Wednesday, September 10, 2008

MBS: Marching into the detention area

While many expected the Treasury to eventually act on the GSEs, few (including myself) expected it to happen over the weekend. So what's the trade in fixed income? There have been four major ideas bandied about Wall Street over the weekend. They are: buy agency debentures, buy agency MBS (mortgage-backed securities), buy credit, and sell Treasuries.

The one I like the most is buy MBS. Yesterday MBS spreads moved dramatically tighter, 55bps in OAS (option-adjusted spread) according to the Lehman index, from +147 to +94. But you haven't missed it yet. For most of the last 10-years, the index OAS has been between 30 and 60bps, so there is plenty of room to tighter further in OAS. I've always felt as though OAS was an over-rated value metric, but today its especially questionable. The Treasury has announced their intentions to buy MBS in the open market, with the clear goal of pushing mortgage borrowing rates lower. Ideally the Treasury would like to set off a refi wave, which would help banks "naturally" delever as well as help separate good loans from bad. In order to get most 2006-2007 borrowers "in the money", mortgage rates probably have to fall to around 5.25%. I think this implies another 50bps of MBS tightening.

Buying agency debentures is less intriguing. Right now non-callable agencies are trading between 50 and 60bps more than comparable Treasury bonds. This spread might fall into the 20's for short-term bonds, but it won't collapse to zero. We don't currently know what the GSEs will look like after 2009, and therefore bonds maturing beyond 2-years shouldn't (and won't) be viewed as truly government-guaranteed.

Credit is highly questionable here. The large-cap financials should benefit significantly from a mortgage refi-wave. This would get the "good" loans off bank's balance sheets, freeing up capital, and clarifying how much each bank truly has in good versus bad loans. But the real catalyst for finance credit will be bank earnings which isn't until next month. Obviously the Lehman situation isn't helping either. Buying new issue bank credits isn't a bad idea as a trade but I'm staying underweight for now.

Non-finance credit makes even less sense. The GSE bailout may have been necessary, but it isn't a panacea for the real economy problems we're facing. I'd stay very high quality within credit.

The direction of Treasury rates is also questionable. There remains significant dollar-related buying, as evidenced by the strong bid for the 10 and 30-year bonds in recent sessions. That is a classic sign of foreign bank buyers, especially given the fact that economically, the yield curve should probably be steeper not flatter. There is also no particular reason to believe the GSE bailout results in dramatically more Treasury supply. Not to mention the simple fact that the economy remains weak which should be a natural support for interest rates. I'm staying close to home on duration.


Sergei said...

A large proportion of 2006-2007 mortgages are probably at least 10% below the property value. Can homeowners refinance their mortgage if the mortgage balance is higher than the property value?

Also, if banks can delever through the refi-wave, where would the mortgage balance end up other than on GSE balance sheets or MBS? Is the conservator willing to allow the GSE to dramatically increase their balance sheet?

Accrued Interest said...

Its obvious that many, probably most, 2006-2007 home buyers are under water and can't refinance. But there is still a large number that can and will. Forget about "solving" the housing crisis and think more in terms of "is this a positive or a negative?"

The Treasury is allowing the GSEs to expand their balance sheets through 2009. But also you could have stronger banks taking some of the new mortgages. I mean, mortgage underwriting in 2008 is probably going to be pretty profitable for the few with the balance sheet to participate.

"Mindles H. Dreck" said...

I suspect the rate of foreclosure will slow, which is about the only positive I see for credit issues outside the agency world. Refis remain possible only for the good loans, as you say.

The fact that the govt. had to put $100B on the line in each case shows how much it takes to drown the shorts. Thinking about, for instance, WaMu, TPG doesn't have the scratch to do it.

PNL4LYFE said...

I think Treasury had to make the number sufficiently large that the market would view it as an ironclad backstop. Between the equity and preferreds that still have to be wiped out and the treasury commitment, it's around 300bn total. I haven't heard any knowlegdeable person suggest that the eventual losses will be higher than that.

Accrued Interest said...

It reads to me like the Treasury is going to put in cash $1 billion at a time as needed. Personally I'd be surprised if it takes more than $20 billion or so. Well, maybe becuase of M2M stuff they need to inject more than that. But in terms of losses? It won't be that large. As PNL said, there is a lot of other equity to take losses before it gets to tax payers.

Accrued Interest said...

Man WaMu looks like toast to me. Which has an easier time finding a buyer? Lehman or WaMu?

PNL4LYFE said...

I think WM is probably a more valuable business than LEH (especially if LEH sells Neuberger). But I would be more comfortable holding LEH securities of any flavor. LEH has the same counterparty links as BSC which should compel the fed/treasury to craft a deal where business can continue as usual and there are no counterparty defaults. In this scenario, the buyer probably pays a low price for the common and the preferred is not impaired. I don't see how a buyer could be senior to the preferred as in FNM/FRE bailout.

On the other hand, WM does not have the same counterparty issues. The FDIC provides a path for an orderly liquidation of a traditional bank. I'm not that familiar with the process, but I think a potential buyer could wait for an FDIC takeover and buy at a lower price. In this case, common and preferred would be wiped out and holdco debt seriously impaired.

Sergei said...

I am not sure that mortgage underwriting would be profitable in 2008, considering that property prices could well fall another 10%-15%. They could be profitable if banks demand and get large downpayments, but not many borrowers can put up large downpayments when buying property. If mortgage rate gets artificially pushed down, that leaves even lower profit margins for the underwriters to cover credit costs.

Accrued Interest said...


I 100% agree that the universe of potential mortgage borrowers is small indeed. But for those that do put up the cash and do buy a new home, I think those mortgages will be pretty profitable. Just because only the best mortgages will be made, and therefore the 2008 vintage is probably going to be a very low default vintage. Especially if one excludes any quasi-mod refis.