Saturday, March 21, 2009

The True State Of The CMBS Market, And Why Billions In New Writedowns Are Coming

In response for requests for information on where the capital markets objectively evaluate commercial mortgage backed securities, and also to demonstrate the recent knee jerk reaction of how CMBX spreads ripped wider once it became clear older vintage, sub-AAA would not be eligible for TALF 1.0 participation, I am presenting the recent trading levels of CMBX 1 through CMBX 5, segregated by tranching (rule of thumb: the higher the chart line, the lower the underlying value, the more MTM pain for sellers of the CMBX tranche i.e. banks). The mid-to-late February explosion was a result of the market realizing these securities would not be eligible for taxpayer support in the TALF 1.0 version, and thus indicative of the true, and very sad, state of the commercial mortgage backed real estate market (some good intro material on CMBX here).

The annihilation in CMBX (for lack of a better word), explains the recent urgency behind the Treasury's moves to provide an iteration of TALF that will return some sense of normalcy to the CRE securitization market. The majority of CMBX tranches trade at levels which imply vast losses at low recovery values on the underlying loans.

As there are hundreds of billions in underlying notional behind the various 1 thru 5 vintages, the mark to market pain experienced by major banks and financial institutions (who would sell CMBX risk to willing purchasers) in the recent widening sprint is likely to generate another round of massive write-downs at all TARP recipients.

Also, for an indication of just how bad bank writedowns will be this quarter, a good proxy is the average levels of the various CMBX indices at Dec 31 and where they are trending now. If the current price levels persists, it will be a bloodbath.






hat tip to JP Morgan for chart data. Sphere: Related Content
Print this post


Anonymous said...

What about the proposed MTM changes to be implemented since the start of the quarter?? That would make the banks look better and not take so severe write downs.

Tyler Durden said...

unless banks manage to squeeze cmbx into a held to maturity asset category likely will not impact their mtm. But it explains the rush to adjust mtm ahead of march 31. Regardless even if they mark everything to par this should demonstrate the huge disconnect between where the mkt values cmbs (all time lows) and the banks. Assuming mkt is correct in its loss recovery estimates even if marked at par once defaults skyrocket,banks will be forces to take huge loss provisions...unless that accouting rule is also adjusted

Anonymous said...

exactly, then are u trading on this? I decided to buy financials near their low (FAS) for the run up their stocks would have once this perceived fix is absorbed by the public. I believe they will rise from their low to a more reasonable valuation once this has been dealt with.

Bicycle Repairman said...

Suspending MTM is not the answer for two reasons:

1) The collateral is distressed. MTM suspension advocates hold the naive view that the collateral is fundamentally sound, but market prices are low due to fear.

2) Imagine the trouble a bank could get itself into if it did not have to acknowlegde market prices. The street wants more transparency, not less.

Not until bas assets are removed frm bank balance sheets and investors can make educated assessments of bank health will they invest in and recapitalize the banks.

Anonymous said...

how big is this market compared to RMBS?

OhmeOhMy said...

"hat tip to JP Morgan for chart data"

'nuff said!

Anonymous said...

Nobel economist Paul Krugman saying their plan to prop up asset prices "isn't going to fly". He also said:

"At every stage, Geithner et al have made it clear that they still have faith in the people who created the financial crisis — that they believe that all we have is a liquidity crisis that can be undone with a bit of financial engineering, that “governments do a bad job of running banks” (as opposed, presumably, to the wonderful job the private bankers have done), that financial bailouts and guarantees should come with no strings attached. This was bad analysis, bad policy, and terrible politics. This administration, elected on the promise of change, has already managed, in an astonishingly short time, to create the impression that it’s owned by the wheeler-dealers."

This just in, Geithner appoints Shamwow pitchman to sell toxic assets to investors.

Said: "We needed a pitchman with Sham experience".

Anonymous said...

We Not Only Have a Shadow Banking System, But also a Shadow Government,

When one considers the comparatively extensive system of congressional checks and balances that goes into the spending of every dollar in the budget via the normal appropriations process, what's happening in the Fed amounts to something truly revolutionary —
a kind of shadow government with a budget many times the size of the normal federal outlay, administered dictatorially by one man, Fed chairman Ben Bernanke. "We spend hours and hours and hours arguing over $10 million amendments on the floor of the Senate, but there has been no discussion about who has been receiving this $3 trillion," says Sen. Bernie Sanders. "It is beyond comprehension." ...


