Tuesday, March 25, 2008
Roots of the Crisis
by Tom Bozzo
The question arose as to what my problem with Adam Davidson's analysis on NPR [*]was, exactly. Davidson fielded a listener question asking how a seemingly small problem like subprime mortgage defaults could lead to a widespread financial crisis. Much of Davidson's answer discussed the role of leverage in amplifying losses, which clearly is part of the story.
The offending part was an introduction that ran with the premise of the question — small cause, big effect — leading Davidson to suggest that the situation was akin to having a failure in the rubber-band industry torpedo the whole economy. That, no doubt, was exaggeration for comic purposes. But the entire underlying premise is questionable at best.
The housing industry may be just one sector of the economy (if one, along with finance, that was strongly leading the growth out of the 2001 recession), but more relevant is that the
housing boom or bubble created what's looking like several trillion dollars' worth of now-vanished "equity" on household balance sheets. Thanks to lax lending and borrowing practices, a lot of mortgages were thus collateralized by assets valued at bubble prices. That makes for a large hit to households' finances and to the holders of mortgages and MBS; a big deal. Of course we've seen hundreds of billions of dollars in asset write-downs so far — with quite a bit more to come if the Nouriel Roubinis of the world continue to be right.
That leads us to the next big cause, which after Dr. Hypercube's coinage is the "matryoshka lemons" problem. That is, all the slicing and dicing of mortgages for MBS and MBS-derivative securities left the market not only unsure who was holding bad (or potentially bad) debts, but in fact all but unable to determine the actual quality of the securities. Hence, once large and liquid markets dried up overnight. The first casualties here were some lenders who relied on the ability to borrow short, lend long, and quickly pack securitized loans off to investors. This is also a big deal, since securitization is the linchpin of postmodern mortgage lending.
Having borrowed lots of money to make bad bets in illiquid markets then is a follow-on consequence of the big systemic problems.
In short, I think a better answer would have been that the triggering events — deflating the housing boom in general, and the securitization market failures — are big and not small. Subprime was maybe the likeliest starting point because so many of those loan contracts were structured to fail in response to the least shock, but ultimately the subprime failures were symptomatic of the general overvaluation of real-estate assets rather than causal.
In effect, Davidson's account (especially in the summary of the response linked above) is that leverage took a small problem in the housing market — i.e., subprime — and made it big. Correctly accounting for the roles of the housing bubble and the MBS market failures suggests that leverage took a big problem and made it fast.
Why does the distinction matter? The lessons and eventual remedies could well vary a lot depending on whether we've been seeing an amplification of problems created in rogue elements of the financial system, or whether the "shadow banking system" is fundamentally broken. I think the evidence points toward the latter situation.
[*] The NPR page now has a summary, but not a full transcript of the proceedings.
(Note, this post has been expanded and edited from the original version posted over the weekend, and bumped. -Ed.)
The question arose as to what my problem with Adam Davidson's analysis on NPR [*]was, exactly. Davidson fielded a listener question asking how a seemingly small problem like subprime mortgage defaults could lead to a widespread financial crisis. Much of Davidson's answer discussed the role of leverage in amplifying losses, which clearly is part of the story.
The offending part was an introduction that ran with the premise of the question — small cause, big effect — leading Davidson to suggest that the situation was akin to having a failure in the rubber-band industry torpedo the whole economy. That, no doubt, was exaggeration for comic purposes. But the entire underlying premise is questionable at best.
The housing industry may be just one sector of the economy (if one, along with finance, that was strongly leading the growth out of the 2001 recession), but more relevant is that the
housing boom or bubble created what's looking like several trillion dollars' worth of now-vanished "equity" on household balance sheets. Thanks to lax lending and borrowing practices, a lot of mortgages were thus collateralized by assets valued at bubble prices. That makes for a large hit to households' finances and to the holders of mortgages and MBS; a big deal. Of course we've seen hundreds of billions of dollars in asset write-downs so far — with quite a bit more to come if the Nouriel Roubinis of the world continue to be right.
