Another sign of market panic: even though Fannie Mae and Freddie Mac are now officially wards of the state and the Treasury has assured that they will not fall into a negative equity standing, the general credit market stress and flight to quality means that their mortgage backed bonds are trading at elevated spreads. However, mortgage rates are lower than in July due to the rally in Treasuries.
From Bloomberg:
Yields on Fannie Mae and Freddie Mac mortgage bonds jumped relative to Treasuries for a second day, pushing spreads above their levels before the U.S. took over the companies and vowed to support the market.
The difference between yields on Washington-based Fannie’s current-coupon 30-year fixed-rate securities and 10-year Treasuries rose 12 basis points to about 204 basis points as of 3:05 p.m. in New York, according to data compiled by Bloomberg. The rise in yields on the bonds this week suggests an increase in interest rates on new home loans of about 50 basis points, 0.50 percentage point….
Mortgage-bond spreads also climbed as expectations for interest-rate volatility, as measured by swaption prices, rose yesterday to the highest since 2003, according to a Lehman Brothers Holdings Inc. index. Swaptions give the buyers the option to enter into swap contracts later. A swap offers a fixed yield in return for floating payments linked to short-term bank borrowing costs….
The yield on Fannie’s current-coupon mortgage bonds rose to 5.91 percent from 5.71 percent yesterday, suggesting mortgage rates are increasing. The yield is up from 5.63 percent on Sept. 5 and 5.27 percent on Oct. 6.
The average rate on a typical 30-year fixed-rate mortgage rose to 5.97 percent yesterday, compared with as low as 5.72 percent last month and a six-year high of 6.51 percent in July.
26 comments
Comments are closed.
How is this even possible? Where is the F/F vs the Treasury Dept. arbitrage? These bonds are AS GOOD as the similar duration government debt.
(sarcasm off)
The yield spread reflects prepayment risk not credit risk.
Increase in volatility increases the value of the prepayment option and increases the spread to eqv durationTsy.
Anon of 5:29 PM,
That is very helpful, but also one of those cases in which derivatives logic (higher implied vol means higher priced option) does not play out in the real world.
It takes time, cost, and effort to refinance. In normal environments, consumers refinance when they see a meaningful fall in their mortgage rate. Sure, rates have been noisy, but there hasn’t been enough consistent downward movement to produce much (amy?) refinancing. And in this declining employment/credit score/housing price environment, the practical barriers to refinancing are high.
Yves,
Apologize if I wasn’t being clear. The spread reflects the fact that duration of the mortgage can change while the duration of the Tsy is fixed (making hedging a nightmare).
Prepayment doesn’t only mean re-financing and a shorter duration.
The likelihood is that the duration of the mortgages can extend to that of a 20yr or 30yr Tsy as homeowners only make their required payments.
The greater interest rate volatility, the greater the value of these embedded options.
Anon of 5:29 PM,
the article says “The rise in yields on the bonds this week suggests an increase in interest rates on new home loans of about 50 basis points, 0.50 percentage point….”
They’re talking about new loans.
Nope, consistently high spreads in pretty much all markets – from interbank lending to mortgages point – point to credit risks, not prepayment.
I think the risk that an investor could be trapped in a GSE bond by a surge in inflation is probably most of the risk. That’s particularly painful, as precisely when being in long bonds is most costly, you get the longest duration as people sit on their low-payment mortgages. Prepayment risk is probably not such a big deal with current interest rates so low. Having some of your money dumped out in 2 years is not much risk, as you probably won’t get prices much below current treasuries ever.
It’s me again, Toto.
“These bonds are AS GOOD as the similar duration government debt.”
Yes indeedy. Just AS GOOD.
“the practical barriers to refinancing are high.”
Some other practical barriers are rising, too.
Catch the “mortgage backed security” news that came out of Chicago?
The Cook County Sheriff announced he’s going to cease enforcing foreclosure evictions.
Golly. That could start a trend. Sheriffs won’t evict, local judges “freeze” the courts in procedural delays and people simply stop paying.
What an idea. Instead of mailing in their keys they mail in your payment books.
