

This transaction was so huge and SEEMED so unlikely that it seemed like easy profit for the trader who wrote the options. But in this transaction, one of the two parties was going to be wrong. And with such a huge position, what are the odds that a major oil exporter was wrong vs. a single commodities trader? The desire to buy those options on the part of the Mexican government was itself a market indication that should have been accounted for. Who buys an option to sell 2/3rds of their oil at a price 50% of what it is currently trading at within a year? The answer is someone who's pretty sure that oil is going to fall quite a bit in value over the next year. The trader thought he was smarter than the Mexican government and could make easy money off of them. What are the odds that this is true?
posted by deanc at 8:11 AM on January 6, 2016 [2 favorites]

The trader thought he was smarter than the Mexican government and could make easy money off of them. What are the odds that this is true?

I'm not too sophisticated on this topic, but isn't the idea that Mexico made the trade as a form of insurance? That is to say, they let someone else assume the risk - which does not mean both sides lose money. Instead the party assuming the risk spends their time and resources hedging their bets with lots and lots of smaller bets. Mexico probably didn't have the expertise or resources to perform this service themselves.

Since both sides of the trade ultimately did profit, neither side of this trade could said to be "smarter" than the other, they just dealt with the market risk according to their own business models.
posted by elwoodwiles at 8:28 AM on January 6, 2016

Yes, Mexico was hedging. Oil exports are a huge part of their economy and they need to be able to budget based on what they think they'll gain from the sale of their oil on the global markets. Except the price of oil obviously fluctuates so they have no idea what the price will be, say six months from now. So hedging is a very normal way for commodity producers to protect themselves from large fluctuations in the market by locking in a minimum price.

It's my understanding that most oil-exporting nations use futures or swaps for hedging, and that options strategies like this one are unusual. I'm guessing the reason Mexico did it (because it's more expensive to hedge this way) is because the banks involved in structuring the options were offering to structure a brand new product that tracked the actual price of Maya crude, rather than using WTI as a proxy, which is what was standard.

The put option gave Mexico the right to sell the oil at $70, so if the oil price falls to $50, Mexico will exercise their option to sell and the counterparties (banks) then have to pay Mexico the $70 agreed-upon price. However, if the oil price RISES, Mexico doesn't have to exercise the option, they can just let it expire (on the agreed-upon expiration date set at the outset) and they'll only be out the premium paid (essentially the cost paid to the seller, which was $1.5bn). So they're limiting the risk on the downside but can still fully participate in the upside, all for the cost of the premium. If they hadn't hedged, the price shock would have decimated their economy.
posted by triggerfinger at 10:09 AM on January 6, 2016

The put option gave Mexico the right to sell the oil at $70, so if the oil price falls to $50, Mexico will exercise their option to sell and the counterparties (banks) then have to pay Mexico the $70 agreed-upon price.

The bank has to be there on the other side of the transaction to make the purchase, and this turned out to be a market-moving amount of oil to cover. The bank thought its exposure would be minimal and that they wouldn't be left holding billions of dollars of oil that they couldn't unload. The fact that Mexico came to the bank at all should have raised a red flag-- because Mexico had nothing to lose: they could just go back to their previous, somewhat more expensive, hedging strategy (assuming they had one), where the risks would have been distributed throughout the market. The trader, who led a desk of maybe a dozen people, was basically betting that he was more clever than a multi-billion dollar national oil industry. His mistake was looking at the market in isolation: what could happen in the future, based on the past. But his model didn't account for one piece of important information in the present: the fact that Mexico wants an option to sell 2/3rds of their oil at $70/barrel sometime over the course of the next year.

When someone specifically asks for a life insurance policy with a Double Indemnity clause for train deaths, maybe you should make sure no one is buying train tickets for the covered person.

Since both sides of the trade ultimately did profit,

Only because, ultimately, the bank probably wouldn't have been able to fulfill their part of the deal, and Mexico figured they were better off preserving the relationship with the guy who wrote the options rather than sucking him and the commodities desk for all they were worth.
posted by deanc at 10:45 AM on January 6, 2016

It appears that in 2015, the hedging strategy paid off for Mexico again. Last month, Mexico's Finance Ministry received a record $6.3 billion from its oil hedge program. This is more than the $5.1 billion it received in 2009.

