Realcent.org

Big Risk: $1.2 Quadrillion Derivatives Market Dwarfs World GDP
By Peter Cohan
This article was written on 06/09/10

One of the biggest risks to the world's financial health is the $1.2 quadrillion derivatives market. It's complex, it's unregulated, and it ought to be of concern to world leaders that its notional value is 20 times the size of the world economy. But traders rule the roost -- and as much as risk managers and regulators might want to limit that risk, they lack the power or knowledge to do so.

A quadrillion is a big number: 1,000 times a trillion. Yet according to one of the world's leading derivatives experts, Paul Wilmott, who holds a doctorate in applied mathematics from Oxford University (and whose speaking voice sounds eerily like John Lennon's), $1.2 quadrillion is the so-called notional value of the worldwide derivatives market. To put that in perspective, the world's annual gross domestic product is between $50 trillion and $60 trillion.

To understand the concept of "notional value," it's useful to have an example. Let's say you borrow $1 million to buy an apartment and the interest rate on that loan gets reset every six months. Meanwhile, you turn around and rent that apartment out at a monthly fixed rate. If all your expenses including interest are less than the rent, you make money. But if the interest and expenses get bigger than the rent, you lose.

You might be able to hedge this risk of a spike in interest rates by swapping that variable rate of interest for a fixed one. To do that you'd need to find a counterparty who has an asset with a fixed rate of return who believed that interest rates were going to fall and was willing to swap his fixed rate for your variable one.

The actual cash amount of the interest rates swaps might be 1% of the $1 million debt, while that $1 million is the "notional" amount. Applying that same 1% to the $1.2 quadrillion derivatives market would leave a cash amount of the derivatives market of $12 trillion -- far smaller, but still 20% of the world economy.

Getting a Handle on Derivatives Risk

How big is the risk to the world economy from these derivatives? According to Wilmott, it's impossible to know unless you understand the details of the derivatives contracts. But since they're unregulated and likely to remain so, it is hard to gauge the risk.

But Wilmott gives an example of an over-the-counter "customized" derivative that could be very risky indeed, and could also put its practitioners in a position of what he called "moral hazard." Suppose Bank 1 (B1) and Bank 2 (B2) decide to hedge against the risk that Bank 3 (B3) and Bank 4 (B4) might fail to repay their debt to B1 and B2. To guard against that, B1 and B2 might hedge the risk through derivatives.

In so doing, B1 and B2 might buy a credit default swap (CDS) on B3 and B4 debt. The CDS would pay B1 and B2 if B3 and B4 failed to repay their loan. B1 and B2 might also bet on the decline in shares of B3 and B4 through a short sale.

At that point, any action that B1 and B2 might take to boost the odds that B3 and B4 might default would increase the value of their derivatives. That possibility might tempt B1 and B2 to take actions that would boost the odds of failure for B3 and B4. As I wrote back in September 2008 on DailyFinance's sister site, BloggingStocks, this kind of behavior -- in which hedge funds pulled their money out of banks whose stock they were shorting -- may have contributed to the failures of Bear Stearns and Lehman Brothers.

It's also the sort of conduct that makes it extremely difficult to estimate the risk of the derivatives market.

How Positive Feedback Loops Crash Markets

Another kind of market conduct that makes markets volatile is what Wilmott calls positive and negative feedback loops. These relatively bland-sounding terms mask some really scary behavior for investors who are not clued into it. Wilmott argues that a positive feedback loop contributed to the 22.6% crash in the Dow back in October 1987.

In the 1980s, a firm run by some former academics came up with the idea of portfolio insurance.

Their idea was that if investors are worried about their assets losing value, they can buy puts -- the option to sell their investments at pre-determined prices. They can sell everything -- which would be embarrassing if the market then started to rise -- or they could sell a fixed proportion of their portfolio depending on the percentage decline in a particular stock market index.

This latter idea is portfolio insurance. If the Dow, for example, fell 3%; it might suggest that investors should sell 20% of their portfolio. And if the Dow fell 20%, it would indicate that investors should sell 100% of their portfolio.

That positive feedback loop -- in which a stock price decline leads to more selling -- boosts market volatility. Portfolio insurance causes more investors to sell as the market declines by, say 3%, which causes an even deeper plunge in the value of investors' holdings. And that deeper decline leads to more selling. Before you know it, many investors are selling everything.

The portfolio insurance firm started off with $5 billion, but as its reputation spread, it ended up managing $50 billion. In 1987, that was a lot of money. So when that positive feedback loop got going, it took the Dow down 22.6% in a day.

The big problem back then was the absence of a sufficient number of traders using a negative feedback loop strategy. With a negative feedback loop, a trader would sell stocks as they rose and buy them as they declined. With a negative feedback loop strategy, volatility would be far lower.

Unfortunately, data on how much money has been going into negative and positive feedback loop strategies is not available. Therefore, it's hard to know how the positive feedback loops have gained such a hold on the market.

