The following is an excellent excerpt from the book “ECONned: How Unenlightened Self Interest Undermined Democracy and Corrupted Capitalism” by Yves Smith from Chapter 9: “The Heart of Darkness: The Shadow Banking System Self-Interest” from page 242 and I quote: “CDS are economically equivalent to credit insurance and are largely unregulated. The party writing the insurance (the “protection seller,” acting in the role of guarantor) will make a payment to the party buying the policy if a corporation (“reference entity”) specified in the CDS contract defaults on its debt or goes into bankruptcy. The party providing the guarantee gets regular payments, just as an insurance company receives premiums on its policies. Originally, banks used credit default swaps to reduce the risk of loans they held on their balance sheets, but as we will discuss shortly, the market morphed into an unregulated casino.
Thus CDS “protection buyers” are shorting the credit risk, poised to profit if the borrower defaults. CDS have become popular precisely because they offer a ready way to approximate shorting a bond (unlike stocks, bonds are often difficult to borrow, so many debt instruments cannot readily be shorted in the cash market).
By contrast, the seller of CDS protection does want the borrower to do well. A “protection seller” has many of the same risks and payoffs as if he purchased a bond. He receives regular income (the insurance premiums) just as a bondholder does (the interest payments) and he suffers a loss if the bond defaults. The biggest difference is that the bond buyer buys a security, while a CDS protection seller merely has to post some collateral against the possibility he has to pay out on the contract. Thus CDS allowed investors to take levered bets on bond risks, since the collateral posting requirement is generally much less than the cost of buying the security.
As a result, the bond and credit default swaps markets have become linked via arbitrage. Indeed, the credit default swaps market, which is more liquid than the bond market, often dictates the pricing of new bond issues.
Another important difference between CDS and the credit exposures they “reference” is that, perversely, credit default swap creation is not constrained by activity in the real economy. A company sells bonds it has a need for funding. By contrast, credit default swaps creation is limited only by the need to find two parties to a transaction. Derivatives expert Satyajit Das has noted, “On actively traded names CDS volumes are substantially greater than outstanding debt.” For instance, when the demand for collateralized debt obligations began to outstrip supply, credit default swaps were used to create “synthetic” CDOs. This process allowed for risks to be taken on a scale that would be difficult, it not impossible, with traditional instruments.
But this “free market” was a quality-free market. Several institutions, most prominently AIG and monoline and insurers such as MBIA and Ambac, ended up selling CDS protection at a rate that far outstripped their abilities to make good on their guarantees if stress conditions developed. These insurers were massively undercapitalized. As a result, if one of these insurers faced more claims than it could possibly pay out, then scores of institutions that had relief on it for insurance against credit default risks would suddenly also find themselves to be undercapitalized. A shock to the system could, and did, set off a death spiral that pulled down other institutions as the failures cascaded.
A second aspect of credit default swaps is that if you own a CDS protection contract, then you are paid off if the underlying bond defaults. Suppose that you hold a fire insurance policy on a house you don’t own. You’d be delighted if the house were torched by an arsonist; in fact, you might lob a firebomb yourself. In fact, that sort of behavior cropped up when insurance was first launched in England. As a result, in 1774, England implemented the “British Life Assurance Act” to limit issuance of insurance policies. Only parties who have a legitimate reason to protect against loss (we now call this an “insurable interest”) can obtain insurance policies. This concept is now a fundamental tenet of regulated insurance, but was notably absent in the credit default swaps arena.
All three of these innovations turbo-charged an explosive growth of credit arbitrage strategies by investment banks and hedge funds, which produced the “wall of liquidity” that fueled profligate lending. Greenspan’s super-low interest rates post-2001 provided the impetus for these new approaches. Together, these innovations and strategies led to an acceleration in the growth of debt that was not fully recognized by regulators, and to the extent that it was, they saw it as the result of market forces and therefore salutary. And yet, these new activities nevertheless were backstopped by the authorities when hit by a classic bank panic. Understanding the role of credit trading strategies is therefore essential to interpreting what came to pass.
Consider this classic Wall Street joke. On a slow day, some market-makers decide to start trading a can of sardines. Trader A starts the bidding at $1. B quickly bids $2, and several transactions later, E is the proud owner of the tin for $5.
