Fortunately, by now the worst excesses of the mortgage market have been obliterated. There are no more CDOs of CDOs, and I truly hope that there never will be again.7 But the financial architecture we live with has become wildly Gothic all the same. Financial derivatives—Warren Buffett’s ‘weapons of mass financial destruction’—are financial structures, typically involving options, that can massively multiply or mitigate asset exposures and/or pricing risks. The value of these products depends on several variables, such as changes in interest rates, currencies, credit, time value, and the like. In contrast to the relatively simple arithmetic needed to perform basic bond math, the valuation of derivatives is a dizzyingly complex and contested subject.
Nevertheless, usage of these potentially beneficial yet complex products has expanded to an extraordinary degree over the past three decades. In ‘over-the-counter’ derivatives (i.e. those that are privately negotiated) alone, there is some $600 trillion worth of notional principal outstanding (a term I’ll define a little later)—that’s over forty times the GDP of the United States.8 Or forget the US. It’s more than ten times the GDP of the entire world. And that’s not all of them. As well as OTC derivatives, there are exchange-traded varieties available too, which add perhaps half as much again to the total stock. As a rough guide, then, the notional principal outstanding on the world’s derivatives market is equal to at least fifteen times global GDP. If there’s intelligent life on Mars, it’s laughing at us.
And these comments assume that the officially collected totals for derivatives are correct. You’d hope they were right, of course: these derivatives have already helped blow up the financial system once, and may well do so again. But because derivatives are contingent liabilities—if you hold them, you may owe something or you may not—they live in the shadows off company balance sheets. Since shadows are a good place to hide stuff, and since plenty of firms have strong incentives to hide their most toxic obligations, you can never be quite sure what lurks in the darkness. Perhaps you think I’m exaggerating? Not so. Lehman Brothers was repeatedly criticized for its ‘Repo 105’ arrangements, which disguised $50 billion of debt under the cloak of phony sales.9 The transactions weren’t illegal but they were deliberately misleading. That’s why they were conducted in the first place. Other firms may be doing similar things now, far away from the sight of the official stats. I don’t know that they are doing so, but I don’t know that they’re not. What I do know is that the regulators lack the teeth and the resources to monitor these things effectively, and that financial institutions—particularly those under the greatest pressure to improve their results—have a strong incentive to indulge in precisely this type of arrangement. That’s scary.
Having said all this, I should probably make a couple of disclaimers at this point. First, I’ve spent much of my career trading derivatives. I think they can be an incredibly helpful tool that can achieve some wonderful things. For example, I’ve pioneered the use of certain derivatives in the carbon-trading market. Those tools have assisted large corporations in the quest to make themselves carbon neutral. When one of their executives flies by jet around the world, those tools will ensure that someone somewhere replaces a coal-fired power plant with wind farms. I’m proud of that. There are plenty of other examples of honorable uses of derivatives. Low-income African farmers can now often use their mobile phones to check agricultural futures prices, and lock in a price today for a crop that won’t be harvested for another three months. That gives the farmer a certainty he never had in the past; and that’s a wonderful thing. Derivatives offer assistance in more mundane ways too. Airlines can protect themselves against a spike in oil prices, or a collapse in profits because of excessive snowfall. Multinational firms can protect themselves against currency swings. Homeowners can fix their mortgage payments. Heck, I once structured and provided the first protection to golf courses in the US for weather-related cancelations. OK, it’s not the same as inventing penicillin, but in a modest way it made those golf courses more financially secure, and thus protected their employees and investors. There’s no reason at all why derivatives shouldn’t be used in ways that help people and firms meet their financial goals. That’s why they exist.
Secondly, I should be clear about the meaning of the phrase ‘notional principal outstanding.’ Many derivatives contracts are based on some notional outstanding amount. So, for example, if you are paying a floating interest rate on some debt, and you’d prefer to pay a fixed interest rate instead, you and I could arrange a deal whereby we enter into a interest rate swap agreement. I’ll pay you a floating rate—for example, three-month Libor. In exchange, you pay me an agreed fixed rate—for example, 4%. In effect, you’ve swapped your floating rate obligation for a fixed rate obligation. That achieves your objective.