But it is not hard to imagine that if a particular investor made huge amounts of money following a positive feedback loop strategy, other investors would hear about it and copy it. Moreover, the way traders get compensated suggests that it's better for them to take more and more risk to replicate what their peers are doing.

Traders Make More Money By Following the Pack

There is a clear economic incentive for traders to follow what their peers are doing. According to Wilmott, to understand why, it helps to imagine a simplified example of a trading floor. Picture yourself as a new college graduate joining a bank's trading floor with 100 traders. Those 100 traders each trade $10 million: They "win" if a coin toss lands on heads and "lose" if it lands on tails. But now imagine you've come up with a magic coin that has a 75% chance of landing on heads -- you can make a better bet than the other 100 traders with their 50-50 coin.

You might think that the best strategy for you would be to bet your $10 million on that magic coin. But you'd be wrong. According to Wilmott, if the magic coin lands on a head but the other 100 traders flip tails, the bank loses $1 billion while you get a relatively paltry $10 million.

The best possible outcome for you is a 37.5% chance that everyone makes money (the 75% chance of you tossing heads multiplied by the 50% chance of the other traders getting a head). If instead, you use the same coin as everyone else on the floor, the probability of everyone getting a bonus rises to 50%.

When Traders Say 'Jump,' Risk Managers Ask 'How High?'

Traders are a huge source of profit on Wall Street these days and they have an incentive to bet together and to bet big. According to Wilmott, traders get a bonus based on the one-year profits of those on their trading floor. If the trading floor makes big money, all the traders get a big bonus. And if it loses money, they get no bonus -- but at least they don't have to repay their capital providers for the losses.

Given that bonus structure, a trader is always better off risking $1 billion than $1 million. So if the trader, who is the king of the hill at the bank, asks a lowly risk manager to analyze how much risk the trader is taking, that risk manager is on the spot. If the risk manager comes back with a risk level that limits how big a bet the trader can take, the trader will demand that the risk manager recalculate the risk level lower so the trader can take the bigger bet.

Traders also manipulate their bonuses by assuming the existence of trading profits before they are actually realized. This happens when traders get involved with derivatives that will not unwind for 20 years.

Although the profits or losses on that trade have not been realized at the end of the first year, the bank will make an assumption about whether that trade made or lost money each year. Given the power traders wield, they can make the number come out positive so they can receive a hefty bonus -- even though it is too early to tell what the real outcome of the trade will be.

How Trader Incentives Caused the CDO Bubble

Wilmott imagines that this greater incentive to follow the pack is what happened when many traders were piling into collateralized debt obligations. In Wilmott's view, CDO risk managers who had analyzed a future scenario in which housing prices fell and interest rates rose would have concluded that the CDOs would become worthless under that scenario. He imagines that when notified of that possible outcome, CDO traders would have demanded that the risk managers shred that nasty scenario so they could keep trading more CDOs.

Incidentally, the traders who profited by going against the CDO crowd were lone wolves whose compensation did not depend on following the trading floor pack. This reinforces the idea that big bank compensation policies drive dangerous behavior that boosts market volatility.

What You Don't Understand, You Can't Properly Regulate

Wilmott believes that derivatives represent a risk of unknown proportions. But unless there is a change to trader compensation policies -- one which would force traders to put their compensation at risk for the life of the derivative -- then this risk could remain difficult to manage.

Unfortunately, he thinks that regulators aren't in a good position to assess the risks of derivatives because they don't understand them. Wilmott offers training in risk management. While traders and risk managers at banks and hedge funds have taken his course, regulators so far have not.

And if regulators don't understand the risks in derivatives, chances are great that Congress does not understand them either.

Source: http://www.dailyfinance.com/2010/06/09/ ... arket-gdp/

that is one scary illustration.


overexposed? a little I think.

I don't think I ever have really got a grasp on derivatives. I always thought they acted like a commodity does in the market. In my simple world, which is probably way off, the price will either go up or down so someone will make a buck and someone will lose a buck. Should be kind of cash neutral shouldn't it?

Where is the money coming from to fund this? I assumed that it was in everyone's 401k and pension accounts and was being invested. This isn't all borrowed money is it? Now if the executives are taking a cash advance off the top on "expected future earnings" that's a totally different issue but in theory if they were traded properly it was my assumption that they would be cash neutral.

I am going to try to wrap my pea brain around the derative concept and educate myself a little better since it is something I have been interested in for a while but honestly don't have much of a clue about.

The thing about the scary huge number of derivatives is that it's not like someone is going to have to write a check that big, ever. It's like asking what is the retail value of all real estate in the world. Interesting perhaps, but no one is ever going to buy it all, nor is it ever all going to come on the market all at once.