E opens his new purchase and discovers the sardines have gone bad. He goes back to A and says, “You were selling rotten sardines.!”
A smiles broadly and says, “Son, those aren’t eating sardines. They are trading sardines.”
Let’s say a large-scale market in trading sardines developed, where the price was $5. But the price of sardines in the real world of eating sardines is only $1. What would happen?
You’d see makers of sardines care only about making trading sardines. They’d ramp up production to satisfy demand at the new miraculously profitable price. According to orthodox theory, the influx of supply would lead prices to drop from the $5 level.
But in the world of trading sardines, the price is $5 not due to normal supply/demand considerations but due to dynamics within that market. As long as cheap funding persists, there is virtually unlimited demand for trading sardines at $5. In fact, remember how asset prices and loans interact. If banks are willing to fund a lot of the purchase price of trading sardines at $5, prices could even rise further, since new investors might want to get in on the action. What happens in that scenario?
Anything that can plausibly be called a sardine will go into the can. The can of rotten sardines in the classic sotry is thus no accident. It is precisely the outcome you expect when manias or other mechanisms that produce sustained price distortions take hold.
Notice how this story is wildly at odds with the prevailing economic theory. First, prices are supposed to be a function of supply and demand. The idea that market participants would want to game prices or would be insensitive to increased supply or falling quality is a direct contradiction to fundamental assumptions such as rationality and perfect knowledge. Recall how Krugman and the vast majority of economists simply refused to look further into burgeoning oil prices and consider mechanisms that might not fit a tidy, simple picture.
That conventional view misses the “trading sardines” versus “eating sardines” dynamic. There is no economic theory of how the financial system interacts with the real economy, save the use of interest rate assumptions as inputs into economic models. The idea that a product could have a very different value in the financial realm (for reasons internal to those markets) than in the real economy is completely ignored. Prices of financial assets are seen as the result of informed decisions. Markets are efficient. Prices are assumed to send valid signals, save some short-term noise. And there is no place in mainstream theory for prices in financial markets driving and distorting activity in the real economy. Strangely, this dynamic is well-known empirically. For instance, influxes of speculative “hot money” have repeatedly fueled asset price booms in emerging economies that implode when the speculators exit. But these phenomena sit uncomfortably outside pristine equilibrium theories.
We see a second result vexing to orthodox theory: market outcomes producing socially damaging results. First, if trading sardines go to $5, anyone who really wanted to eat sardines faces much greater costs and much poorer quality. Second, undue resources are devoted to sardine production.
And misallocation of resources is precisely what happened in the credit-glutted United States. Most commentators have focused on the dynamics in the real economy, of seemingly unending rises in housing prices, typical bubble signs of overheated buying, and bad practices, particularly predatory lending. Yet that view misses the impact of the trading-fuelled demand for high yield loans for all sorts, which was particularly acute for risky U.S. mortgages but extended to other credit instruments, such as takeover lending. Cheap funding similarly played a major role in the breakneck pace of mergers and acquisitions, which became more and more frenzied until the onset of the credit contraction, in the summer of 2007. Global mergers for the first six months of 2007 were $2.8 trillion, a remarkable 50% higher than the record level for the same period in 2006. And takeovers for the full year 2006 ran at a stunning seven times the level seen four years prior.
The bubble was in the debt markets, particularly for high-spread, high-risk loans that could be dressed up to look safer than they were. Its effects extended far and wide into the real economy. Credit was the driver, the related asset bubbles mere symptoms.
We will focus on collateralized debt obligations and credit default swaps because they were arguably the most destructive of these new credit products.
We refer here specifically to so-called ABS CDOs, meaning CDOs composed primarily of asset backed securities. These CDOs were based on cash flows from loans (rather than purely from credit default swaps); they have produced catastrophic losses and had clear links to increased lending. When the financial press has discussed CDOs in recent years, it almost always refers to this type unless stated otherwise.
A key driver of the rapid growth of the CDO market was that demand for AAA securities exceeded supply.
Insurers, pension funds, and banks all had reasons, due to regulatory treatment of their holdings, to buy AAA instruments. They were also popular with foreign investors, particularly foreign central banks, who had to hold ever-growing amounts of dollar assets to fund America’s unrelenting trade deficit.
