Sub-Prime! An Idea for a Master's Thesis

I have developed a hypothesis regarding sub-prime that I will be researching more carefully over the next month or so, with the idea of writing an article about it. I don’t think there’s a full book in the idea – but I can quite easily see it becoming somebody’s master’s thesis, or a chapter in a book.

Hypothesis: A sub-prime-crisis-equivalent has almost happened before. It has also been deliberately avoided before

Abstract: So far, the nasty bits of the crisis have not had much to do with sub-prime mortgages themselves, or even their RMBS (Residential Mortgage Backed Security) pools, but with highly leveraged pools of RMBS – see, for example, the sad stories of Cairn High Grade Funding I and Global DIGIT. These vehicles failed not through any problems with the RMBS themselves, but through the fear of such problems and the subsequent evaporation of a market for ABCP (Asset Backed Commercial Paper) with such underlying security.

This may be compared with the market on Bank Perpetuals in the ’80’s. The basic idea behind the perps was for a bank to offer a 100-year Floating Rate Note at a spread to the appropriate floating rate. At the time (I confess, I’m not sure about right now) such issues could be considered equity by the banks; simultaneously portfolio managers could consider them short term instruments due to their floating rate coupon. At least one such issue made its way into a money market fund.

I wasn’t a market participant at the time these things were sexy, which was somewhere around the 1983-88 period when rates were coming down quickly and banks wanted long money but didn’t want to pay long rates for a long time. But very scrappy recollections involve the idea that the problem with these things was that the market was too homogeneous … all the issuers were all issuing them for the same reasons; all the buyers were all buying them for the same reasons. When the reason for the buyers went away, so did the market.

Nowadays, for instance, I’m seeing a chunk of RBC FRN Oct 1, 2083, which pays CDOR + 40bp  being offered at $94.50. I suspect that a buyer willing to take the whole block could put in a stink bid and be filled, but what do I know? Anyway, a salesman whom I respect (at a firm other than the one stuck with these turkeys) had to be prodded to remember the existence of such things. It would appear that they are the by-appointment-only traders to end all by-appointment-only traders.

Now, in the ’80’s when they were sexy, rates were still pretty high. What if they’d been low (like they have been, relatively, for the past five years)? Would it not have been a tempting idea to put together a vehicle that, with $100 equity you went out and bought $1,000 of these perps and financed with $900 commercial paper at, I don’t know, CDOR + 10? That’s a pretty good return, provided you can roll your paper for 100 years!

There are two reasons I can think of right away that this didn’t happen:

  • There was not so much innovation in the markets then. Hell, at that point the junk bond market hadn’t even been invented.
  • There was lots and lots of government issued short term paper at the time, which (to an extent) was crowding out more innovative issues

Food for thought.

The part where I’m thinking this was deliberately avoided before is the bond futures market. Bond futures are not the easiest things in the world to analyze quantitatively. The big problem is that there’s a basket of deliverables and a cheapest to deliver. The Cheapest to Deliver bond can change since the bond delivered against a contract is at the seller’s option, which gives the contract negative convexity. There are also delivery options that confuse the issue.

It is my understanding that this complexity was introduced deliberately when the contract was being designed. The CBOT wanted complexity so that it would be hard to analyze so that the fair price would vary by a bit depending on what assumptions you plugged into your model. This ensured that there would be disagreement over what the thing was worth depending on your views, which in turn ensured that there would be an actual market with some depth.

After all, if something’s known to be worth $110.87 and everybody knows that, who’s gonna trade? It will be quoted at $110.86-88, zero volume, forever.

The complexity helped develop a heterogeneous market which has been quite successful, to say the least.

The evaporation of the ABCP market has a tipping-monkey (or is it “100th monkey”? You know, where there’s a very rapid change of state of a large system once you reach the critical point) feel to it.

I’m thinking it would be most interesting to compare ABCP with the above two markets – especially the Bank Perps market – with a view to isolating the similarities and seeing if any conclusions may be drawn relating market risk and market homogeniety … and who knows, maybe doing a little forecasting!

Leave a Reply

You must be logged in to post a comment.