That leads us to the next big cause, which after Dr. Hypercube's coinage is the "matryoshka lemons" problem. That is, all the slicing and dicing of mortgages for MBS and MBS-derivative securities left the market not only unsure who was holding bad (or potentially bad) debts, but in fact all but unable to determine the actual quality of the securities. Hence, once large and liquid markets dried up overnight. The first casualties here were some lenders who relied on the ability to borrow short, lend long, and quickly pack securitized loans off to investors. This is also a big deal, since securitization is the linchpin of postmodern mortgage lending.
Having borrowed lots of money to make bad bets in illiquid markets then is a follow-on consequence of the big systemic problems.
In short, I think a better answer would have been that the triggering events — deflating the housing boom in general, and the securitization market failures — are big and not small. Subprime was maybe the likeliest starting point because so many of those loan contracts were structured to fail in response to the least shock, but ultimately the subprime failures were symptomatic of the general overvaluation of real-estate assets rather than causal.
In effect, Davidson's account (especially in the summary of the response linked above) is that leverage took a small problem in the housing market — i.e., subprime — and made it big. Correctly accounting for the roles of the housing bubble and the MBS market failures suggests that leverage took a big problem and made it fast.
Why does the distinction matter? The lessons and eventual remedies could well vary a lot depending on whether we've been seeing an amplification of problems created in rogue elements of the financial system, or whether the "shadow banking system" is fundamentally broken. I think the evidence points toward the latter situation.
[*] The NPR page now has a summary, but not a full transcript of the proceedings.
Labels: Housing Bubble
Comments:
<< Home
Those concentrated in subprime mortgages did end up with a lot of egg on their faces, and everyone else seems to blame them for the bits of egg that have fallen on their own faces as well.
But while the subprime sector may have been juggling way more eggs than they could handle, everyone else is guilty of the mistaken belief that the smaller number eggs they themselves were juggling would also never come down.
But while the subprime sector may have been juggling way more eggs than they could handle, everyone else is guilty of the mistaken belief that the smaller number eggs they themselves were juggling would also never come down.
# posted by 
: Sun Mar 23, 06:56:00 AM
: Sun Mar 23, 06:56:00 AM
The biggest lender in 2001 was Countrywide. The biggest MBS issuer in the early 1990s (and up until recent years) was The Old Firm. Those business models worked for many years.
I still don't believe it was a bailout, since the alternative was clearly worse. (If it is done again, then it's clearly a bailout, and the first was probably intended to be--but that doesn't change that the actions last weekend were not intended to [a] save shareholders, [b] save jobs, or [c] support The Old Firm. It was all about contagion effect, and the key actions didn't come from either the Fed or JPMC—they were from S&P and CITIC, probably not in that order.)
But I'm less inclined to throw the blame on subprime than Mr. Davidson appears to be.
Let's be clear. Subprime was a creation of the market—Say's Law still applies—as a result of hedge funds that asked for (relatively) high-return assets in a market when EM, Corporates, and traditional MBSes were fairly commoditized.
And subprime itself did not (and likely could not) cause the collapse.
This is a case of fixed costs and variable income.
Post a Comment
I still don't believe it was a bailout, since the alternative was clearly worse. (If it is done again, then it's clearly a bailout, and the first was probably intended to be--but that doesn't change that the actions last weekend were not intended to [a] save shareholders, [b] save jobs, or [c] support The Old Firm. It was all about contagion effect, and the key actions didn't come from either the Fed or JPMC—they were from S&P and CITIC, probably not in that order.)
But I'm less inclined to throw the blame on subprime than Mr. Davidson appears to be.
Let's be clear. Subprime was a creation of the market—Say's Law still applies—as a result of hedge funds that asked for (relatively) high-return assets in a market when EM, Corporates, and traditional MBSes were fairly commoditized.
And subprime itself did not (and likely could not) cause the collapse.
This is a case of fixed costs and variable income.
<< Home
