Monday, September 03, 2007

Housing and CDS

We are undeniably in a period of high volatility with whipsaws every day, or at least as often as my blog entries. Having mentioned several positive factors and indicators in my last post and enjoyed last Friday’s uptick as much as any long, it’s only appropriate to look at the other side of the coin.

Housing remains the bogeyman. Tim Iacono of TheMessThatGreenSpanMade wrote a great piece to rebut one of the common arguments bulls use to dismiss subprime worries:

He starts by quoting an article Ben Stein wrote in New York Times:

The total mortgage market in the United States is roughly $10.4 trillion. Of that, a little over 13 percent, or about $1.35 trillion, is subprime — certainly a large sum. Of this, nearly 14 percent is delinquent, meaning late in payment or in foreclosure. Of this amount, about 5 percent is actually in foreclosure, or about $67 billion. Of this amount, according to my friends in real estate, at least about half will be recovered in foreclosure. So now we are down to losses of about $33 billion to $34 billion.
The rate of loss in subprime mortgages keeps climbing. In time, perhaps it will double, maybe back to $67 billion. This is a large sum by absolute standards, and I would sure like to have it in my bank account.
But by the metrics of a large economy, it is nothing. The total wealth of the United States is about $70 trillion.

Then he proceeds to rip apart that argument.

According to the latest Federal Reserve Z1 report, the total value of owner-occupied housing is about $23 trillion. As shown in the chart below, the total is up considerably from just a few years ago as indicated in blue, however, the curve looks like it's beginning to flatten - keep an eye on that.
Who's been setting a lot of these prices at the margin?
Subprime borrowers.
They've been pushing home prices up from the bottom, paying more for starter homes, enabling more typical homeowners to cash out and trade up and so on, until home prices in your neighborhood are affected too.
Over at Pimco, they call them Plankton. If it turns out that people have been paying way too much for real estate in recent years because money has been so cheap and lenders so lax (it's looking more likely every day), markets will revert to the mean and there could be hell to pay.
A 10 percent drop from that $23 trillion total would be a loss of $2.3 trillion in national real estate wealth - a 20 percent decline would result in almost a $5 trillion hit.
That's a lot less home equity withdrawal and a lot fewer reverse mortgages - consumer spending and ultimately the broader economy will suffer. Maybe a lot.
These are big numbers - this is the giant thing hiding in the closet.

Goldman Sachs (via CalculatedRisk) just issued a report that estimated a 7% drop in home prices in both ’07 and ’08 basis the Case-Shiller index. GS definitely is not a permabear, so this report carries a lot of weight.

Credit default swaps
The other skeleton in the closet is CDS (credit default swap) which is basically a insurance against default on the underlying debt instrument [Here’s a short primer on CDS from which I will quote below]. For example, the holder of a high yielding junk bond can buy CDS to hedge the default risk of the bond. As long as the cost of the CDS is less than the spread between the bond yield and treasury yield, he has “safely” enhanced his return above that of the treasury. However, this safety is dependent upon solvency of the CDS issuer (aka the counterparty) just as an insurance policy is only good if the insurance company can pay up.

CDS has been a boon to the credit market. It allows risk to be transferred from one party to another and less risk concentration is certainly a good thing. However, one can’t expect Wall St. to leave a good thing alone. CDS "has grown from a $1 trillion industry a few years ago to a $26 trillion industry today." That’s because punters (mostly hedge funds) can buy or sell CDS on an asset that they don’t own – simply as a bet on the perceived default risk.

Heavy concentration of CDS in a few hands makes a disastrous "chain reaction" possible:

CDS doesn't exist on an exchange, much of the volume transfers through few hands -- namely the big banks like Goldman Sachs, JP Morgan, Morgan Stanley, and the like. Nobody on the outside knows how much CDS these banks hold and what kind of directional exposure they have. If hedge funds hold $100 billion of CDS protection and Morgan Stanley took the other side of the trade, if that $100 billion of debt defaulted with no recovery, Morgan Stanley has to pay $100 billion to those hedge funds. I highly doubt Morgan Stanley has $100 billion in loss reserves set aside. Should Morgan Stanley be unable to pay its debts, holders of Morgan Stanley CDS would then come into play -- and suppose Lehman Brothers is on the short end of $50 billion of Morgan Stanley CDS. You can see where this is going. It sounds absurd to say, but it's not inconceivable that a CDS domino effect could destroy the debt and credit markets as we know them.

This is why Warren Buffet called it (and other derivatives) financial weapons of mass destruction.

The burning questions are, “How much of CDS on subprime MBS was issued?”, “Are they considerably more than the amount of subprime mortgages outstanding?”, and “Who are the bag holders?”. Before they can be answered, “containment” is but a pipe dream.

A relief rally to sell into?
I’ve harped on separating the liquidity problem with weakness in the underlying economy. It’s a simplification because tight lending standards will put a crimp on access to credit and consumption. Nonetheless, if you believe that the Fed will be successful in restoring confidence in the credit market before the proverbial “second (or third, fourth…) shoe” drops, then there would be a relief rally to sell into. We’ll know this week when all the bigwigs are back from the Hamptons whether last Wednesday (8/29) marked the start of this rally.