Planet Ponzi

by Mitch Feierstein

  After that, the theory stops and the grim accounting begins. We’re only halfway through the credit crisis and the worst is yet to come.

  10

  How to hide a neutron bomb in ten easy steps

  The previous chapter talked about the evolution of financial markets from the simplicity of Joe Schmoe’s mortgage to the horrendous complexity of Joe and Joella’s. We saw how the Joe and Joella credit risk was ground up so fine that no single investor held more than a passing exposure to that fine couple. We saw how new ways to insure yourself against risk were created and deployed. We also saw that despite the proliferation of these ‘risk minimization’ technologies, when it came to the crunch they turned out to be risk multiplication technologies. Those financial neutron bombs are still scattered thickly across our financial system today. As a matter of fact, those bombs are so numerous and so dangerous, I can’t even list them all. What follows is something like a top ten‌—‌or, in fact, a recipe for how to create a financial neutron bomb. It’s a recipe that Wall Street has gleefully followed, with its normal high-energy zeal and creativity. Oh, and bear in mind that many of the most dangerous explosives sit in the grey land ‘off balance sheet’‌—‌that is, not directly recorded in company accounts‌—‌which makes it almost impossible for third parties such as myself to quantify, or even fully describe, the risks involved. Needless to say, when those explosives detonate they have a curious habit of becoming ‘on balance sheet’ realities very fast and very destructively.

  But we start with one risk‌—‌credit risk‌—‌that doesn’t form part of that top ten, simply because credit risk is inherent in any financial system. Indeed, it’s the point of the system; the reason why banks exist in the first place. So I don’t have a problem with credit risk; it’s what you do with it that counts.

  And there was and is and always will be just one way to handle credit risk properly: by sitting down with all the data (and ideally eyeball to eyeball with the borrower) and figuring out whether the risk makes sense. The Joe Schmoe / Jefferson Smith banking system didn’t eliminate credit risk; it just handled it properly. Called for evidence. Considered income flows. Weighed up the value of the collateral. Made an assessment of character. Produced a decision.

  The slice’n’dice approach to credit risk‌—‌the way Wall Street operates today‌—‌means that everyone ends up exposed and no one ends up doing the analysis. In the mortgage market, no one ever really wondered whether Joe and Joella were good for the money they borrowed. The inevitable consequence was that some awful credit decisions were made.

  That same lack of fundamental analysis is still true across huge acreages of the financial system today. When investors bought Greek sovereign debt, did they really do the math? Question the data? Consider economic fundamentals? Evaluate the politics? Figure out how a not very large, rich, or innovative European country could afford to run a railway system that racked up losses of more than $1 billion on sales of $253 million?1 Or allow hairdressers to retire at fifty because they have a ‘dangerous’ occupation?2 Or decide to back a system that mis-states its accounts, has gaping holes in pension provision, and routinely raids social security coffers for current spending?

  Of course they didn’t. No one could have made a sober analysis of the risks and decided to go ahead and invest. People who bought Greek debt did so because they didn’t do their homework‌—‌and now they’re reaping the rewards of their own idleness.

  You can’t eliminate credit risk from finance, and you shouldn’t even try. What you can do is never take on a risk without thinking about it. That’s not a fun game; it’s not sexy, it’s not innovative, and most of the time the result of all your hard work will be to confirm exactly what you guessed in the first place. Well, tough. If sexy is what you want, go into the movie business. If good credit decisions are what you want, you need to put in the hours.

  Because credit risk lies at the heart of everything, many of the risks that follow trace back in some way to this one central issue. Nevertheless, as we’ll see, the multiplying variety of risks means a massive increase in total risk.

  Groupthink and reliance on authority

  One of the major contributory factors in the mortgage crisis was the reliance of end investors on the credit ratings awarded by the major ratings agencies (Moody’s, S&P, Fitch). Investors thought they didn’t need to perform their own credit analysis, because someone else was doing it for them.