But ever since the Modigliani-Miller theory posited that the value of a company did not depend on its credit ratings, fewer and fewer companies bothered to make the extra effort needed to maintain as AAA. The only native AAA credits were a handful of U.S. and foreign companies and a few sovereign credits.
The elevated level of demand for AAA instruments meant richer prices and skimpier yields than some investors were willing to accept, even when they had institutional imperatives to hold high quality investments. Should they buy AAA paper, and receive lower income due to the keen demand, or buy lower-rated securities that offered more yield, even though they would really rather take less risk (and for investors subject to capital requirements, put up more equity)?
Starting in 2001, conditions fell into place that made premium-yield AAA instruments as well as riskier options seem very attractive, even to those who should have known better. Remember, in finance, if something seems to be too good to be true, it is. The idea that an instrument could pay more interest than other AAA bonds and still legitimately be as safe was questionable from the outset. But a lot of investors had good reason to delude themselves.
When Greenspan dropped the Federal funds rate 1.25%, then 1.0%, the result was a negative real yield, meaning lenders could not change enough, at least on short-term, safe loans, to compensate for the loss of the purchasing power due to inflation. They had to take more risk, either by lending longer term, or by seeking out riskier borrowers who would pay more interest, just to keep up with the erosion in the value of their money.
Put the dilemma in such simple terms and it was pretty obvious creditors and investors were in a no-win situation. They could go hunting for more income, but only if they were willing to accept higher odds of loss.
A new, large source of AAA instruments that offered a higher yield than, say, Treasuries, would therefore be a prized new choice.
Securitization was one way to create AAA securities out of raw material that was not AAA, such as subprime loans. Assembling together a large number of subprime loans and then subdividing the result into tranches made it seem at least plausible that many of the higher tranches would never be exposed to default risk and so deserved the AAA designation.
But the exposure to subprime was a considerable and underappreciated danger, one grossly underestimated by rating agencies, and hence by most investors. They assumed default rates would not be very high, based on the short history of subprime debt, which did not include a recession. In addition, they were too optimistic about the housing market and believed housing prices would continue to rise. For instance, Robert Rodriquez of First Pacific Advisors recounted a 2007 conference call with a rating agency Fitch about subprime mortgages:
“They were highly confident regarding their models and their ratings. My associate asked several questions. ‘What are the key drivers of your rating model?” They responded, FICO scores [consumer credit ratings] and home price appreciation (HPA). . . . .My associate then asked, “What if HPA was flat for an extended period of time?” They responded that their model would start to break down. He then asked, “What if HPA were to decline 1% to 2% for an extended period of time?” They responded that their models would break down completely.”
As we know now, the fall in prices was far worse than 2% for several years running.
In the 2001-2003 period, subprime mortgage bonds seemed like a solution to the yield-hungry investors’ problem: the cash flows from the underlying loans could be sliced and diced to produce instruments that fetched an AAA yet offered more interest than Treasuries. The very best of the AAA subprime bonds, the first three of four AAA-rated tranches, were sold to pension funds, insurance companies, and other investors who were keen for relatively high-yield AAA securities.
On the other hand, the very bottom (equity) tranche of a subprime bond could also find buyers. The securitization was structured so that the equity layer would not only pay a high yield, but it would get paid off very quickly, due to the fact that it had a cushion of extra interest available for the equity holder. The short duration and realtively small size of these bonds made them very attractive to a certain class of hedge fund investors. Since they paid off so quickly, these investors could load up on new ones as the ones they already owned paid down. As a result, the equity pieces of subprime bonds also became easy to sell.
But there was little appetite for the AA through BBB layers of a subprime mortgage bond, which accounted for nearly 20% of the total value. There was a cohort of sophisticated investors that were interested. But the small size of this group limited the amount of subprime that could be securitized, and consequently made these investors fairly powerful. Although the theory was, as we have seen, that structured securities would be popular because each tranche would end up finding its niche, the fact that in practice some tranches were harder to sell would have significant repercussions.
CDOs were originally devised as a way to dress up these junior layers and make them palatable to a wider range of investors, just as unwanted piggie bits get ground up with a little bit of the better cuts and a lot of spices and turned into sausage.
