Part 2 (2/2)

Chapter 3.

Plate Tectonics.

A familiar account of the current economic crisis goes something like this: A housing bubble in the United States got out of control sometime around 2005 or 2006. People took out mortgages they couldn't afford and eventually defaulted on them. Having been securitized, however, those mortgages went on to infect and topple the global financial system.

This account blames a few bad apples, subprime borrowers, for the catastrophe. It's rea.s.suring but wrong: while the housing bubble rested in part on subprime mortgages, the problems were more pervasive and widespread. Nor were these problems of recent origin; they were rooted in deep structural changes in the economy that date back many years.

In other words, the securitization of bad loans was but the beginning; long-standing changes in corporate governance and compensation schemes played a role too. Government also shoulders some share of the blame, most obviously the monetary policies pursued by Alan Greenspan. So too do decades of government policies favoring home owners.h.i.+p.

In the end, however, the significance of government intervention was dwarfed by the significance of government inaction. For years federal regulators turned a blind eye to the rise of a new shadow banking system that made the entire financial system dangerously fragile and p.r.o.ne to collapse. These new financial inst.i.tutions battened on the easy money and easy credit made available not only by the Federal Reserve but by emerging economies like China.

These changes may have been invisible to most market watchers, or at the very least, their importance was not fully recognized. Subprime mortgages were but the most obvious sign of a deep and systemic rot. This fact underscores a cardinal principle of crisis economics: the biggest and most destructive financial disasters are not produced by something so inconsequential as subprime mortgages or a few reckless risk takers. Nor are they merely produced by the euphoria of a speculative bubble.

Rather, much as with earthquakes, the pressures build for many years, and when the shock finally comes, it can be staggering. In 2006-8 it was not simply the subprime securities that collapsed in value; the entire edifice of the world's financial system was shaken. The collapse revealed a frightening if familiar truth: the homes of subprime borrowers were not the only structures standing on the proverbial fault line; countless towers of leverage and debt had been built there too.

Financial Innovation.

Many bubbles begin when a burst of innovation or technological progress heralds the dawn of a new economy. In the 1840s Great Britain endured a mania driven by a new technology: the railroad. In 1830 the first commercially successful railroad began carrying pa.s.sengers between Manchester and Liverpool; thereafter investors bought shares in companies that would build even more profitable lines. During the height of the boom in 1845-46, share prices of railroad stocks soared, and corporations built thousands of miles of track, much of it redundant and unnecessary. While that boom ended with a brutal bust, it was justified in part by fundamentals: a new technology begat new business opportunities. Even though most of the railway companies of the 1840s went bankrupt, they left behind a new transportation infrastructure that was essential to the nation's economic expansion throughout the nineteenth century.

The same argument could be made for the dot-com boom of the 1990s. Though it swiftly became a speculative bubble, it was at least partly justified by a new technology-the Internet-and its many promising applications. When this bubble collapsed, plenty of new companies survived, as did a new communications infrastructure of coaxial cable lines, cell phone towers, and other tangible technological improvements.

The recent crisis, by contrast, has left behind few tangible benefits: abandoned housing subdivisions in Las Vegas are next to useless. Worse, no technological revolution underpinned the housing boom: houses built in 2006 were no different and no more efficient than houses built a decade or two earlier. The most recent boom was that rare creature, a boom without any change in fundamentals. It was a speculative bubble and nothing more.

But if technological innovation wasn't driving the housing boom, what was? In fact, there was plenty of innovation-that's the good news. The bad news is that most of it percolated in one sector of the economy, the financial services industry. In itself, this was not a problem. After all, plenty of financial innovations in centuries past-insurance, for example, and commodity options-have proved their value again and again, enabling market partic.i.p.ants to manage and contain risk.

At first that same spirit animated the current crop of financial innovations. Indeed, they were attempts to improve on the older model of making loans. Several decades ago banks that made home loans followed the ”originate and hold” model. A prospective homeowner would apply for a mortgage, and the bank would lend the money, then sit back and collect payments on the princ.i.p.al and interest. The bank that originated the mortgage held the mortgage; it was strictly a transaction between the homeowner and the bank.

Financial innovation changed that. In the 1970s the Government National Mortgage a.s.sociation (better known as Ginnie Mae) put together the first mortgage-backed securities. That is, it pooled mortgages it had originated, then issued bonds on the basis of that pool. Consequently, rather than waiting thirty years to recoup the proceeds from a mortgage, Ginnie Mae could receive a lump sum up front from the purchasers of the bond. In turn, the investors buying these new bonds would receive a certain portion of the revenue stream from the thousands of homeowners paying off their mortgages.

