Felix Salmon

Are you sick of the

liquidity put yet? You shouldn’t be, because it’s only getting more and

more interesting. There are three articles worth reading on the subject today,

starting with David

Reilly’s WSJ column on whether Citigroup should move its $41 billion of

CDO exposure onto its balance sheet. The answer is that yes, of course it should,

since Citi’s exposure to these vehicles is significant. But there also seems

to be a very good chance that it won’t take all this nuclear waste

onto its balance sheet, since there’s a colorable case to be made that accounting

laws don’t require it.

Reilly also moves forward the question of who owns the equity in all these

trades: the answer, it seems, is no one. If you recall the chart

I published on Friday, it shows that the equity investor puts in no money at

all, which essentially means that the bank sponsor has at least as much exposure

to the structure as anybody else. I thought that maybe this extreme structure

was just a Canadian thing, but apparently not:

Like other banks, Citigroup structured these vehicles so they wouldn’t be

included on its books. The vehicles are created as corporate zombies that

ostensibly aren’t owned or controlled by anyone.

But is this just a Citi issue? According

to Peter Eavis, writing on the newly-relaunched fortune.com, no. He says

that Merrill Lynch and Bank of America have similar exposure, although they

might have accounted for it slightly differently. And there might be more banks

we don’t even know about, too:

Almost every major banks has significant conduit exposure. But if conduits

are becoming a problem, banks are not saying much about it in their financial

statements.

I like the way that Eavis lumps in CDO vehicles and SIVs as just different

types of conduit, with much the same kind of balance-sheet risk – I think

that’s exactly right. In other words, a bank’s SIV exposure is always no more

than a lower bound for its overall conduit exposure. And now that HSBC

is taking its SIVs onto its balance sheet, banking-industry best practice

clearly shows that all such conduit exposure belongs on the balance sheet and

not off it.

Finally, Antony Currie explains how this situation came about in the first

place, in an excellent backgrounder

on CDOs in general:

As the market boomed, greed set in. By 2006, investors buying the riskiest

slices of ABS CDOs, called the equity, could virtually dictate their own terms.

And investment banks were willing to bend over backward if it helped win the

business of arranging a deal. One example: Rather than funding the highest-rated

portion of a CDO with long-term debt, banks often used much cheaper short-term

commercial-paper programs.

That saved the CDO money, meaning more interest could be paid to the equity

investors. Some banks even agreed to step in if lenders suddenly snubbed the

short-term debt. That is exactly what happened, and this "liquidity put"

has landed Citigroup with $25 billion, and Bank of America with $15 billion,

of exposure to commercial paper backing CDOs.

Were liquidity puts confined to Citi and BofA? For the time being, it seems

as though yes, they were. But no one knows for sure.

On Wednesday I said

that the notorious "liquidity put", which was allegedly responsible

for tens of billions of dollars in Citigroup losses, was "really nothing

more than a CP backstop". Today we’re learning a lot about something known

as leveraged super senior trades, or LSS. And it’s now becoming clearer exactly

went down with these trades, and what the liquidity put really involved.

Think about it this way. Remember the Bear Stearns hedge funds which imploded?

Their big mistake was to buy highly-rated mortgage-backed paper with borrowed

money. When the paper fell in value, the funds’ leverage meant they were wiped

out. To this day, we still don’t really know what happened to the banks which

lent money to those funds, although I suspect that most of them were paid off

by Bear Stearns. In an LSS trade, however, there was no Bear Stearns to bail

out lenders, and so those lenders ended up taking a massive bath.

In an LSS, investors made a leveraged bet on the super-senior tranches of mortgage-backed

securities. But the leverage wasn’t the kind of leverage that the Bear Stearns

hedge funds used. The Bear Stearns funds simply went to their banks (or "prime

brokers", as they’re known in the hedge-fund world) and borrowed the money

against the value of their portfolios. When those portfolios dropped in value,

the prime brokers started making margin calls, forcing the funds to sell their

paper at a loss.

An LSS, by contrast, made much the same trade, but didn’t have any prime brokers

breathing down its neck. That’s because it borrowed the money to create its

leverage by issuing asset-backed commercial paper, or ABCP. Investors would

lend money at very short maturities – less than 90 days – against

the assets of the LSS. And those investors had two reasons to be sure that they

would get repaid in full. The first was that the assets of the LSS, being super-senior,

were therefore super-safe. (That one didn’t work out so well.) The second was

that the banks which created these structures, like Citigroup, promised that

they would step up and buy the ABCP if no one else would. That is the

famous liquidity put.

After the super-senior tranches of mortgage-backed securities started plunging

in value, the owners of those tranches found themselves having to roll over

their ABCP, repaying their original lenders and finding new lenders to fund

their positions. At that point, there were no takers for that kind of paper

at any price – there was no liquidity in the ABCP market. And so the liquidity

put was triggered, and Citigroup was forced to buy the ABCP itself.

Now the ABCP is asset-backed, so Citi could and presumably did take possession

of the super-senior paper which was held by the LSS vehicle. The original investors

in the LSS will have been wiped out, left with nothing. But the value of that

super-senior paper as now fallen so far that it’s worth much less than Citigroup

paid for the ABCP.

