Earlier this week, FT Alphaville brought you a structured finance index primer. Now that everyone’s up-to-speed, let’s take a look at the latest price action, look at some (gasp!) fundamentals, and stir the debate on subprime versus prime — a party that we are admittedly fashionably late to.
As the above chart shows, the PrimeX indices have been experiencing something of a deterioration lately. This is potentially misleading, however, given that the indices are relatively illiquid. In terms of total amounts outstanding, the ABX indices (untranched) have $19.8bn net notional across 3,038 trades versus the $3.8bn net across 718 trades for the PrimeX, according to DTCC data that’s current as of last Friday.
The amount of trading activity for all these indices can vary widely week by week. Over the last month, here are how many trades were either put on (or terminated) per week, starting with the most recent week for which data are available (last week): 54, 28, 57, 9. There you have it, one month of trading activity = 148 trades. The PrimeX.FRM.2, which can be seen tanking in the above, had just one trade last week.
But anyway, we’ll stop teasing you and compare PrimeX to those lovable ABX subprime indices. Here goes:
Here you can see that the ABX indices are substantially below par – something that will surprise no one. It also can be seen that the more liquid AAAs have been trending downwards; one of the AA’s had a particularly exciting downward move. Before anyone gets excited about that ABX.HE.AA.06-2 chart though, beware that it didn’t even have a single trade last week and it only has 92 trades outstanding. Chances are that some trades were done this week, though, and everyone updated their marks to reflect that trade (or trades). Good old-fashioned price discovery in action.
Moving on, there are a number of possible PrimeX explanations. Such as:
(a) PrimeX is finally pricing in fundamentals correctly, and was deluded before.
(b) Negative media coverage influenced sentiment, shifting prices. In which case, the move may be temporary.
(c) Some longs happened to back out of the trade now for some idiosyncratic reason and/or in order to de-lever in the face of the sovereign crisis and global slowdown.
To investigate a bit further, FT Alphaville spent ages hogging the Bloomberg terminal in the FT Newsroom (sorry…) looked up some information on the mortgage pools that the most liquid PrimeX references – the ARM.1 that recently dipped below par (see above chart).
Of the 20 mortgage pools referenced, we selected the four that currently have the worst performance statistics per metric examined. After all, it wouldn’t be in keeping with the current mood to look on the bright side of life. Here goes:
The percentage of mortgages in the pools that are currently 60 days delinquent is quite volatile and generally hasn’t moved much above two per cent. Again, a reminder that these will be just the four out of 20 possible pools that performed the worst in September. We thought 20 lines on a graph would make Excel 2003 cry.
The percentage of the mortgage pool that have had foreclosures has been increasing, but levelling off somewhat (again for the worst performers).
The cumulative losses experienced in the pools has been steadily increasing over time.
Right. So we ask again: why the downward movements now? Why not before? Do we know that much more than a month or so ago? Barclays doesn’t think so:
• Fitch downgraded a large percentage of bonds in the prime space, including several bonds in PrimeX recently.
• However, concerns about these downgrades are overblown, in our view, as almost all of the securities had previously been downgraded by Moody’s/S&P.
Ouch. If we were Fitch, our feelings would be hurt. Like, “what do you mean our ratings actions don’t move the market?” Here’s a snipped from Fitch’s report, while we’re here (Emphasis ours):
The negative rating actions reflect the continued stress in the sector driven by the ongoing correction in the housing market and challenging macroeconomic conditions. The rating actions also reflect the application of the agency’s new prime RMBS loan-level loss model.
So part of the Fitch downgrades were down to fundamentals and part due to methodological changes.
Back to Barclays for a bit more investment insight though, here on the “technicals risk”:
• This is potentially more credible than other risks, especially if investors used a lot of leverage and are forced to unwind PrimeX longs as it goes down.
• However, we believe that at current levels, investors are compensated to take this near-term mark-to-market risk.
In other words, if the longs are highly leveraged, relatively small movements could cause them to unwind, and as everyone rushes for the door at once, the price to close out by going short could make the trade unprofitable overall. And the “other risks” cited are: credit risk, prepay risk, and the earlier quote covered downgrade risk.
PrimeX may well prove an interesting one to watch. Perhaps for the fundamentals. But almost certainly for the technicals.
Related Links:
Mortgage Credit in ‘Quiet Crash’: DoubleLine’s Gundlach – CNBC
Fitch Completes Review of U.S. Prime RMBS Ratings Using New Model – WSJ MarketWatch