This scheme was revolutionary. Thanks to what was quickly dubbed securitization, illiquid a.s.sets like mortgages could now be pooled and transformed into liquid a.s.sets that were tradable on the open market. These new instruments had a name: mortgage-backed securities. In time other government agencies like Freddie Mac and Fannie Mae joined the securitization business. So too did investment banks, brokerages, and even home builders, all of whom brought together growing numbers of home mortgages into new and ever more profitable pools. Investors around the world snapped them up. After all, according to conventional wisdom, home prices never went down.

Investment banks typically guided the creation of pools of mortgage-backed securities. Working with whoever had originated the pool of mortgages-a bank, a nonbank lender, or a government-sponsored ent.i.ty-an investment bank would help set up a ”special purpose vehicle” (SPV). The SPV would then issue bonds, or mortgage-backed securities, selling them to investors. In theory, everyone got what he wanted with this system. The homeowner got a loan, and the mortgage broker and the appraiser earned their fees. The mortgage lender made a tidy profit without having to wait thirty years. The investment bank earned a fat fee for its a.s.sistance even as it unloaded the risk of the mortgage onto someone else. And last but not least, the investors who purchased the securities looked forward to receiving a steady revenue stream as homeowners paid off their loans.

Though mortgage-backed securities became increasingly popular in the 1980s, it was not until the 1990s that they really took off. In an ironic twist, the savings and loan (S&L) crisis cemented the popularity of securitization. In that debacle more than sixteen hundred thrifts went bust because they'd made a bunch of bad residential and commercial real estate loans that they'd kept on their books (as ”originate and hold” transactions). That would not have happened had the loans been securitized-or at least that's the lesson that many bankers drew from the S&Ls' collapse. The new thinking was simple enough: far better to sell off the loans and pocket a tidy profit up front than hold the loans and risk having them go bad later. Distributing the loans to those better able to shoulder the risk-pension funds, insurance companies, and other inst.i.tutional investors-could lessen the risk of a systemic banking crisis. ”Originate and distribute” replaced ”originate and hold.”

It's a sound principle as long as the buyers of the securities can accurately a.s.sess the risk inherent in them. But if you're a bank selling off newly minted mortgages via the securitization pipeline, your primary objective is to unload as many mortgages as quickly as possible. Each sale gives you more money with which to make more loans. Unfortunately, because the bank no longer faces the consequences of making bad loans, it has much less incentive to properly monitor the underlying risk of the mortgages it originates. When originate and hold becomes originate and distribute, a bad mortgage is pa.s.sed down the line like a hot potato.

As securitization became increasingly commonplace in the 1990s and 2000s, mortgage brokers, mortgage appraisers, ordinary banks, investment banks, and even quasi-public inst.i.tutions like Fannie Mae and Freddie Mac no longer subjected would-be borrowers to careful scrutiny. So-called liar loans became increasingly common, as borrowers fibbed about their income and failed to provide written confirmation of their salary. Most infamous of all were the ”NINJA loans,” in which the borrower had No Income, No Job, (and no) a.s.sets.

Securitization did not stop there. Financial firms oversaw the securitization of commercial real estate mortgages along with many kinds of consumer loans: credit card loans, student loans, and auto loans. Corporate loans were securitized as well, such as leveraged loans and industrial and commercial loans. The resulting bonds-a.s.set-backed securities-proved popular, and securitization soon spread elsewhere. As one textbook on risk management concluded in 2001, ”Sometimes it seems as though almost anything can be securitized.” That was no exaggeration: by the time the crisis. .h.i.t, securitization had been applied to airplane leases, revenues from forests and mines, delinquent tax liens, radio tower revenues, boat loans, state and local government revenues, and even the royalties of rock bands.

Many of these newfangled products suffered from the same problems and temptations a.s.sociated with the first generation of mortgage-backed securities: the bank or firm originating the securities had little incentive to conduct the oversight and due diligence necessary to confirm that the underlying loans would be paid off. The investment banks that had midwifed the birth of these pools of securities did not perform this duty either: they intended to sell the bundled loans and thereby move them off their balance sheets.

In theory, the ratings agencies-Moody's, Fitch, Standard & Poor's-should have sounded the alarm. But relying on the ratings agencies was much like relying on the fox to guard the henhouse: they had every possible incentive to give a high rating to the securities under review (see chapter 8). Doing so earned them a nice fee from the very ent.i.ties they were evaluating and the promise of future business; giving a realistic a.s.sessment, by contrast, could mean losing the commission, along with any future commissions. Far better to award a bank the financial equivalent of a Good Housekeeping Seal of Approval and hope for the best. On the eve of the crisis, the ratings firms made upwards of half their profits from handing out AAA ratings, many of which were undeserved, to exotic structured finance products.