What's interesting to me is that Mexico's hedging program is unusual. According to this Bloomberg article: "Few other commodity-rich countries have followed suit with similar hedging programs. Ecuador locked in oil sales in 1993 and, after losses triggered a political storm, never tried it again. Colombia, Algeria and even Texas have experimented with locking in prices."
posted by cynical pinnacle at 12:34 PM on January 6, 2016

Note that Mexico is one of the few oil exporting countries that have not suffered serious domestic unrest in the wake of oil prices falling.

And that a few weeks ago they had a visit from a cat-5 hurricane without a single life lost.

They're doing something right.
posted by ocschwar at 12:44 PM on January 6, 2016 [2 favorites]

Yeah, I always assume governments don't stand a chance when negotiating with ultra-sophisticated financiers (how can they afford to hire anyone with those skills?), but it does seem Mexico knows what it's doing.
posted by miyabo at 7:02 PM on January 6, 2016

I'd be interested in hearing what somebody who'd lost $200 million had to say, but this person didn't lose anything personally. As far as I can tell this story is about how their ambition allowed them to mismanage other people's money at an eventual cost of $200 million. It concludes with him receiving his bonus at the end of the year anyway.

"Golly what a year it was and also look at all these moneys" is not much of a moral insight, in my opinion.
posted by mhoye at 7:38 PM on January 6, 2016 [1 favorite]

deanc: "But his model didn't account for one piece of important information in the present: the fact that Mexico wants an option to sell 2/3rds of their oil at $70/barrel sometime over the course of the next year.

When someone specifically asks for a life insurance policy with a Double Indemnity clause for train deaths, maybe you should make sure no one is buying train tickets for the covered person."

Also, from Guys and Dolls (1955):

One of these days in your travels, a guy is going to show you a brand-new deck of cards on which the seal is not yet broken. Then this guy is going to offer to bet you that he can make the jack of spades jump out of this brand-new deck of cards and squirt cider in your ear. But, son, do not accept this bet, because as sure as you stand there, you're going to wind up with an ear full of cider.

posted by mhum at 3:07 PM on January 7, 2016 [1 favorite]

Back when I worked for a hedge fund, one of the most important lessons I learned was that in finance, the expression for being "over a barrel," or being in a tight spot, or just being fucked, is being "short an option."

Everyone in finance has learned Nassim Taleb's lesson that options are something you should buy, not issue. (If you issue and sell an options contract, by definition you are short that option.)

But if you do decide to go into that shark tank and issue sell options at $70, then you damn well better buy options at (say) $40 so that your potential losses are limited to $30. (And they guy who sold you that options contract at $40? He better buy some at $20, if that's how much he can afford to lose. And so on up until that guy who sells an options contract at $5 and is good for that $5.)

And you should buy those options before you execute that sell. Of course, if your options contract doesn't actually exist, and you have to create something that will perform like that options contract, for your customer, than you might have a hard time finding one that will sell you that same contract you just cobbled together, at a different price. But if you can't afford to lose all the way down to zero, and you can't find your own hedging counterpart, then you should not be signing your employer to that deal. SO I see a big operational cockup here. Still, no wrongdoing. This was an investment bank. He lost other people's money, but those other people had an affirmative duty to be able to withstand losing every money they put into that pot.
posted by ocschwar at 7:52 PM on January 7, 2016 [5 favorites]

Part of the cockup was a failure to realise that issuing options on such a huge part of Mexico's production meant that he was clearly signalling his position. But, Mexico also made a huge error: ignoring the fact that a single trader with an unusual option mix couldn't necessarily cover the options. Things worked out OK for both of them, but it was good luck more than good management.
posted by Joe in Australia at 10:12 PM on January 7, 2016 [2 favorites]

Can we give these people their own economy to play with, so that they can leave the one we need to buy food and pay rent alone?
posted by Legomancer at 11:23 AM on January 8, 2016