  Sure enough, those ratings agencies called the mortgage market spectacularly wrong. Securities which were rated AAA turned out to be junk. Securities which carried a sub-investment-grade rating oftentimes turned out to be worthless. Now, to some extent, that’s just human error. If smart people look at enough problems, there’ll come a time when they call one wrong. The agencies do, I believe, do a decent job in trying to get things right, but in the end they’re staffed by humans‌—‌and humans who, as we’ll see in the next chapter, face an uphill struggle in making the right calls.

  In particular, they tend to be slow in adjusting their opinions to a swift-changing reality. Their failures in the mortgage market are very well known, but you can take pretty much any credit crisis in history and find that the ratings agencies have been slow to catch up with reality. At the time of writing, Fitch has just downgraded Greek debt to CCC‌—‌that is, ‘extremely speculative’ and only one notch up from ‘in default.’ Fitch is damn right. And damn slow. Greek debt has been extremely speculative for a long while. Its debt to GDP ratio has been hovering at or above 100% for more than ten years. It has a history of falsifying its national accounts. Deficits have been high. Corruption is rife, tax avoidance ridiculous, politics dysfunctional, and the economy remarkably lacking in strong, competitive, export-oriented companies.

  These longstanding features of Greek finances more or less define the notion of speculative sovereign borrower. What’s more, in the years prior to the creation of the euro, markets knew perfectly well that Greek debt was speculative in the extreme. In 1994, the cost of borrowing money in drachmas for six months was over 25%. In the following years, it stayed in the mid-teens.3 These huge rates tell us that investors knew perfectly well the Greek government could not be trusted to repay its loans without destroying value. Back then, the destruction happened via devaluation. Once Greece entered the eurozone, however, default was left as the only plausible option.

  These things are obvious. However little you know about financial markets, you know that someone who can only borrow money at interest rates of between 16% and 30% is scarcely a trustworthy borrower. Yet, as you can see from figure 10.1, even after the credit crisis started to break in 2007–8, every single one of the major ratings agencies rated Greece at least A−, or ‘upper medium grade.’ How could that possibly be? The Greeks lied about their national accounts‌—‌deliberately falsified them‌—‌and on the kind of scale which would have brought about criminal prosecutions in any private sector company.4 To me, no such borrower can be judged investment grade (that is, BBB or higher), at least until some years after the problem has been thoroughly exposed and addressed. Given that a country’s credit standing rests on its capacity to levy taxes, I’d also suggest that a culture of outrageous tax avoidance‌—‌promoted in part by the authorities’ blind acceptance of it‌—‌would tend to suggest that the country’s actual revenue-generating capacity falls far short of its theoretical capacity for gathering taxes.

  Figure 10.1: Greek debt rating history, 1999–2010

  Source: ‘Greek rating cut, but bail-out rate agreed,’ Financial Times, April 9, 2010.

  Once again, the conclusion is remarkably simple. Investors need to do their own homework. They need to avoid groupthink by practicing the habit of independent thought, evaluating risk for themselves. However hard and honestly the ratings agencies attempt to conduct their business‌—‌despite a biased and inappropriate set of incentives‌—‌they can’t always be right. Very often, they’ll prefer to opt for comfort rather
than truth. They were wrong about the mortgage market. They are only now starting to catch up with problems in the sovereign debt market. The only protection for any investor is to start from the beginning every time, to do the homework, to think for themselves.

  Liquidity risk

  In the old days‌—‌the days of Joe Schmoe and Jefferson Smith‌—‌credit was, for the most part, not tradable. A loan was made, and the same two counterparties would then remain locked in a credit relationship which would last until the loan was repaid or the borrower declared bankruptcy. Since both parties knew they’d be tied together for the term of the loan, this arrangement forced discipline on both of them. Borrowers went to banks they trusted. Banks chose borrowers they deemed reliable. Simple good sense.

  The development of a large and liquid corporate bond market in the late 1970s and 1980s brought many good things in its wake. Because borrowers could go straight to the ultimate investors, without needing an old-fashioned commercial bank to act as middleman, they could borrow more cheaply. That helped them make investments and develop their businesses. In a small way, we all benefitted.