With derivatives, it's not like half the people are on one side and the other half of people are on the other side, like when you bet on a baseball game. It's more like being the bookie, taking bets on both sides of the game. If you find yourself with a lopsided position where if the Cubs lose then you're out $10,000 more than you took in for bets, then you hedge in the opposite direction. You call your bookie friend in Chicago who has a similar but exactly opposite lopsided position where he is out $10k if the Cubs win. You put a big bet for the Cubs to lose, or "lay off" the risk to someone else (he is doing the same, laying off his risk on you). So the Cubs lose, you pay your gambling clients $10k more than you took in on the game, but you have one big bet from the other bookie in which you won, so you use that money to pay back your winners. Your bookie friend is out $10k to you, but he took in that $10k from his local gamblers.

When they get that huge scary number for derivatives, it's not really all that scary when you break it down. Like our bookie scenario. Let's say you took in $30k in bets; $10k payout if the Cubs win and $20k payout if they lose. Your Chicago bookie friend took in $30k also, but $20k payout if the Cubs win and $10k if they lose. To get that big scary number they add up the sum of all the gamblers; the fans gambled $30k in your town, $30k in Chicago, the bookies gambled $60k taking the other side of those bets, plus $10k each between you and your bookie friend. Add that all up, (30 + 30 + 60 + 10 + 10 = $140k). That's called "Headline Math". Police and other law enforcement will use it too when they bust up your "$140,000 interstate gambling racket". Even though you and your bookie friend probably only made $1500 each on that game, taking a 5% vig on the action.

Liquidity problems are the real risk. Let's say your bookie friend is real slow in sending you your winnings. You are slow in paying your winners because you don't have the cash. That causes 5 guys to be late on their child support and alimony. Now all of a sudden you have 20 people all short on cash, late on paying their rent, worrying about their next meal. Bam, you've got a news story, community leaders start clamoring for reform, and everybody gets all worked up. You could see how it would be advantageous for your local mob boss to send in a loan shark and make an emergency loan to you so you could make your payments; they don't want police being pressured to crack down on illegal gambling and other assorted mob business. They want to keep the game going. You are in a real jam so the loan shark gets to extract a high interest rate to keep things working smooth.

This just happened on a federal level, and with a lot more zeroes: TARP.

Thanks for the above example. The concept should work as long as the flow of money isn't disrupted. If it is disrupted it could be from something other than the derivatives that caused it but will still affect it.

Now lets put this into a trading example. Bet your $1 and if you are correct you get back $1.95 and the bookie keeps his 5%. How does this apply too derivatives? If you bet that energy stocks will go up and they do, how much of a cut does the average brokerage outfit typically get? If they are leveraged properly and the cash flow holds out they should just take their commission and roll on not really caring if the market goes up or down. I'm sure it can't be that simple.

Would this be similar in any way to how annuities or the bond market works? If interest stays flat then you get what you get but if interest goes up it cuts into your earnings since the bond price drops and if it goes up enough you could lose money.

When you bet your $1, you are not a "market maker", you are a "market participant". All you can really do is bankrupt yourself, because your broker won't let you bet more than you have, or at least more than your margin allows. If a position turns against you, then you might end up in the poor house but life goes on for the rest of us.

Sometimes a market maker will have a dealing desk; they will take the other side of your position. You want to buy an ounce of gold (or a share of stock or a bushel of wheat, whatever), they will sell it to you even if they don't have it. They will buy it later at hopefully a lower price so they can make even more off your trade. Even if they are wrong, let's say you pay a $5 commission to buy and another $5 to sell. They can be wrong up to $10 on the price when they cover and still make money, because you are paying them that commission. (They might have to pay a few pennies to the exchange for their commission because they do such a huge volume they get a price break.)

Sometimes the brokerage house is a little wrong when they take the other sides of these transactions, and management gets a smaller bonus. Sometimes they are very wrong and the brokerage goes bankrupt and their other clients get hosed as the BK court sifts through the ashes looking for anything they can sell (like MFGlobal). Sometimes all of the Too-Big-Too-Fail guys are spectacularly wrong at the same time, so the fed gov't has to step in keep the entire world from falling apart, so you get what we had a couple years ago.

Commissions on these complex derivative and swap transactions could be a fraction of a percent up to who knows, 50%? More? Whatever those guys can get. It's not like buying a loaf of bread from the store for a dollar, and knowing that the store bought a truckload of bread for 50cents a piece. It's more like trying to figure out your salesman's commission when he sells you a car. The dealer might get a year-end rebate from Ford Corporate if they sell enough cars, you might have bought the special undercarriage treatment so they get a piece of that action, maybe the salesman gets a double commission on every car over 5 he sells in a week, but to make the sale he had to give you floormats and a free wash and wax so he gets a clawback on that, if you financed it then the credit union gives them a kickback, too many blue cars on the lot so the shift manager gives a $100 spliff for moving a blue car by Friday, etc..

I'm explaining this at a real simple level, so I'm glossing over a lot of ugly stuff. Plus I don't work in that part of finance, so I don't have a complete knowledge of how the nuts and bolts fit together over there. But I know enough to say that parts of our system are very broken.