Now, if you borrowed $100 million in the first place, and if our interest rate swap was for that full $100 million, then the notional principal outstanding is $100 million. That doesn’t, however, mean that anything like that amount of cash is at stake. It isn’t. In practice, the two sets of payments (my floating rate payment to you; your fixed rate payment to me) are set against each other. So if, for example, the floating rate chosen is 1% below the fixed rate we’ve agreed, you’ll just pay the difference to me, according to our pre-agreed settlement terms. Under these circumstances, the amount of money changing hands in a given year is 1% of $100 million, or just $1 million – a tiny fraction of that scary-looking notional principal outstanding. (The practical reality can be a lot more complex than this illustration suggests, of course.)
Nevertheless, and even taking these things into account, the purely speculative volume of derivatives contracts is extraordinarily huge. What’s worse is that most people who get involved in this stuff don’t understand it. Typically, a Wall Street banker comes along with a slick pitch and a credible story. The rewards seem huge and certain, the risks remote and improbable. The Wall Street banker never quotes you his fee for arranging the deal (because his fee is buried in the overall transaction structure), so while the banker will know to a penny how the contract has been priced and put together, the buyer likely doesn’t have a clue. Plus, of course, the banker is a professional, high-end salesman, with huge incentives to get the deal done. Time after time, I’ve seen clients entering into ridiculous contracts which they don’t understand—and paying an extortionate sum for the privilege. There are plenty of well-known examples of firms and organizations that have been all but destroyed by these contracts.
And buried right at the heart of this huge and dangerous mountain is risk quantification. Wall Street banks have the smartest, best-funded, best-resourced quantitative analysts (‘quants’) in the business, yet when it came to the first credit crisis, they simply didn’t know the value of the stuff they carried on their balance sheets. They didn’t know, that is, until the market ripped their bogus assumptions into a million small pieces, obliterating two leading firms (Bear Stearns and Lehman) and leaving all the other brokerages in fear of their lives—and on government life support. When the ground opened up in front of Citigroup, when the collapse in the mortgage market started to generate huge losses, the firm barely understood what had hit it. Financial Times journalist Gillian Tett, in her book Fool’s Gold, reports one senior Citi manager saying bitterly: ‘Perhaps there were a dozen people in the bank who really understood all this before—I doubt it was more.’10 That Citi manager was wrong. There were no people in the bank who really understood it. By definition. If any of them had understood the true riskiness of these contracts, they’d never have entered into them. This was a huge bank. Taking huge risks. Incurring vast losses. And no one in the bank understood how it had happened.
At least Chuck Prince, Citi’s CEO, had the grace to apologize profusely for his mismanagement of the firm when he appeared in front of the federal Financial Crisis Inquiry Commission. Robert Rubin—chairman of Citigroup, co-CEO of Go
ldman Sachs, former Treasury Secretary—refused to apologize, stating that the executive committee, which he chaired, ‘wasn’t a substantive part of the decision-making process.’11 If it wasn’t, it’s not altogether clear what entitled him to draw more than $100 million in pay and benefits from the firm during his decade at the top. And Prince, for all his remorse, hasn’t felt obliged to return any of his assets. On Planet Ponzi, grace goes only so far. It stops when it reaches the wallet.
And that’s just Wall Street. That’s the smart guys. The problem everywhere else is even more mountainous. If you want to know how a Ponzi scheme can stay alive so long, it’s because you can pile up a hideous amount of losses in other people’s balance sheets—and those people don’t even know it’s there.
Risk quantification. It sounds trivial. It sounds boring. It sounds like a case of just buying a bigger computer. But risk management kills firms. And there’s only one way to avoid it: only play with stuff that you really, truly understand. And stick close to actual market pricing, because only then do you stay close to reality.