  Investors benefitted too. Previously, investors had, broadly speaking, three choices. They could buy equities on the stock market. They could acquire bonds issued by the government or a few very large banks and corporations. Or they could put their money on deposit in a bank. As the securities market opened up, they became able to build entire portfolios of corporate bonds. They no longer had to restrict themselves to blue-chip borrowers, the likes of General Electric and IBM. The creation of a market in sub-investment-grade debt‌—‌the junk bond market‌—‌meant all kinds of borrowers could now access investors directly, and vice versa.

  As these markets developed, investors began to realize that this new form of credit relationship didn’t have to be a lasting one. You could buy IBM bonds one day, wake up the next day and decide you wanted to invest in something else. So you sell IBM and buy whatever. A couple of phone calls and you’re done. Investors no longer had to be the boring guys in finance. They could be traders too. They could be players.

  And the very best thing, from an investor’s perspective, about this new world? Risks were slashed. If you bought an AAA-rated bond issued by the well-known company American Widget Corp, you’d be aware that even a strong company carries some risk. If AWC ran into bad times‌—‌if competitors attacked its markets, prices tumbled, technology changed in adverse ways‌—‌then likely enough the ratings agencies would start to downgrade the bond. From AAA to AA+. From AA+ to AA, then maybe AA−. And at every stage, you’d have the option to bail out. If your investment fund wanted to concentrate only on the very strongest credits, you could simply have a rule that you’d sell up if a bond was downgraded below AA−.

  Sure, you’d most likely take a hit as you made that sale. The price of a bond is inversely related to its yield, or interest rate. (That’s a mathematical truism. If you don’t understand why the rule exists, just trust me on it.5) Since less creditworthy borrowers have to pay higher interest rates than stronger borrowers, when a bond is downgraded, its price will fall. But you don’t mind taking a hit. Losing a little money when you sell out of a bond is the price you pay to protect yourself against further weakness. In the old days, if a bank made a ten-year loan, it knew it was going to be on the hook for ten long and dangerous years. In the new world, if an investor bought a ten-year bond‌—‌or even (why not?) a thirty-year bond‌—‌they knew they weren’t making a commitment that had to last as long as the bond. They could sell the bond whenever they wanted. Pass the risk on to someone else. This was the joy of the bond market‌—‌the joy of liquidity.

  Don’t get me wrong. Liquid bond markets have indeed brought huge benefits. They reduce the cost of finance. They make it easier for ordinary companies to do what they do. We all benefit. But you don’t get innovations without risk; and in this case, the existence of liquid bond markets tempted banks and investors to believe that those markets would always be available. And virtually no market is always available. Why would it be? Markets exist when you have plenty of willing buyers on one side, plenty of willing sellers on the other. If circumstances alter, so that there are many more would-be sellers than would-be buyers, you no longer have a market. You have Torschlusspanik‌—‌a German word meaning literally ‘door-closing-panic’: a stampede for the exit, under conditions when the stampede is the very thing that’s blocking up the exit.

  Circumstances like this are anything but theoretical. When Lehman Brothers was starting to go under, it tried to exit the mortgage market positions that were causing it so much trouble. But as Lehman started selling assets, the market learned that it was selling. Because Lehman was such a significant player, that knowledge inevitably forced the price lower. Other investors figured they’d better sell before the price crashed completely. So the price fell lower. So Lehman was under still greater pressure to sell. So prices crashed until the market effectively locked up. These were firesale prices, only no one was buying.

  Huge numbers of investors were caught in the crossfire. They’d bought shedloads of AAA-rated mortgage bonds containing subprime and other mortgages‌—‌in the belief that (1) the credit agencies couldn’t be that mistaken, and (2) if a problem arose they could always sell up and get out, suffering only a modest loss. Wrong! And wrong again! The consequences of those false beliefs were massive. They lay at the very heart of the first stage of the credit crisis, and inflicted losses of (depending on how you count them and whom you believe) between $1.5 and $3 trillion.

