A February 3, 2012 article in Businessinsider (link is here) alerted us to a piece of new research from Fitch on the repo market. The Fitch report (link is here) notes that the share of structured finance instruments being financed in the repo market is at near pre-financial crisis levels, tracking at nearly 20% of total repo collateral. Structured finance assets include, among other things, Collateralized Debt Obligations (CDOs) and Residential Mortgage-Banked Securities (RMBS). Within the RMBS pool are Alt-A and Sub-Prime mortgages which now account for 42% of the the structured finance assets being financed.
Fitch looked at Money Market funds tri-party repo collateral holdings over time, with the last data as of August 2011. Their sample was $90 bio of financing trades or just over 5% of the tri-party trades outstanding. We do wonder how appropriate it is to extrapolate out to the rest of the market from the 5% or, put another way, how idiosyncratic the large Money Market funds tri-party investing strategy is relative to the rest of the market. But we hope some data is better than no data.
This brings up many questions and more than a couple flashbacks. The repo market nearly shut down in 2008 over these kinds of assets. Alt-A and Sub-Prime MBS were lumped in with asset-backed securities on (at least one that we know of) tri-party agent collateral schedules. There was no granularity for ABS and a simple “check the box” approach included anything that could be considered an ABS. So cash providers who had, back in the day, said “yes” to more benign forms of ABS like auto and credit card securitizations could also be given Alt-A and Sub-Prime paper as collateral. We hope the schedules have been made more granular and cash providers a bit wiser about their collateral.
Since haircuts on tri-party trades are hard wired in agreements & systems, traders can’t renegotiate haircuts at the drop of a hat (like bilateral deals can). In 2008 when life became uncertain, the market simply shut down. Researchers have likened it to an e-coli breakout. You don’t know which vegetable is infected or how badly, so you avoid veggies altogether. They have also called it the repo version of a bank run. We wrote about this a couple days ago in a post where we highlighted a recent paper by Yale economists Gorton and Ordoñez (“Collateral Crises”) and their discussion of “information insensitive” securities. These were securities which had high costs to get information for, so investors relied on cheap information like ratings. We know how that worked out. We urge you to go back and read the post and the paper. Finadium wrote a research piece last year focused on tri-party and looked at many of those issues. A link to the synopsis is here.
In 2008, hedge funds and the like were doing non-recourse term financing deals in the MBS paper. When there was recourse, there wasn’t much capital to back up the trades, making them non-recourse for all intensive purposes. It seemed like a good idea at the time: the paper was AAA and traded with tight bid-ask spreads. That was, until it didn’t. Once the market broke, it wasn’t unusual to see 10 or 20 point differences on the marks from one pricing service to another on the exact same paper. The pricing differences exceeded the haircuts on the repos, often by a multiple. Executing a liquidation was done on a wing and a prayer. The question remains on how these new deals are being marked and the nature of the structured repos creating the underlying financing. Fitch says that median haircuts on structured finance paper from their sample was 5%, but that is what the broker/dealers are providing to fund themselves in tri-party, not what the underlying beneficial owners are paying. Hopefully, non-recourse deals on paper with this kind of risk (in the CDS market they call it “jump to default”, maybe we can acknowledge that repo is by its nature wrong-way risk and call it “jump to correlation”?) are no longer being done and a robust mark-to-market process is in place. If history is a guide, the cash borrowers need deep enough pockets to survive some pretty big potential margin calls.
Cash lenders using tri-party should check their schedules and really understand their collateral. The tri-party reforms will make it easier to understand what kind of paper is populating their “shells” (the term for the approved collateral types) and price differences on the marks. But if cash lenders don’t analyze and manage the information in the reports, well, lets just say “caveat emptor”.
It should be said that sub-prime paper didn’t work out all that badly. The Gorton and Ordoñez paper reminded us that “the realized principal losses on the $1.9 trillion AAA/Aaa subprime bonds issued between 2004 and 2007 to be 17 basis points as of February, 2011” and “the Financial Crisis Inquiry Commission (FCIC) Report (2011) noted that with respect to subprime mortgages: ‘Overall, for 2005 to 2007 vintage tranches of mortgage-backed securities originally rated triple-A, despite the mass downgrades, only about 10% of Alt-A and 4% of subprime securities had been ’materially impaired’-meaning that losses were imminent or had already been suffered-by the end of 2009’…” Just like the e-coli, panic about the unknown can get exaggerated. Perhaps the paper is no longer as informational challenged as it once was? Time will tell.