Long tail risk
There’s a variant on risk quantification which carries its own separate hazards. Back in the 1990s, JP Morgan—then and now, one of the best-run banks on the market—invented a risk management technology which measured ‘value at risk’ or VAR. That is, it could tell you how much money you would stand to lose on your entire portfolio if interest rates rose a little, or if the yen fell a little, and so on. The technology was a terrific innovation, and became widespread across the market. But it had a limitation. It could only predict likely losses under likely scenarios. It was a way of measuring the losses you’d be exposed to 95 times out of 100, perhaps even 99 times out of 100.
Of course, that’s information any decently managed financial institution needs—an essential part of the information that enables it to manage its ordinary risks as accurately as possible. Ninety-nine times out of a hundred, when you come into work, you won’t find a tornado flying around your dealing room.
But ordinary risks are not the risks which are going to bury your firm—and the whole of Western capitalism—under a mountain of excessive debts and lousy assets. VAR technology was so dangerous because the technology was so good—95% of the time. As it happened, JP Morgan was never suckered by its own creation. The firm remembered what it could do and what it could not do. Management wanted to build a ‘fortress balance sheet’ that would withstand the 1 in 10,000 chance, as well as the 95 in 100 one. So they did. When the first credit crisis hit, JP Morgan shipped some water, but not much. The safety of the firm was never jeopardized. The firm was profitable right through the financial crisis.12
Other institutions weren’t so smart. They believed too much in the power of a technology—and way too little in the power of ordinary common sense, consistently applied. It was another example of too much reliance on computers, too little respect for the way markets can make suckers of us all.
Funding risk
As all these risks multiply, they generate new forms of danger—and dangers that are typically underestimated by those taking the risks.
The flipside of liquidity risk, for example (primarily a concern of investors) is funding risk (a concern of borrowers). This is the risk that the markets you rely on for your funding simply dry up, cease to operate, deny you funds at any reasonable price. The overconfident, overaggressive British bank Northern Rock failed in 2007 because it could no longer obtain funds. Historically, retail banks have always been careful to fund themselves largely by attracting customer deposits, because such deposits have historically proved a reliable and stable form of financing. Northern Rock threw that tried-and-tested model out of the window, relying in very large part on short-term money market financing. That funding structure was insane in at least two respects. First, it generated a huge maturity mismatch between Northern Rock’s long-term assets (primarily loans made to customers) and its very short-term funding. Secondly, it placed vastly excessive reliance on the continued willingness of commercial lenders to lend to it. It was simply asking for trouble—and in 2007 that’s what it got. It took £26 billion of government loans to keep Northern Rock afloat, until the bank was effectively nationalized in 2008.13
That story was only the first in a chain of failures. The German bank IKB set up some off-balance-sheet investment vehicles that were funded exclusively in the commercial paper market (the main form of short-term borrowing). When those markets closed up, IKB was left with nowhere to turn. Dumb financing, maturity mismatches, disastrous outcomes.
Citigroup itself was felled by the same thing. When that manager spoke bitterly about how almost no one in the bank even knew about the thing that had eventually laid it low, the problem was only indirectly the collapse in the mortgage market. That collapse would have bankrupted some of Citi’s off-balance-sheet vehicles, but Citi itself should have survived. It didn’t, because it had guaranteed to provide those vehicles with funding, if normal sources of funding ever dried up. A simple guarantee to offer—but one that brought the company to its knees—and but for government support would have killed it.