  You can see the first shivers of Torschlusspanik in the sovereign debt market now. In mid-September 2011, the cost to Greece of borrowing money for two years rose to over 80% per annum, and market rumor placed some yields in excess of 100%. I know loan sharks who would be embarrassed to charge that much. Indeed, Greece could get a better rate than that by borrowing the cash on some ripoff store card deal. In effect, that 80% interest rate is simply a sign that funding is no longer available, that the country’s debt market has lost all liquidity.6 (As a rule of thumb, when interest rates rise to more than 7% or 8% payback becomes impossible, because you can never manage to pay back principal when you’re spending so much on interest payments.) If you were one of the investors who figured that you didn’t have to think about credit risk because you could always sell your bonds and move on, you’d have found that the modest hit you had been prepared for had turned suddenly into a knockout blow.

  You’d think that maybe these experiences would have taught investors to think differently about the risks they take on, but nothing seems to teach investors anything. The incredible speed with which Wall Street returned to business as normal after the first credit crisis has seen credit and leverage returning to pre-Lehman levels. Each new problem is seen as a local matter. Oops! Bit of a screw-up in the mortgage market. Yikes! Got my fingers burned in Greece. But none of that is going to be a problem for my Spanish bonds (or my Italian bonds, or the bonds issued by banks who hold this stuff), because I can always sell out if things get sticky.

  Risk quantification

  So far, the risks we’ve spoken about would have been recognizable to the bankers of the Joe Schmoe / Jefferson Smith era. Back then folks didn’t rely too much on ratings agencies and liquidity risk was much less prominent. Jefferson Smith would have been baffled by the idea of risk quantification. As far as he was concerned, the amount of money he had at risk was the amount he’d lent to Joe Schmoe minus the amount that Joe had repaid. The only risk to which Jefferson Smith was exposed was that some idiot in the office might not be able to do simple arithmetic.

  The situation is utterly different today. For all that securities are designed to be tradable‌—‌a commoditized asset class using largely standardized legal language, conferring the same rights no matter which investor owns the security‌—‌they oftentimes don’t trade much at all. Sometimes that doesn’t matter too much. If you own an A-rated corporate bond, and you’
ve done your homework and reckon that the A-rating is appropriate for the borrower in question, you can get a pretty good idea of the price your bond will trade at simply by looking at the prices at which other A-rated bonds are traded today. There’ll be a margin of error, of course. Investors have different sentiments about different companies. There’ll be things to think about in terms of liquidity, duration, currency, and so on. Nevertheless, if you’re reasonably smart about these things, you should be able to do the math with a fair degree of confidence.

  Those comments, however, apply only to bonds that come in plain vanilla, and it’s increasingly rare for any large financial institution to stick to plain vanilla. In the heyday of the mortgage market, many of the most complex securities hardly traded at all, and when they did the transaction prices weren’t disclosed. (A major difference from the equity market, by the way. You can go online right now and check the buy-price and the sell-price on any major stock in the world. That publicly available information just doesn’t exist for most parts of the bond market‌—‌a big reason why Wall Street finds it easy to rip off customers and investors.) Yet banks and others needed some way to value their portfolios‌—‌so instead of relying on actual prices paid in actual transactions for actual securities, they built mathematical models of how those securities ought to trade.

  Now, if you ever wanted to illustrate how really, really smart people could be really, really stupid, those mathematical models would be a prime exhibit. They were wonders of the mathematician’s art: a pleasure-garden filled with Gaussian copulae, stochastic probability projections, and Monte Carlo simulations. And they’re bullshit. The only thing that matters is actual prices in actual situations. In any situation of market turbulence‌—‌the moments, in other words, when pricing really, really matters‌—‌you can bet your life that those models will fail. They’ll fail because they’re based on a whole heap of historical data, generated at a time when the market was not under stress, when pricing was not being generated in a live combat situation.