Figure 9.2 on page 248 is a simplified version of a typical ABS CDO structure. Going from left to right in the figure, subprime loans first went into a pool. The principal and interest payments were then allotted to various classes of securities, the “subprime mortgage bonds” rated AAA through the “BB/NR” with NR being “not rated” or “equity” layer.
The key difference between these CDOs and other types of structured credit is that they were resecruitizations, made largely of unwanted pieces of subprime bonds. The CDOs that took the better of the unpopular pieces, the junior AAA, AA, and A layers, were called “high-grade” CDOs. We will focus on the other type, so-called “mezz” ABS CDOs, or simply “mezz” CDOs, which used the BBB or “mezzanine” layer from subprime bond issues. We’ll use the mezz CDOs since the synthetic versions mimicked them, although the same general principles apply to the high-grade variant.
Here again, the magic of structured credit alchemy took pools of loans, and turned them into instruments (tranches) that got different credit ratings. As in other structured securities, the bulk of the value of the resulting CDO, meaning the total cash paid to purchase each of the various tranches, was far and away in the AAA-rated tranches, which typically accounted for 75% to 80% of the total proceeds (versus around 90% for the high-grade variant).
The very worst tranche, the ones to take losses first, was the equity tranche, so-called because it was not rated. It usually accounted for 4% to 7% of the value of the deal. Next up were the mezzanine tranches, rated somewhere in the BBBs, the lowest investment grade, and usually 10% of the deal’s value.
Finally, until the later stages of the credit mania, the BBB tranches of these CDOs were again securities that virtually no one wanted. Often, these unpopular pieces went into later CDOs (the rating agencies tolerated a surprisingly high percentage of pieces from other CDOs in CDOs, up to 40% of high-grade in a so-called high-grade CDO and 10% of mezz in a mezz CDO). In other cases these pieces were attractive to rather exotic investors. But the fact that there was always a problem placing some pieces of the otherwise sought-after CDOs meant that CDOs, in some respects, resembled a Ponzi scheme.
To recap: in these second generation pools, the riskier cash flows from the original subprime bonds were again allocated to various tranches, many of which were then rated AAA. In other words, these CDOs took the worst exposures from weak mortgages and used financial technology to create new instruments of which anywhere from weak mortgages and used financial technology to create new instruments of which anywhere from 75% to 90% was designated AAA.
ABS CDOs were the financial equivalent of turning pigs’ ears into silk purses, and in the end, they worked about as well. How could anyone at the time have convinced himself that these junior exposures to low credit quality instruments could produce AAA-rated paper? The problem is that procedures that made some sense on first generation securitizations were dangerously misleading here.
It’s easy to blame rising real estate prices and ratings agencies, but the real roots lie again in flawed economic models.
Recall the discussion of correlation risk from Chapter 3. The theory, developed by Harry Markowitz and William Sharpe, was that investors could create an optimal portfolio that suited their appetite for risk. But to do that, they needed to find investments whose prices moved differently, and they needed to have precise information about how these prices would move in relationship to each other (“covary”) in the future. In other words, this was a clever idea that would seem to have little practical application, except that a whole industry of faux science was constructed on this flawed foundation.
The way this approach was applied to structuring collateralized debt obligations was particularly dubious. The ratings agencies, the monoline insurers, and many investors looked at the risk of default using correlation models. But correlation is a concept in financial economics used to estimate overall portfolio risk based on price movements of the instruments in that portfolio in relationship to each other. If the price of one holding increases 5% in a day, another could change in a whole range of ways: up even more, up but not as much, no change, or down a lot or a little.
But if one loan defaults, the next will either default or not default. Only simple binary outcomes are possible. Thus using Markowitz/Sharpe-type models to analyze defaults was fundamentally wrongheaded.
Now it is possible that even an inappropriate correlation model could have been brute forced to this task, just as the handle of a screwdriver can be an effective hammer. But the users of these tools made a second error: their correlation models showed that the diversification of resecuritization, of going from a single subprime bond to a new vehicle composed largely of risky bits of subprime bonds, reduced risk in a meaningful way. But that was bogus.
