Take a shot of whiskey for courage and then let’s get started.
The story starts with Mortgage Backed Securities. A mortgage backed security (MBS) is a bond backed by a pool of individual mortgages. The government-chartered mortgage brokers Fannie Mae and Freddie Mac (FMFM) have always had the discretion to create them. In the old days they worked like this.
I took out a mortgage from a bank. The bank sold that mortgage to Fannie, recovering capital they can use for more investments. Fannie packaged most of those mortgages into MBS’s which they sold to investors. Fannie only accepted mortgages that met certain strict criteria. That effectively regulated the entire market, creating requirements around credit ratings, loan to equity ratios, inspections, insurance, etc. This stability and predictability translated into real estate prices that appreciated at a slow, stable rate throughout the post-war period.
This system limited the creativity of the mortgage market. Since FMFM funded well over half the total mortgage market, their rules set the bar. If you needed flexibility on some term (self-employed with income difficult to establish, private contractors, short credit history, etc) then a mortgage became much more expensive because it had to go through a different channel.
Mortgage backed securities have a quirk that used to limit their use beyond FMFM. They suffer unpredictable returns. This is less due to the risk of default (historically small) than because of the significant risk of early repayment, which dramatically reduces the total revenue from the note.
Along came an additional innovation in the ‘90’s – the Collateralized Debt Obligation (CDO). A CDO is a pool of diverse assets, in this case perhaps mortgage backed securities of different qualities and ratings, carefully aggregated and arranged to hedge the risks from the various MBS’s. These were usually divided into tiers called “tranches.” Buyers of the riskiest tier of a CDO got the highest interest rates, with lower returns for the buyers of the safest tranches.
The bottom tranche typically would be wiped out by the first 3-5% of loses on the CDO. But the assumptions about the history of the real estate market suggested that this could almost never happen. And the higher tranches were often considered as secure as anything in the investment market.
But even these instruments were tough to properly rate and price. Plus, by the late nineties they were beginning to attract regulatory scrutiny. Then in 2001, two important things happened. There was a new President who embraced a radical withdrawal from financial regulation. And a new mathematical model was developed that allowed rapid pricing of arcane derivatives like CDO’s.
The weakness with this new model was that it was reliant on historical trends. When applied to the real estate market that weakness became a hidden trap. The real estate market was exceptionally stable over a long period of time which under the model made CDO’s based on MBO’s look like a very safe investment. However, once Wall Street got its teeth into the real estate market they dramatically changed the facts on the ground and the model did not adjust.
In the ‘aughts, a new generation of companies began to capitalize on the potential to market mortgages to Wall Street in the form of MBS’s packaged into CDO’s. Opening up mortgage lending to broad Wall Street investment created new capital for unconventional lending – lending to borrowers whose mortgages could not have qualified to be sold to FMFM.
This could have been a good thing. It made it easier, for example, for lenders to work with contract workers, solo professionals, and minorities.
There was still more innovation to come. First, CDO’s could be held by federally regulated entities only if the CDO’s maintained the highest rating, AAA. What’s more, if they achieved this rating, they could be held essentially off the books, like treasury bills. As mortgage practices on the ground deteriorated, with companies searching high and wide to find buyers to meet Wall Street’s fresh demand for MBS’s, traders needed to get creative to try to achieve the necessary AAA ratings.
Bond crafters responded by making the tranches of CDO’s ever more complex and opaque. The two major ratings agencies, Moody’s and S&P were earning enormous profits from this new business. Though analysts at times complained that they were not getting enough insight into the nature of the actual mortgages being diced up into the various CDO tranches, superiors simply pointed to the model and told them to assign a rating.
Now buckle up, because this gets even more complex. Here comes the weapon of mass destruction.
Firms looking to free up capital and hedge against risk decided that CDO’s, especially the lower tranches with higher returns and risk, could be made even more profitable if they could insure against the modest risk they thought they were taking. The Credit Default Swap (CDS) was the perfect instrument.
A CDS is essentially an insurance policy. Assume I own a $1m dollar bond paying 8% interest. There is a small, but real risk that the bond issuer might fail. So I contract with, say AIG, to insure that bond. I pay AIG a modest premium based on their assessment of the bond’s risk, perhaps $20K a year. In the event that the bond issuer fails to pay the bond, AIG is on the hook to cover the loss. The arrangement lasts for the length of the bond’s term. During that time AIG has to post collateral. If the rated risk of the bond (the rating coming from the rating agencies, or the value assessed by the market) shifts, AIG might have to post more collateral or the bond holder might have to pay a higher premium.
Still with me?
A few important details here. First, a CDS functions just like an insurance policy, but it is not an insurance policy, hence is not subject to the regulations that exist to keep insurance credible, solvent, and ethical. Why wasn’t it brought under the regulatory umbrella? Did I mention who was in charge during this period? Why didn’t Democrats force the matter? The unprecedented new lending to low income and minority borrowers provided a sort of regulatory hedge. Think of it as a big fat political steak waved under the nose of the watchdog, making it easier to mislead or co-opt lawmakers.