Let’s say that the value of a super-senior tranche falls from 100 to 65. The

tranche was originally bought by an LSS, where investors paid in 10 cents of

their own money and borrowed the other 90 cents by issuing ABCP. That ABCP eventually

gets rolled over to Citigroup, which also pays 90 cents for it. So the original

investors lose their 10 cents, but Citi ends up paying 90 cents for something

which is now worth only 65. In other words, the lender’s losses – 25 cents

– are significantly bigger than the losses of the people who bought the

riskiest equity tranches in the LSS, and who lost all their money.

Now I hasten to add that I’m pretty sure this is an oversimplified explanation

of what went on in reality. Alea has a more

nuanced and complex take on this whole trade, which involves not only credit

default swaps but even Canadians. And here’s how Charlie Calomiris

describes the LSS trade:

The CDO problem became magnified by the creation of additional layers of

securitisation involving the leveraging of the “super-senior”

tranches of CDOs (the AAA-rated tranches issued by CDO conduits). These so-called

leveraged super-senior conduits, or “LSS trades,” were financed

in the asset-backed commercial paper (ABCP) market. Some banks structured

securitisations that levered up their holdings of these super-senior tranches

of CDOs by more than 10 times, so that the ABCP issued by the LSS conduits

was based on underlying organiser equity of only one-tenth the amount of the

ABCP borrowings, with additional credit and liquidity enhancements offered

to assure ABCP holders and ratings agencies. When CDO super-senior tranches

turned out not to be of AAA quality, the leveraging of the CDOs multiplied

the consequences of the ratings error, which was a major concern to ABCP holders

of LSS conduits.

I strongly suspect that the "credit and liquidity enhancements" Calomiris

talks about here are exactly the same as the "liquidity puts" which

did in Citigroup. But it would certainly be nice if someone did some more reporting

on this.

Update: Alea points me to a

May piece from Pimco’s Edward Devlin, who explains the LSS and even provides

a helpful diagram. According to this (admittedly Canadian) structure, the liquidity

put was funding not the 90 cents at the base of the pyramid, but rather the

10 cents at the top! Which means that a 10-cent drop in the value of the super-senior

tranches would wipe out the provider of the liquidity put entirely. Meanwhile,

the equity investor only has upside, and makes no investment at all, so has

no real downside whatsoever. You can click on the image for a bigger version,

or just go to the Devlin article for more detail.

Jenny Anderson today runs down the list of winners

and losers in terms of subprime losses. Winners (or at least banks with

relatively small losses): Goldman Sachs; Credit Suisse; Lehman Brothers; JP

Morgan. Losers: UBS; Merrill Lynch; Citigroup. Anderson concludes:

Some of the Street’s safest institutions — or those that hoped

to be perceived as safe — turned out not to be, while some perceived

as risky are so far sailing through.

Anderson doesn’t go into a lot of detail about why this should be the case,

but for me it’s quite intuitive. The reason is that most of the pain has been

felt not by institutions taking on aggressive risk positions, but rather by

institutions who have seen the value of AAA-rated securities fall out from underneath

them. If an old-school banker has AAA-rated paper on his balance sheet at par

(or the risk of being forced to buy such paper at par), he knows that it might

fall in value to 99 or maybe even 97 cents on the dollar. It never even occurs

to him that it could be worth only 50 cents or less. A bet on AAA-rated paper

plunging in value and behaving more like equity than debt? That’s the kind of

bet engaged in only by houses with much more robust risk appetite.

From an anonymous fund manager, via

John Carney:

"B of A is the Stupid Bank. Citi is the Incompetent Bank. JP Morgan

is Villainous. The super SIV is Stupid, Incompetent and Villainous."

The FT’s Sam Jones puts

two and two together today and finally explains what the

notorious liquidity put is. You might recall

that a couple of weeks ago he told

us about CDOs which issued commercial paper. I noted at the time that given

the problems with CDOs, and given the problems with asset-backed commercial

paper, the intersection of the two – call it CDO-backed ABCP – would

have to be particularly toxic.

Well, it turns out that the snark was a boojum. Or, rather, the liquidity put

is that CDO-backed ABCP, and it’s making billions of dollars of Citigroup’s

capital softly and silently vanish away.

A liquidity put is really nothing more than a CP backstop. But let’s back up

for a second, here. A CDO is at heart just a collection of bonds, be they mortgage-backed

or otherwise. In order to buy those bonds, it needs to raise money. Historically,

it has raised that money from long-term investors who in return get the income

from the bonds which the CDO owns. But in 2005, Citi started to create CDOs

which raised their money not only from investors but also by issuing ABCP. Explains

Jones:

To mitigate any CP rollover risk, Citi entered into a series of “agreements”

which forced it to buy the CDO CP if no one else would. As Mr Rubin calls

them, “liquidity puts”.

The problem with these CDOs is exactly the same as the problem with SIVs (or,

for that matter, with banks in general): that the structure’s assets have a

much longer duration than its liabilities. If the CDO or the SIV can’t roll

over its ABCP, then the sponsoring bank is in trouble, since it invariably has

agreed to buy that CP if no one else will do so. Of course, there’s normally

a very good reason why no one else will buy that CP, and the bank ends up having

to take enormous mark-to-market losses on the CP it’s forced to buy at par.

And according to Jones this is not just a Citigroup problem.

It seems that Bank of America had similarly structured CDO deals in place.

In their 10Q, viewable here,

there’s an admission that:

The Corporation is obligated under the written put options to provide

funding to the CDOs by purchasing the commercial paper at predetermined

contractual yields…

And as we understand things, those obligations forced BofA to buy $12bn

of CDO CP from off balance sheet CDOs.

Stay tuned: I think this story is going to develop even further on Monday.