What’ll stop them? The shining example of moral probity, honesty and patriotic self-sacrifice set by lower Manhattan?
Hi Yves and Anon
The predominant form of prepayment of a home loan takes place on sale. Refi and true prepayment are secondary.
If the economy stinks, look for folks to sell their MacMansions and move to a desert box in Phoenix.
8:34 — Sales at foreclosure auction count as prepayments, do they not? I think I recall that from somewhere in the mists.
The judges stopping foreclosure proceedings, or at least delaying them, is a good thing.
People, they always get in the way. There are still people out there, probably even more than there were yesterday.
Interest rate risk is something that got lost over the last few years. Remember when bond yields were inverted? In actuality, maybe they still are. I think at this point you can throw out t-bills as a measurement of anything beside panic. What is the real 3 month yield? Probably somewhere between 2 and 6 (apy) percent, that is the best analysis I can come up with, any one have anything more substantial to add? Clearly trying to measure short term rates now is very difficult, but how could you begin?
Good think we wiped out GSE shareholders to get here.
Yves
I’ve been informally studying the “credit default swaps” market for some time. It is said that there are $58 trillion of these outstanding. Though some are off setting many may be built on leverage and unreserved for by the underwriter.
What will happen as a result of the Lehman auction today and similar events is that those solid institutions that underwrote CDS contracts which they than turned around and hedged with an off-setting CDO will still be obligated contractually for the CDO they underwrote while losing their protection to a bankrupt counter-party. They will than have unhedged exposure and may be taken down by the same bankruptcy or another depending on whom they underwrote. A few major bankruptcies will take down the whole $58 trillion edifice like a “house of cards.”
The only way to resolve this is for the U.S. government to declare “force majeure” and annul these contracts. Parties will be relatively less scathed financially as they will lose only the premium flow and un-callable protection where they have underlying securities to protect.
An Economist
anon 10:26 (economist)
Say I’m a hedge fund that bought protection on Morgan Stanley 3 years ago from Goldman Sachs. I’ve dutifully paid my premiums. I don’t own any Morgan Stanley paper.
My CDS has a theoretical value of $6,000,000. I’m about as senior as they come to recover by $6 Milion if Goldman goes belly up (granted Uncle Hank can change the rules).
Are you saying that it is okay for the govt. to take my $6MM and, in effect, give to the shareholders of Goldman Sachs because I made the right call and because the present admin is otherwise incomptent?
Toto, please read more on the Chicago Cook County story – the reason he refuses to evict is because oftentimes the evicted are honest renters who live by the rules and are current on their rent. They have no control over landlord’s mortgage situation.
The predominant form of prepayment of a home loan takes place on sale. Refi and true prepayment are secondary.
If the economy stinks, look for folks to sell their MacMansions and move to a desert box in Phoenix.
This is just a reiteration of the “credit risk” – bonds will have to take a hit at the moment of sale because of lower RE prices.
Yves said,
“even though Fannie Mae and Freddie Mac are now officially wards of the state and the Treasury has assured that they will not fall into a negative equity standing”
I think you are misunderestimating (to coin a phrase) the value of semantics here. Assurances are not guarantees. Given the certain insolvency of FNE/FRE without government support, any withdrawal of such support, or more likely a reduction of such support by requiring haircuts of all stakeholders would send all GSE debt quickly below par. I believe the spread between treasuries and GSE debt is a quite rational discounting of such a possibility.
Just checked, the spread between FNM and FRE backed MBS is consistent with that of GNMA backed MBS.
Prior to the Hankilization of FNM/FRE, their MBS traded at spread premium to GNMA.
You can view all these MBS as amortizing callable/extendible Treasuries.
The spread just reflects the uncertainty as to the timing of return of principal (and total interest) not uncertainty as to return of principal.
Anon of 10:26 PM,
We are well aware of counterparty risk and were writing about the risks posed by CDS well before they were mainstream. See this selection of many examples “Is Systemic Risk Underestimated” (and note the day, May 2007, before the credit contraction began), “Fitch Points to Credit Derivatives as Possible Accelerant in Credit Downturn” (July 2007), “Are Credit Default Swaps Next?” (August 2007) and “Counterparty Risk Problems With Credit Default Swaps?” (November 2008).