Second, notice the importance of the ratings agency in this mix. If the raters are fooled, co-opted, or negligent, bad things can happen. Very bad things.
But up to this point, the situation is still not quite out of control. Here’s where it goes nuts. CDS’s themselves were traded like bonds. And since they were for all practical purposes unregulated, there was nothing to keep people from selling me a Credit Default Swap on assets I don’t even own. In effect, it was as if I could issue or buy insurance on your car. This was important by about 2005 because the market was running out of decent new mortgages to bundle into MBS’s, assemble into incomprehensible CDO’s, and issue CDS’s on. So they began to spin up CDS’s in a sort of circular form, rising into trillions of dollars (estimated at possibly $50tr+ for the entire market by 2008).
Apparently, even the bond issuers got into the CDS game. It is possible to use a CDS to effectively bet against a particular bond. Late in the game some CDO issuers started investing in CDS’s against the mortgage market. In essence, building lousy CDO’s, then taking out insurance that would pay when the market failed. This is documented in Michael Lewis’ book and is broader than the SEC’s similar fraud case against Goldman Sachs.
So to review, we have mortgages which have been assembled into bonds called MBS’s. Those bonds are diced up into CDO’s, which are officially classed into different tranches to reflect the varying degrees of default risk, but in fact are being obscured and manipulated so that by 2005 many of them are junk, top to bottom. Wall Street firms, banks, and insurers are issuing CDS’s to insure the holders of these CDO’s, but are also just issuing CDS’s on top of CDS’s on top of CDS’s in order to profit from the churn. And each CDS ties up a small, variable chunk of collateral from the issuer.
Under the pressure of so much swirling money, even Freddie and Fannie got into the act by 2004.
And why would they engage in this risky behavior? Money, of course, but more to the point real estate never declines in value, right? Much of this activity would be solid if the assumptions about past real estate performance still made sense. Plus, FMFM were beginning to lose a significant portion of the market and they wanted a piece. But of course, practices on the ground had changed dramatically. These were no longer traditional mortgage investments and it was not your grandfather’s mortgage market.
We’re almost done. But I’ve left out an important element. The mortgage lenders, although they sold off the mortgages to be assembled into MBO’s, generally faced a contractual provision to keep them “honest.” They had to hold a portion of the riskiest tranches of the loans they generated for a term, usually two years. This meant they only needed buyers to keep paying and not refinance for two years and they were in the clear.
They couldn’t care less what you as a borrower might do with your completely unaffordable mortgage after the interest rate reset three years down the line. Someone else’s problem.
Except for one thing – by the end of 2005 many of the new loans were so extraordinarily lousy that buyers started defaulting after only a few months. By the end of 2006 many of the largest mortgage lenders were collapsing under the weight of failed mortgages. If ratings agencies had responded appropriately, there would have been painful collateral calls all up and down the stack of MBS’s, CDO’s, and CDS’s. This would have triggered an S&L-style banking crisis. In other words, a serious, but manageable crisis. That didn’t happen and it continued to get worse.
We still don’t know exactly why the ratings agencies failed to respond, but it’s not hard to guess. It isn’t easy to say no to another billion dollars in fees. Besides, the obsolete model they were all pointing to still indicated that these investments were solid. Collateral requirements and bond ratings did not adjust while the underlying assets turned foul. The industry continued to ramp up higher and higher on a foundation of garbage.
By the time it was over, there were so many trillions of dollars of capital wound up in this idiotic system that its sudden collapse would have threatened not just banking or capitalism, but civilization itself.
As one trader related in a quote from Michael Lewis’ book:
“When banking stops, credit stops, trade stops, and when trade stops – well, the City of Chicago had only eight days supply of chlorine on hand for its water supply.”
It finally ended when failures of MBS-denominated securities became too visible to ignore and collateral requirements were raised by the ratings agencies. Arguably the first major casualty was Bear Stearns in July 2007. For a long time insiders were able to claim that this was a classic “bank-run” where a failure in confidence caused withdrawals, but of course that myth didn’t hold up for long. By the time AIG went down in the fall of ’08 there was no more room to pretend. It was over and the government stepped in to cushion the fall for all of us.
As a post-script, I still chuckle when I hear someone on TV say that mortgage holders should be continuing to pay their lousy loans on over-valued homes out of some misplaced ethical obligation. The companies that built the crisis and understand the true nature of their obligations are laughing and doing the opposite.
Incidentally, there is an alternate-reality version of what caused the financial collapse promoted by the right flank of the GOP and the banking industry. It blames the insidious 1977 Community Reinvestment Act, a law banning discriminatory banking practices like red-lining, for the crash.
How did an obscure law passed thirty years ago cause the crash? Where do UFO’s come from? Who’s hiding Obama’s birth certificate? Why let facts interfere with ideology? Here’s a description from an enthusiast.
Thanks for riding along through this one. Naturally this misses a lot of detail. What am I leaving out that’s important?
From the Web:
Busted, by Edmund Andrews
And Then the Roof Caved In, by David Faber
The Big Short, by Michael Lewis