The typical pool backing a subprime mortgage-backed bond would have had typically 4,000 to 5,000 mortgages. So the pool in theory was already diversified, although it could have some concentrated exposures, for instance, by geography (were the loans only in few states?). But the flaw was in thinking that there was much diversity in these pools to begin with. By contrast, when this same approach was applied to corporate loans, there was bona fide lowering of overall default exposure, since, say, a chemical company faced different business challenges and had different management than a telecom.
But with so many mortgages in a single mortgage bond, it would resemble what in stock investing would be called an index with a tilt, for instance, an equity market proxy with an overweighting of a particular sector considered desirable, such as technology or high dividend stocks. So the resecuritization would not reduce the underlying risk save by eliminating any skewing and moving the exposure closer to a subprime index. The only other risk reduction would be due to the fact that a portion of a CDO would need to consist of lower-risk assets, which could be better quality mortgages (although some, like Alt-As, in fact proved to be no improvement) or completely different types of loans, like equipment leases.
Thus the risk reduction of going from a subprime bond tranche to a CDO based on subprime bond tranches was in fact very minor. The market reaction to the original subprime deal should have been a tipoff. Remember, pretty much no one wanted those BBB to AA bits. That means possible buyers thought the default risk was high, and they weren’t getting paid enough to take it.
So those very same pieces were put into these subprime CDOs. And remember, the “mezz” variety of this type of CDO took the BBB pieces. That slice gets wiped out if losses on the underlying pool of loans reaches 8% to 12%. The models optimistically treated that outcome as impossible across a whole bunch of subprime bond deals. Yet a Moody’s report as of September 2009 shows “pipeline” losses of 22% and expected losses of 25.3% across all the 2005 to 2008 subprime mortgage pools underlying the bonds they track.
Thus, unless the CDO manager did a spectacular job of selecting bonds, the investors in ABS CDOs took massive, often total, losses. For instance, a Bloomberg run shows 1,590 ABSCDO tranches that originally had an “investment grade” rating from Standard & Poor’s, meaning BBB or higher. As of this writing, only 29 are still investment grade. And remember, 75% to 90% of the original value of these deals was in AAA-rated tranches.”
(YVES SMITH MENTIONED THE AUTHOR SATYAJIT DAS, WHO WROTE THE BOOK “TRADERS, GUNS & MONEY: KNOWNS AND UNKNOWNS IN THE DAZZLING WORLD OF DERIVATIVES” WHO THIS AUTHOR (YVES SMITH) USES AS AN EXPERT IF SHE HAS ANY QUESTIONS ABOUT DERIVATIVES. THIS SEGMENT TALKS A LOT ABOUT CREDIT DEFAULT SWAPS WHICH GOT AIG INTO TROUBLE AND HAD TO BE BAILED OUT BY THE GOVERNMENT. AND THE FACT THAT STANDARD & POOR’S AND OTHER RATING AGENCIES WERE VIRTUALLY IN BED WITH THE BIG INVESTMENT BANKS AND HEDGE FUNDS. SEC (SECURITIES & EXCHANGE COMMISISON) WAS SUPPOSED TO MONITOR THE DERIVATIVE TRADE BUT WERE LAX. AT THE RATE THE DERIVATIVES ARE GROWING, WITH THE VARIOUS NEW RISKY CONTRACTS THEY ARE DEVELOPING IN THEIR COMPANIES, WITH THE USE OF CORRUPT LAWYERS AND ACCOUNTANTS, YOU CAN SEE WHY 2008 HAPPENED. WE COULD GET A MUCH WORSE FINANCIAL CRASH LIKE THOM HARTMANN REPORTED IN HIS BOOK “THE CRASH OF 2016.” THIS IS THE MAIN REASON THAT BERNIE SANDERS WAS SO POPULAR IN THE DEMOCRATIC PRESIDENTIAL PRIMARY. IF HILLARY CLINTON CAN’T GET THIS PROBLEM SOLVED, WHICH COULD BANKRUPT THIS COUNTRY, AS WELL AS THE WORLD, AFTER BRITAIN VOTED THEMSELVES OUT OF THE EUROPEAN UNION, WE’RE ALL IN TROUBLE.
LaVern Isely, Progressive, Overtaxed, Independent Middle Class Taxpayer and Public Citizen Member and USAF Veteran
isely 