One of the really astonishing aspects of these stories is how little those involved seem to have learned. The chairman of Northern Rock, Matt Ridley, was interrogated by British parliamentarians in the wake of the crisis—a crisis that had prompted the first bank run in Britain for 150 years. Instead of sounding apologetic or humble about his failures, Ridley sounded plaintive. He complained: ‘The idea that all markets would close simultaneously was unforeseen by any major authority. We were hit by an unexpected and unpredictable concatenation of events.’14
Well, duh! Of course you don’t expect the unexpected. That’s why you take precautions and prudently manage your risks accordingly. The fact that Britain hadn’t experienced a bank run for 150 years suggests that those precautions were hardly unknown to Ridley’s predecessors. But in the era of the Ponzi scheme, why burden yourself with common sense, when you can just press the pedal to the metal and see how far you can travel before the engine blows? (In Ridley’s case, the answer was around three years at a salary of £300,000 per year.15 Those three years ended up adding around £100 billion to the British national debt.16)
Counterparty risk
We ended the last chapter by noting a puzzling fact. The global financial system boasted core capital of a massive $3,500 billion and yet was laid low by a bankruptcy, Lehman’s, to the tune of just $129 billion, of which a substantial portion—maybe half?—was owed to nonbank institutions. The risk minimization, hedging, and dispersal technologies of which Wall Street was so proud turned out to be risk maximization, multiplication, and concentration technologies. In this chapter so far we’ve addressed a good many of those risks—and of course, they all overlap and interact—but the white-hot core of the problem turned out to be counterparty risk.
That $129 billion which Lehman owed was a net figure: the value of its assets less the (rather larger) value of its debts. But the net figure was irrelevant. That wasn’t what Lehman’s creditors were thinking about. They were thinking: ‘Am I going to get my money back?’ And since there were $768 billion worth of creditors, there were a lot of worried people.
Even that $768 billion figure itself understates things. When Lehman (like any other institution) drew up its balance sheet, it noted down in the right-hand liabilities column only things that were actually owed. But Lehman sat in the middle of a massive network of derivative contracts—contracts that took the form, ‘If X happens, you pay me Y.’ Those obligations are called contingent liabilities, because they’re only triggered if certain events happen. But that was hardly going to reassure all the thousands of creditors who had taken out various forms of derivatives insurance with Lehman. If Lehman had gone, their insurance had gone. Very significant financial assumptions now needed to be reconsidered.
What’s
more, although banks do need to report their contingent liabilities in a footnote to their financial statements (and trust me, that footnote tends to include some scarily huge numbers 17), for management purposes they tend to focus on their net liabilities. So Lehman, for example, would have been in the middle of some interest rate bets which would pay out if interest rates rose and other bets which would pay out if interest rates fell. On a net basis, Lehman would have ensured that its exposure to movements in interest rates would have remained at tolerable levels. But that’s to look at the world from Lehman’s perspective. And on that fateful day in September 2008, Lehman no longer existed. Its creditors—gross, not net; contingent as well as actual—suddenly realized they had no idea if they’d ever get their money back.
The neutron bomb exploded outwards from there. If Lehman couldn’t be trusted, then what about AIG? What about RBS? What about Merrill Lynch? What about Citi? The interbank market froze.18 Everyone was suspicious of everyone else. No one knew how much those other parties had been injured by the succession of calamities (mortgage market problems, Lehman’s failure, interbank funding constraints, massive excess leverage). So the ‘strength’ of Wall Street’s risk management system—the way everyone was now connected to everyone else—turned into its biggest weakness. It was as though all the passengers on the Titanic were roped together, and the fattest guy had just fallen overboard.
Accounting risk
I hope that at least a few of you have read the foregoing pages with a gathering sense of puzzlement. Sure, you might be thinking, counterparty risk matters. You need to know who you’re dealing with and whether they’re good for the money. That’s the kind of risk even Jefferson Smith could have understood. But since the kind of counterparties we’re dealing with are generally large, sophisticated financial institutions, all of which produce detailed accounts on a quarterly basis, why wouldn’t you just call up the latest set of accounts and take a look? If you like what you see, you’ve got a counterparty you can do business with. If you don’t, then just say no. What’s so hard about that?
Planet Ponzi Page 13