But there is more to the story than corrupt ratings agencies. In fact, the ratings agencies may have had a genuinely difficult time figuring out the likelihood of defaults on the loans pooled into these securities, as very little historical data about the new subprime mortgages and their default rates were available. This was particularly the case with the new, exotic, and complicated mortgage-backed and a.s.set-backed securities first crafted by investment banks in the 1980s. These securities go by different names and different acronyms: collateralized mortgage obligations (CMOs), collateralized debt obligations (CDOs), and collateralized loan obligations (CLOs).

All of them worked according to the same principle. Anyone holding a plain-vanilla mortgage-backed security necessarily took on a certain amount of risk: the homeowner might default, for example, or simply prepay the loan, thereby depriving the lender of the additional interest payments it would earn if the loan was paid off on schedule. Financial ”engineers” on Wall Street came up with an elegant solution: the CDO. The CDO would be divided into slices, or tranches. The simplest CDOs had only three tranches: equity, mezzanine, and senior. The purchasers of an equity tranche got the highest return but took on the greatest risk: if any homeowners in the underlying pool defaulted, the holders of the equity tranche would see losses before anyone else. The mezzanine tranche was less risky, but its purchasers would still suffer losses if a larger percentage of homeowners in the underlying pool defaulted. At the top was the senior tranche. While it paid the lowest rate of return, it was supposed to be risk free or pretty close to it. The holders of the senior tranche got paid first and sustained losses last.

This impressive edifice of structured finance rested on shaky foundations. It depended on a sleight of hand: a bunch of dodgy and risky BBBRATED subprime mortgages would be bundled into a BBB mortgage-backed security and then sliced into tranches in which the senior tranche-approximately 80 percent of the total underlying a.s.sets-would be given an AAA rating. The process transformed toxic waste into a gold-plated security, even though the underlying pool of mortgages was just as risky as it was before.

Securitization achieved even more bizarre levels of complexity. It became fas.h.i.+onable, for instance, to combine CDOs with other CDOs, then split them up into tranches. These CDOs of CDOs (sometimes called a CDO2) paled next to the more baroque products coming out of the labs on Wall Street: CDOs of CDOs of CDOs, better known as CDO3; and synthetic CDOs, which a.s.sembled a bunch of credit default swaps to mimic an underlying CDO. Some of these more esoteric products had far more than three tranches: they might have fifty or even one hundred, each of which represented a certain level of risk tolerance.

In hindsight, the peril of this kind of financial innovation is easy to understand. Slicing and dicing credit risk and transferring it around the world suffused the system with financial instruments that were exotic, complex, and illiquid. These creations became so fiendishly complex and unique that it became difficult to value them by conventional means. Instead of market prices, financial firms resorted to mathematical models to value them. Unfortunately, these models relied on optimistic a.s.sumptions that minimized measured risk. The net result was an utterly opaque, impenetrable financial system ripe for a panic.

This state of affairs may seem unique and unprecedented, and it was, but only in the particulars. Lack of transparency, underestimation of risk, and cluelessness about how new financial products might behave when subjected to significant stress are recurrent problems in many crises, past and present.

Moral Hazard.

While the financial engineers who gave us monstrosities like the CDO3 deserve plenty of blame, many other problems were acc.u.mulating that went far beyond the obvious flaws in the securitization food chain. The faulty ways in which financial firms governed themselves helped lay the groundwork for the recent crisis as well.

The key to understanding this situation is the concept of ”moral hazard.” Simply put, moral hazard is someone's willingness to take risks-particularly excessive risks-that he would normally avoid, simply because he knows someone else will shoulder whatever negative consequences follow if not bail out those who took those risks. For example, someone who has auto theft insurance may be more willing to park his car in a place where it might be stolen, or neglect to buy an anti-theft device, than someone who lacks that insurance. The car owner knows that the insurance company will cover the loss; the problem will fall on someone else's shoulders. Likewise, someone who leases an automobile with a service contract is more likely to drive in ways that subject the car to wear and tear than someone without such a contract. Again, the lessee knows any damage will be someone else's problem.

Moral hazard played a significant role in the recent economic crisis. In the securitization food chain, a mortgage broker who knowingly brought a liar loan to a bank got compensated for his efforts but bore no responsibility for what would happen as the mortgage moved down the line. Likewise, the trader who placed enormous bets on a CDO would be rewarded handsomely if he succeeded but was rarely punished if he failed. Even if he was terminated, he would get to keep whatever compensation he'd accrued over the years. The fallout of his decisions became someone else's problem-namely, the company that employed him.

