This is my very non-technical understanding of why everything blew up. It’s based on articles from Paul Krugman, Tyler Cowen, newspaper reporting about the crisis, and particularly from This American Life episode 365. While it’s doubtless more complex in the details, it’s not really a tricky story and there’s no math. It just takes some time to untangle.
In the beginning
Back in 1999, the Senate votes 95-0 to deregulate CDSes. [CDS= Credit Default Swap]. At the time, this made perfect sense. Why? Well, let’s figure out what a CDS started as.
Let’s say that you have some money in savings but you’re not happy with the interest rate. If you’re very conservative [as an investor], you put the money in Treasury Bills. They don’t pay much interest, but the government has never failed to repay them and they can always print more cash if they have to. You can also put your money in a bank CD, or you can start looking at stocks and bonds.
Bonds are safer than stock– they’re debt, something like IOUs, and you’re paid off first if the company goes bankrupt. Of course, the buisness could go bankrupt and have nothing left at all, in which case your bonds are worthless… but that’s rare.
The CDS was created as a way to insure bonds. If I buy CompanyX bonds, but worry that CompanyX may oneday go bankrupt, I might look for insurance against that small chance. That insurance is a CDS. Basically, I pay FredCorp a little money every month and they promise to pay me if the company issuing the bond can’t pay it. That’s what was going on in 1999 when they decided that these instruments are complex and restricted to financially savvy people, so there’s no harm in deregulating them.
Why would you buy a CDS in 1999?
If I buy CompanyX bonds, I can make more money when they pay out their dividends than if I bought Treasury Bonds. If I’m a little worried that CompanyX might not make it, I can buy a CDS to insure me. It cuts into my profit a little, but if the dividend minus the CDS payment is higher than the Treasury Bond interest, I make a little more and I’m still completely safe. (Well, 99.9%….)
How it developed
It is now 2004 or 2006– sometime later after CDSes mutate in purpose. CDSes are still unregulated, but the CDS market is a lot bigger. It’s hard to tell how much bigger, because CDSes are private trades between companies, not tracked on a regulated exchange. Some people still use CDSes for their original purpose– to make sure their bonds are insured against a company’s collapse. But there’s a new use for CDSes that promises lots of money, though it’s very risky.
The new strategy is to buy CDSes for buisnesses when you don’t own any bonds. So I’m buying “insurance” against a company’s collapse, even though I don’t own anything that company issued. Why do you do this? Well, you can make a lot of money. The following is a paraphrase of the This American Life example that really illuminates this.
I’m Joe’s Hedge Fund, worth $100 million dollars. I tell people that I’m willing to issue CDSes for CompanyX. I get lots of interested people, who say, I’ll take it. They promise to pay me $20 million a year for my insuring their $1 billion of bonds. I’m very happy– I can double my money in five years. Because there’s no exchange, no one knows how many companies I’ve promised to insure, and because there’s no government oversght, I don’t have to have the $1 billion in case of collapse.
Why does Joe’s Hedge Fund do that?
Did you see that Joe’s Hedge Fund could double in value in just five years? As a CEO, that’s awfully tempting… look at all that growth you can report to investors. Besides CompanyX won’t go out of buisness (fingers crossed), so I’m not really at risk. [Particularly if CompanyX is perceived as “too big to fail”, implying the government will step in and keep them afloat anyway.]
Why do you buy the CDS from Joe’s fund?
For $20 million, you could get $1 billion. It’s a bet– if you’re right, you’ll make 50 times what you invested.
There are a lot of things that people did to reduce their risk. One common thing was hedging. Let’s continue the example above. Let’s say Joe’s Hedge Fund doesn’t like having a billion dollars hanging over their heads. They can buy a CDS from someone else. If they buy a $1 billion CDS for CompanyX from FredCorp for $15 million, there’s no risk left. Joe’s Hedge Fund is being paid $20 million at the risk of paying out $1 billion– but if they have to pay out the $1 billion, then FredCorp has to pay them $1 billion, which they can just pass on. See, no risk and an easy $5 million a year!
What brought it all tumbling down?
Well, remember the last paragraph of how it developed? No one knows who is insuring what CDS defaults and some people don’t have the money to pay what they’re insuring. [Remember, Joe’s Hedge Fund has only $100 million when they offer the CDS, well short of the $1 billion that they promise to pay.]
Lehman Brothers was an investment bank that has been around since 1850. It made large investments in the subprime mortgages that have been defaulting at high rates. On September 13, 2008 several companies were gathered to see if anyone wanted to buy Lehman Brothers. No one was interested in the company as a whole. The next morning they filed for bankruptcy.
By this time, the CDS market was insuring over $45 trillion dollars– in the ballpark of the world’s yearly production. One company, AIG, happened to have $400 billion of promises to pay if Lehman Brothers defaulted. They didn’t have the cash on hand, which meant that everyone depending on AIG to pay their part was now short what they needed to pay, even if they’d hedged their bets.
It sounds like it just snowballed from there. Each person’s failure to pay (even though they were insured against just this event!) rolled over onto the next person, who also couldn’t pay…
There were other problems that cropped up from Lehman’s failure, including the losses they inflicted on people who were loaning them a day’s funds on the commercial paper market.