As for the Federal government voiding these contracts, it has no standing. CDS, as contracts, are governed by state law (even if they were deemed to be insurance contracts, they’d still be governed by state law). The Federal government has no basis for interfering in state law matters. A move along the lines you suggest raises serious Constitutional issues and would be challenged, particularly given the number of deep pockets who would be adversely affected.
Anon of 10:26 PM,
There are also practical reasons why the move you suggest would cause simply colossal dislocations. To wit: some users bought CDS to hedge bonds or exposures they have. If you take away the guaranteee, they have to mark down the position. That swap used as a hedge in turn was in many cases hedged by the protection writer with a party that was simply punting, so you cannot simply preserve the CDS related to hedges of the underlying. The product is too enmeshed with the entire financial system to be able to isolate hedgers.
Similarly, AIG wrote CDS that enabled many Europen banks to evade statutory capital requirements. Void those swaps and the banks are even more undercapitalized (there are reports that the ECB applied considerable pressure to Washington to make sure they bailed out AIG). Many structured products relied on CDS for their credit enhancement. Take that away, and the structured credits are worth considerably less, distributing losses over the financial system (particularly banks, who are substantial holders of structured credit paper).
Yves, I used to manage mortgage bonds. Typically I compare yields over swap rates. That’s where the banks ordinarily fund. It’s still a record spread there, but most of it due to an increase in swaption volatility, which increases hedging costs.
yves,
three thoughts.
1. while the mortgage-OAS spread is wide, as merkel pointed out, realized volatility is through the roof, which explains most of spread widening. basically, there is so much uncertainty about the future path of interest rates (inflation -> up, deflation -> down) that the market is requiring a significant yield premium to the risk free rate to assume the negative convexity associated with a pass through (FNM) or participation certificate (FRE).
2. in theory, the GSE's were supposed to help compress agency spreads by ramping up their portfolio activities (where they would arbitrage the price difference between MBS and agency debt). however, it appears that on this front, things are not going according to plan.
the crux of the issue is that agency buyers are not willing to buy longer dated agency debentures in size, despite the nationalization. the reason is that the Treasury's plan involves the winding down of the retained portfolios.
according to the Paulson plan, 2 years from now – when the crisis is over – the retained portfolios are going to be run down to $200B over a 10 year period. the principal buyers of agencies are foreign central banks and money center banks who use agencies as a way to invest surplus liquidity. they need this market to be deep and highly liquid to continue their participation.
if the GSE retained portfolios start shrinking in 2 years, the agency market will shrink and become highly illiquid. no central bank wants to own a 10 yr agency bond that is going to turn into an off the run, illiquid security in 3 years time.
so right now, the agencies are having a lot of difficulty funding farther out the curve. most of their issuance is of the overnight and 3/6/9 month maturity.
this is creating problems for treasury. they would like to run a duration-neutral portfolio, just as the companies did when they were private. however, the agency market is not going to allow them to. so their only options are to shrink the portfolio (uh oh) or to allow the duration gap to widen dramatically.
my understanding is that the plan for now is to go ahead and lever up the balance sheets by funding the $150B of incremental purchases via short-term funding. this is a terrible idea, given the massive current account and budget deficits the US government is running.
the irony is that paulson & lockhart are going to put on the massive carry trade that the administration always falsely accused the GSE's of running.
3. a similar chain of reasoning also applies to the core MBS business. agency MBS have contractual maturities of 30 years. actual durations right now are in the 4-6 year range. if interest rates back up dramatically, those durations could extend to 7-10 years.
the paulson plan goes out of its way to emphasize to the market that the fundamental role of the GSE's in the mortgage market is going to change in 2010 and beyond. it's no secret to anyone who covers the mortgage industry that the goal of treasury is to move the US to the covered bond model used in europe.
i'm not going to opine here on whether or not this is a good or bad idea in the abstract. however, if i'm a mortgage buyer, why on earth do i want to buy an agency MBS yielding 5.5 – 6% if there's a chance that in 5 years liquidity in the agency MBS market is drying up because no new MBS or PC's are being issued?
i need to be paid extraordinary rates of return in order to be compensated for the illiquidity that i know is coming in a few years time.
