The Big Short, by Michael Lewis, is an amazing book about the banking crisis of 2008. Having watched the events unfold over the course of about a year, and not really understanding everything involved, the tragedy of situation wasn’t quite as impressive to me at the time, as it is having read Lewis’ concise, clear and compressed explanation of it. While I’d encourage everyone interested to read the book, I’m going to try to summarize the story here.
A bond is a promise — usually from a government or corporation — to make regular interest payments on borrowed money, and, eventually, to pay back the borrowed principal. For generations, financial markets have traded bonds.
Given that a bond represents an income stream based on borrowed money, Wall Street, in the late 1980s realized that it could create “bond-like” financial products from other debt-based income streams like credit cards, student loans and home mortgages.
The “mortgage bond” was born, and became another financial product bought and sold by Wall Street investment banks, such as Goldman Sachs, Merrill Lynch, Bear Sterns, JP Morgan and Morgan Stanley.
The mortgage bond collects thousands of home mortgages, purchased from lenders, and packages their associated income streams (monthly mortgage payments) into a financial product, that can be bought and sold like a bonds. Mortgage bonds, however, suffer from a couple of unique problems, related to the fact that home owners often refinance their debt during periods of low interest rates, prematurely repaying the principal.
To address this, Wall Street structured mortgage bonds into stacked layers (called “tranches”) — the lowest layer representing the first N mortgages to be paid off early, and the highest layer being the last N mortgages. Investors seeking the higher returns on their money (and accepting the highest risk) could invest in the lower “tranches”, and those wishing lower return (and lower risk), could invest in the higher tranches.
In the 1990s, Wall Street firms began to create mortgage bonds from “subprime” mortgages, i.e. mortgages of much higher risk, but paying much higher interest rates, made to borrowers with lower levels of credit. The structural “tranches” of the mortgage bonds built from subprime mortgages, at this point, represented not only pre-payment, but also outright default.
With increasing demand from Wall Street to buy subprime mortgages, lenders became motivated to place ever more subprime loans (since they were no longer at risk, should the loans fail), and began to push messages like, “refinance your home, unlock all that equity, pay off your credit card debt, and go on vacation.” Often, lenders convinced those without credit and who can’t afford a mortgage at all to get one anyway. To entices these consumers, a new type of mortgage was created — variable rate, with extremely low (even zero) initial interest rates, which later reset to higher levels.
Americans took on these mortgages in masses, not realizing that the real estate bubble forming around them was being fueled by their own actions.
As the underlying mortgages became of lower quality, Wall Street’s mortgage bonds became inherently riskier, which should have made them more difficult to sell to investors. Why? Because buyers of Wall Street products look to the rating agencies Moodys and Standard & Poors for guidances, through their ratings, and risker products are supposed to receive lower ratings.
But there’s an inherent conflict of interest between Wall Street and the rating agencies, since it’s Wall Street who pays the agencies to rate their products. Likely due to this conflict, the rating agencies assigned surprisingly high ratings for these ever-riskier mortgage bonds.
Using models provided to them by the Wall Street firms, the agencies would rate mortgage bonds based on the average borrower FICO scores. This allowed Wall Street firms to structure bonds to contain mortgages from both high and low FICO borrowers, to increase the overall bond rating. It never occur to the rating agencies that the solvency of a bond composed of 10 borrowers of score 680 is dramatically different than one with five 700s and five 670s, since only a relatively small percentage of the underlying mortgages needed to default for the bond to fail.
Despite the boom in mortgage bonds, Wall Street’s desire for ever more profits grew stronger led them to focus on the relatively lower ratings of the bottom (riskiest) tranches of the mortgage bonds. They came up with a clever idea. If they could package the bottom tranches of hundreds of different mortgage bonds together, then on the principal of diversification, perhaps they could convince Moodys and S&P to give higher ratings to the collection as a whole.
That’s exactly what happened, and the “Collateralized Debt Obligation” (CDO) was born. What in retrospect seems unthinkable, the rating agencies gave CDOs a rating of triple-A (AAA) — communicating a risk rating equivalent to US Treasuries. This was based on the premise that if one group of Americans began to default on their mortgages, it would be unlikely that other groups would.
Just think about that.
These AAA ratings opened the door to a huge market for Wall Street firms — allowing them to sell CDOs to organizations such as state and private pension funds, whose bylaws prohibited them from investing in anything other than AAA-rated financial products.
Rather than focus on the culprits, the The Big Short tells the story through the eyes of the few who foresaw the coming collapse, and made fortunes as a result. One was Mike Burry, a young man with autism and a glass eye (from a childhood cancer tumor).
Mike studied what was happening, and performed deep analysis of the underlying mortgages. He recognized that the enormous demand by Wall Street for mortgages drove the lending process, which in turn artificially drove up housing prices, creating an unsustainable real estate bubble from the fabric of financially fragile American consumers. When that bubble would eventually burst, through massive defaults, he realized it would result in the collapse of the entire mortgage-backed financial markets.