This observation is pretty familiar. Less well known is the fact that moral hazard was especially rife in the financial services industry because of the way these firms provided compensation. Rather than simply paying employees a salary, the traders and bankers who worked at investment banks, hedge funds, and other financial services firms were rewarded for their performance via a system of annual bonuses. While bonuses have long played a role in compensation at these firms, they soared in recent years, and all the major investment banks-Goldman Sachs, Morgan Stanley, Merrill Lynch, Lehman Brothers, and Bear Stearns-paid ever more staggering sums. In 2005 the big five firms paid $25 billion in bonuses; in 2006 they paid $36 billion; and a year later, $38 billion.

More to the point, the ratio of bonuses to base pay skyrocketed. In 2006 the average bonus accounted for 60 percent of total compensation at the five biggest investment banks. In some cases, the figure was much higher: bonuses ten or even twelve times the size of base salaries became commonplace in many firms at the center of the meltdown. Even after these firms ended up on life support, they continued to pay bonuses.

The bonus system, which focused on short-term profits made over the course of a year, encouraged risk taking and excessive leverage on a ma.s.sive scale. Nowhere was this more evident than at AIG, which specialized in selling insurance on events-the bankruptcy of Lehman Brothers, for example-that were unlikely to materialize in any given year. In the short term, this willingness to wager huge amounts of money insuring against catastrophes yielded large revenues, profits, and bonuses for traders and banks. In the long term, the inevitable happened, and when it did, companies like AIG nearly collapsed. The consequences of these gamblers' decisions ended up being shouldered by someone else-namely, American taxpayers.

In theory, this outbreak of moral hazard should have been prevented, but it was not. Why? The answer lies with what economists call the princ.i.p.al-agent problem. In large-scale capitalist enterprises, the princ.i.p.als (the shareholders and board of directors) must hire other people such as managers (the ”agents”) to carry out their wishes and mind the store. Unfortunately, the agents invariably know more about what's going on than the princ.i.p.als and can pursue their own self-interest to destructive effect.

Think, for example, of the problem of a store owner who has employees minding the cash registers. This is a very basic example of the princ.i.p.al-agent problem. It's obviously in the interest of the store owner to have employees behave honestly and not line their own pockets. But the store owner is not omniscient; he can't see everything that's going on beneath him. He suffers from what economists call the asymmetric information problem, in which the princ.i.p.al (the store owner) knows less than the agent (the cas.h.i.+er). The store owner needs to make his employees serve his interests, and that's no easy task.

Now imagine this problem multiplied many times, with many layers of employees or agents, all of whom have the ability to pursue their own interests at the expense of the ”princ.i.p.als” who oversee them. Moreover, many employees are both princ.i.p.als (responsible for overseeing people below them) and agents (responsible for answering to someone above them). Worse, the problem here is no longer that employees will steal, but that they will use the firm's resources to place outsize, risky bets in order to collect the maximum bonus, even if that means putting the firm in jeopardy.

This, more or less, is the structure of a typical financial firm, and the dangers of this arrangement became increasingly evident over the course of the recent financial crisis. The collapse of AIG may be the most extreme example of the dangers of moral hazard, princ.i.p.al-agent problems, and asymmetric information. There, the actions of a small group of employees based in London brought the entire company to its knees, along with the global financial system.

In theory, shareholders should be able to prevent such disasters: they are the last link in the chain, the ultimate owners of the financial firm. But in fact, shareholders generally don't have much incentive to rein in reckless bankers, traders, and managers. Why? Financial firms rely far more heavily on borrowed money to finance their operations than do ordinary corporations, so when it comes to the firm's day-to-day operations, shareholders don't have much skin in the game. They have little incentive to steer traders away from taking big risks; in fact, they have plenty of incentive to do the opposite. If those risks pay off, shareholders win big. If they don't, shareholders may end up losing their small stake in the company. That's bad news, for sure, but when compared to the potential gains realized by playing with other people's money, it's a risk worth taking. Thus, shareholders with little skin in the game were ”gambling for redemption.”

In theory, one final firewall exists to keep moral hazard in check: the people who lend money to banks and other financial firms. If any party has a strong incentive to monitor banks, they do. After all, they stand to lose their s.h.i.+rts if the bank does something stupid. Unfortunately, this is another example of the law of unintended consequences. Funds lent to most ordinary banks come in the form of deposits. However, most deposits are subject to deposit insurance. So even if a bank recklessly gambles with depositors' money, the depositors can sleep well at night knowing that deposit insurance will make them whole. That removes any incentive for them to take actions that might punish the bank for its bad decisions.

<script>