The Less so Bleedin’ Obvious – CDO’s Created Crisis

After yesterday's blog
about the state of the markets and the concluding line: "When you see a
way of making money with no risks, you see the fool’s gold", I guess
the obvious question is: Why didn’t the banks see the fool's gold? 

After
all, these are institutions that go back over five centuries, with
departments that deal with market risk, credit risk, operational risk
and have Advanced Risk Measurement Systems. 

Forget the idea of systemic risk
as, after all, these highly advanced, highly automated institutions
would never get involved in such folly surely (apart from the odd algorithmic explosion).

Some of our community came up with comments though, with Andrew Reynolds saying: “Was
it bank policy to write the loans and hope to profit from the capital
appreciation on foreclosure or were the incentives for the brokers such
that they did not care about the credit quality?”

So
here's my take on it.  The current crisis is mainly related to the
latter and, to a great extent, the spread of and widespread usage of
CDO’s which hid the risks.  That's why the banks didn't see the fool's
gold.

Now I warn you, before you read any further, that this is
a long piece.  It's not funny and it's a bit academic, and so you may
want to click onto the next news item ... but if you're really
wondering why our banks are in such a state, read on.

CDO’s
– Collateralized Debt Obligations – allowed banks to take on higher
risk lending strategies because they could get rid of the risk to other
banks and corporates.  A paper was written in depth on this area by
Joseph Mason, an associate finance professor at Drexel University's
business school, and Joshua Rosner, a managing director at research
firm Graham Fisher & Co.  The paper, titled “How Resilient Are Mortgage Backed Securities to Collateralized Debt Obligation Market Disruptions?”
(37 page pdf), came out in February and has some interesting stats on
this market showing that the growth of CDOs has been explosive.  In
1995, there were hardly any rising to more than $500 billion by 2006 of
which 40% were backed by residential mortgage and almost three quarters
of those were in subprime.

CDO’s allowed the banks to lay
off more risk to others, and were created specifically as a complex
credit derivative to allow greater lending liquidity.  As a result,
banks could lend excessively and relax their credit risk control
measures to a large extent by laying off the risks to other banks and
corporates through CDO’s. 

The fact that these risks were
minimised through collateral, a house, that was worth more than the
loans meant there was no problem ... apart from a housing price boom
that didn't exist and was fuelled by such lending practices.

This is fool’s folly and now we’re seeing the consequences.

And we have seen it all before.

What?

Sure.

I mean, it rings bells with me going back to America’s thrift crisis of 1989.  Funnily enough, here’s a 118 page report from 2003 looking exactly at the systemic risks in the US mortgage markets and how to avoid this (fat lot of good that did).

But
there’s also solutions from those who did learn their lessons and, for
me, this crisis has most similarity with the Lloyd’s of London
insurance losses in the 1980’s.   That crisis almost meant the closure
of a marketplace that had existed since the 17th century ... ring any
bells with where I started?

What was that crisis all about?

Exactly the same as today’s issue ... just with different nuances.

The Lloyd’s crisis began because they took on more and more risk during the early 1980’s and thought they had it covered. 

The
Lloyd’s underwriting syndicates took on the risks because they could
lay-off the risk to external organisations, who reinsure and cover the
exposure.  Eventually, at the top of the tree, you had an excess of
loss syndicate who covered the greatest exposures if there were ever
disasters.

The thing is there were rarely any disasters of
the magnitude that affected those excess of loss firms.  As a result,
those excess of loss firms were the most profitable Lloyd's syndicates
because they never had to pay out any claims.

Therefore the names joined those excess of loss syndicates and the Lloyd's desks took on more and more risk.

Then it all went wrong and Lloyd’s almost went bust.

What happened?

This is quite complicated but relates to our markets today very strongly.

The
way Lloyd’s excess of loss worked was that you might have an oil rig or
property development worth $1 billion.   That’s a huge risk policy so
the firm would go to Lloyd’s to get the cover as Lloyd’s specialised in
it.

The client would go to an underwriter, and ask to get $1
billion of cover and the underwriter would happily take this on because
they could lay-off the risk.,

They would lay-off $900 million of that risk to reinsurance firms outside
the Lloyd’s markets, so it was no problem.  The trouble was that the
reinsurer laid off their risk to another Lloyd’s syndicate, and so $750
million of the risk came back into Lloyd’s.

Therefore,
the total risk to Lloyd’s was $850 million: $100 million with the first
syndicate and $750 with the second.  Lloyd's troubles started there, as
we shall see later, as they thought they just had $100 million of
exposure.

The cycle would probably go round once more,
with $600 million of the $750 million of the second syndicate being
reinsured outside Lloyd’s.  Then the reinsurer comes back into Lloyd’s
and places $400 million with an excess of loss firm.

This
excess of loss firm was happy to take that last $400 million because
they never had to pay.  Bear in mind, it would have to be a total
disaster to pay out the $1 billion and, even if there was a disaster,
the first $600 million of risk was covered by the others.

That’s why excess of loss firms were always profitable.

But
then the disasters hit.  Asbestosis and pollution claims, combined with
natural disasters like Hurricane Betsy and Piper Alpha, meant that
Lloyd's total risk exposure was impacted.

You suddenly had
reinsurance firms paying out $350 million, and Lloyd’s paying out $650
million of which $400 was with the final desk, the excess of loss desk.

Lloyd’s never knew that would happen because every time the
risk was reinsured outside the Lloyd’s markets, they thought they had
got rid of it.  So what they thought they had exposed themselves to was
$100 million here, $150 million there and $400 million at the end of
the game on another deal that was unlikely to ever be paid.

Suddenly, they realised they were up for $650 million for a single incident! 

The
excess of loss firms also banked on rarely having to pay a claim and,
if they did, it would be once in a blue moon.  After all, total disasters rarely happen.

Then
several total disasters all happened at once, and so the $650 million
exposure was not just for one or two risks, but for several that
over-lapped and were inter-related.  In other words, Lloyd's exposures
due to these disasters meant that one market was covering virtually the
whole world's risks. That's an awful lot for a small market, and almost
meant the market collapsed. 

But Lloyd’s got through it.

A
few syndicates collapsed. Others learnt their lessons and now have
systems that look at risk across the whole market, rather than on an
exposure by exposure basis.

Banks will do the same ... eventually.

So going back to where I started: why didn’t banks see the fool’s gold?

It was because, as with the total risk on a single policy at Lloyd's, the fool’s gold was hidden. 

CDO's
combined with other Credit Default Swap Derivatives and exotic
derivatives, meant that, like Lloyd’s in the 1980’s, the risk got moved
out and then back in.  It got moved out of the bank through their
investment desk’s derivatives operations and moved back into the bank
by allowing them to take on huge mortgage debt risk.

Problem for
tomorrow will be: how will the banks ever get to the stage of seeing
the total risk across all of the markets and all of their operations?

Mmmm ... maybe think about that one next week.