Considering how he could profit from this collapse, Mike went to Deutsch Bank and asked, “Can I buy insurance against the failure of a mortgage bond?” Deutsch Bank obliged, and when a couple more people wanted such insurance, an industry standard product was conceived, and the “Credit Default Swap” (CDS) was born.
A credit default swap is an insurance policy, against something you don’t have to own yourself. It’s a mechanism to speculate. To purchase a credit default swap, Mike (and others) paid regular insurance premiums to insure massive dollar amounts of CDOs and mortgage bonds (again, which they didn’t own themselves). As long as the CDOs and bonds didn’t default, the insurers made profits on the premiums. If, over the life of the CDOs and mortgage bonds, they collapsed due to defaults, then Mike (and the others) would be paid fortunes.
From 2003 through 2007, Mike and a few others built up large portfolios of credit default swaps, paying their regular insurance premiums and waiting for the day their investment would reap fortunes.
They often wondered who was on the other side of their bet — i.e. who was selling the insurance. Turns out, it was the world’s largest insurer, AIG. Somehow, through utter incompetency, the world’s largest insurer insured massive amounts of CDOs and mortgage bonds. Rather than deeply analyze the internal makeup of these CDOs, they were content to trust the AAA ratings of Moodys and Standard & Poors. To them, it was like insuring US Treasuries, and so they considered their regular income stream of insurance premiums to be easy profits.
An important aspect of the credit default swap market was the absence of regulation, the way insurance is normally regulated. For example, AIG wasn’t required to post a percentage of insured asset as collateral.
Around 2007, the market for mortgages was drying up, home prices were leveling off, and defaults were already on the rise. Rather than seeing the obvious by now, the Wall Street investment banks remained focused on the continued sale of CDOs. But with their source of mortgages drying up, what income stream could they collect, and pack into these CDOs?
The answer was the income streams from credit default swaps. So the banks themselves got into the business of selling credit default swaps, and packaging those income streams (the insurance premiums) into new CDOs.
By 2008, the Wall Street investment banks were making obscene profits, but held large amounts of CDOs and mortgage bonds waiting to be sold. They were also on the liability end of huge amounts of credit default swaps, often sold and exchanged between themselves. AIG was on the liability end of billions of dollars worth of credit default swaps. Pension funds around the world had huge investments in CDOs and mortgage bonds. And then everything collapsed.
Real estate prices began to drop. American began defaulting in masses on their mortgages. The investment banks began seeing losses on their CDOs and mortgage bonds, and the market for hedging credit default swaps collapsed. Reports of potential insolvency of Bear Stearns surfaced and its stock collapsed, triggering the same for the other financial firms. The government allowed Lehman Brothers to go bankrupt, which triggered more panic in the markets. Commercial lending froze, paralyzing businesses in American and across the globe. Many Americans lost their jobs, savings and retirement funds.
With AIG and other of the world’s largest financial firms facing collapse, the US government stepped in, and bailed them out, paying off their debts, assuming their liabilities, and placing a burden of debt on the American people that likely won’t be paid off through the lifetimes of our grandchildren.
And after it was all over, the executives of these same Wall Street firms went home with billions of dollars of taxpayer money in their pockets, in the form of bonuses.
Wall Street, in its search for profits, created products based on ever riskier mortgages, and found ways to sell them as secure investments.
The rating agencies, Moodys and S&P, gave risky assets gold-plated ratings, equal to US treasuries, which opened the door to a huge market of CDO buyers.
Having someone to sell their mortgages to, lenders were no longer concerned about whether a borrower could pay them back. Freed of risk, and in search of profits, lenders used deceitful tactics to convince Americans to take out mortgages that they couldn’t afford.
Americans, with their insatiable desires for consumption and for whom the average savings level had dropped to less than 1% (consider that the Chinese save 30%), accepted these mortgages in masses, giving up the equity in their homes.
When everything collapsed, the US government rescued the Wall Street firms, passing an unimaginably large bill to the US tax payers, and creating a debt burden that will take generations to repay.
Much of the US tax payers’ rescue money went in the private pockets of the Wall Street executives, in the form of bonuses.
It seems hard to uniquely place the blame.
I’ve always been a believer in the free-market philosophies of Milton Friedman, and I wonder whether, in the long run, it would have been better to allow the banks to fail, even if it meant the collapse the American economy. (I also now question my own beliefs against government involvement in business through regulation.)
Perhaps all we did is simply defer that collapse (and perhaps a more consequential version of it) to the lifetimes of our children, or their children. At least in allowing a collapse, our children could have likely looked back, and have seen the dire consequences of irresponsibility and unbounded greed.
Reading the thesis work of A.K. Barnett-Hart, discussed in this WSJ article, she believes the AAA ratings from the rating agencies wasn’t due to a conflict of interest:
The errors of the rating agencies stemmed from neither conflicts of interest nor preferential treatment given to certain banks. The true culprit behind the rating agencies’ failure was the outsourcing of credit analysis to computer models and the low level of human input used to rate CDOs.
Extremely interesting article. And the thesis is directly downloadable here.