——-
the sad thing is that i'm not sure treasury understands any of these things. philip swagel, one of the intellectual architects of the GSE nationalization stood up at an IMF panel yesterday and proclaimed that the GSE nationalization was a resounding success because the "conforming mortgage market is working fine".
i suppose that, taken in isolation, that statement is factually correct. of course:
– mortgage rates are now higher than they were relative to swap spreads
– a significant number of community banks (and farmer mac) who held GSE preferreds have been wiped out.
– the GSE action triggered a run on lehman. this ended up taking down AIG, WaMu and Wachovia (all the bank executives have told the street that the deposit runs started on Sep 15, the day Treasury let Lehman fail)
– the market for bank preferred stock and hybrid debt is shut permanently
– the equity capital markets are closed to most financial institutions because most investors have concluded that this group of regulators (treasury, fed & FDIC) are bad partners.
– the icelandic banks have all been nationalized and the country is on the brink of failure.
– the high yield and investment grade bond markets have COLLAPSED.
– working capital is being cut off to corporates around the world.
– the global equity markets are in the midst of a historic crash.
some success.
the treasury completely mis-diagnosed the scale, scope and extent of the de-leveraging that was going to take place. it also had absolutely no idea what it would unleash on the global economy when it decided to play investment banker and start re-organizing the US financial system.
Yves, Anon,
Thanks for your comments on my CDS post re: “force majeure.” Though I am not at all a Constitutional lawyer or even very conversant in contract law issues, the President may have the authority under something called “International Emergency Economic Powers Act.”
Whether he or who in the case of the States – – where were these underwritten? – – has the jurisdiction I have no idea.
I was not convinced by your arguments that cascading defaults are not in the cards, so to speak. If as in the example you gave Goldman Sachs pays you for a Morgan Stanley default but is not in turn paid for an off-setting contract it bought from a smaller house which was not adequately reserved, there goes Goldman! You may or may not be paid in that case even if the instrument is senior which I am not sure of and the way these instruments were underwritten will probably be litigated until your money is worthless.
I am not in favor of annulling contracts, taking your money or anything of the sort for that matter. Unfortunately trillions have just been taken in the stock market and through poor governance;, that’s the nature of joint venture, limited liability capitalism. My point was that you may not be paid anyway if your underwriter is taken under so it would be best to break the chain before it started. I hope I am wrong actually and that there are not massive cascading defaults but I don’t believe so. Once the chain starts to go, the sooner this is done probably the better.
An Economist
An Economist,
I am troubled that I do the respect of replying to your comment at some length, provide links to earlier work, and you do not extend a similar courtesy by reading, or at least skimming them.
I have repeatedly, long before it was popular, have written of the risks of cascading counterparty defaults in the CDS market, Indeed, I have even said it represented the biggest single risk to the financial system. Contrary to everything I have written, you contend that I have said cascading defaults are “not in the cards”. The risks may by happenstance be distributed so it does not happen, or few enough big players may wind up being impaired that they can be backstopped, but I have NEVER minimized the risk of CDS counterparty failures. Indeed, I have also pointed to the risk of mere CDS operational failures, which were also widely discounted.
To say the ripping up CDS contracts, which is what you advocate, will also cause colossal disruption is a completely different statement.
What about the non-callable bonds? For example, it looks like the 2.875% note from Fannie Mae maturing on 10/12/2010 is non-callable but still has a huge spread to treasuries. Since it’s non-callable, what would explain that spread?
Have to think the spreads will come down to earth. Surely part of the “bailout” must focus on getting the housing market back on track, meaning that the government must somehow counteract the massive borrowing necessary for the bailout itself. How? I don’t even want to speculate, nor does it appear that any specific government action will calm fears in the short term.