Anonymous said...


While I agree there is little doubt that the loosened underwriting standards in the commercial mortgage market over the last few years will contribute to higher default and loss rates for those recent vintages, I also conclude that the cumulative ten-year loss rate for the entire CMBS conduit market is a manageable 2.53 percent for our baseline scenario.

Further analysis shows that, while the 2006 and 2007 vintages are likely to experience more than twice the losses of the 2002 and 2003 vintages, the losses in these worse vintages should not affect the principal repayments of the investmentgrade CMBS tranches rated "A" and above, even though some of the recent BBB-/BBB tranches will be under stress.More specifically, the aggregated future collateral losses for the CMBX deals are forecast to be 2.5 percent, 3.6 percent, 3.3 percent, and 3.7 percent for CMBX.1, CMBX.2, CMBX.3 and CMBX.4, respectively.

All these baseline analysis loss numbers are significantly below the double-digit loss projections built into the current CMBX spreads. In fact, in a severe financial stress scenario that has a mere 5 percent chance of happening, the highest ten-year loss rate is 8.5 percent for CMBX.4, still beneath what the current CMBX pricing is calling for.


Tyler Durden said...

can you share your assumptions in getting to these losses

Tyler Durden said...

because i am curious why the market disagrees

OhmeOhmy said...

Assumptions provided by Turbo Timmy & JPM.

daddyo said...

I wouldn't use CMBX levels as an indicator of any real world expectation.

For anything below AJ, the flows are nearly non-existent, and limited almost entirely to people hedging books.

For the AJ/AAA, you have a fair mix of real CMBS players, hedgers, and fast money that is just making a "cheapish" bet that things will slide to ABX territory.

I would also argue that most banks are probably hedging CMBS books completely at this point, and possibly even using CMBX as a CRE whole loan hedge.

Insurance companies and other non-bank investors will feel CMBS MTM pain, but I don't think the banks will take significant losses in that sector going forward.

Now, the unhedged CRE whole loan books are a different story. That's a big twinkie.

Anonymous said...

I have yet to find anyone who has even tried to justify market levels to me from a loss perspective, particularly higher in the capital structure. There are a million answers I hear on the technical side.

What kind of return would you want on a risky asset?...particularly one that is exposed to being pummeled by the markets? How do these required returns reflect back on implied losses? Are those really implied losses or just a reflection of the return required by certain investors? Care to hold these bonds on your books and show poor short term returns to your investors? Do you think redemptions might happen if that were the case? How about leverage, some? If not, what unlevered returns might you want? Where are all the buy and hold real money accounts that were the main buyers of these products, banks, life co, pensions? You think they are active buyers in their condition? If you have taken a MtM beating time and again, care to keep coming back for more? How about competing products,...what kind of unlevered loss adjusted returns can you get in competing sectors like RMBS? If you are being called for redemptions and needed to raise cash, what would you sell to raise cash?...something valued at 30 cents or 90 cents.

Here is a CMBS example: A1 bonds, are the front principal pay bonds, they are currently paying down as normal with any loan amortization, and if there are any loans that default, they pay down (unless loss severity is 100% on defaulted loans). Try to run a model where they don’t make at least 10% returns at today’s prices. You need 99% loan loss severities with huge default rates, anything less and you get paid off at par for your discounted price purchase bond.

People like to think markets are efficient, and while they may be in some ways, from a standpoint of smarts, they are anything but that. We just aren’t very smart. We buy internet stocks on firms that have no revenue, we buy homes when we could rent the same for 1/3 the monthly payment, we buy oil at 150 with people claiming 200 to 500 oil, only to have it go to 40 on us, we buy CMBS BBB- bonds at a spread of 85 only to watch it go to 10000. We pick and chose when to believe market levels as being "correct". Today isn’t any different. If you have been pessimistic about the markets and think we took on to much risk in the past, then at the time you didn’t think the levels were an accurate reflection of risk did you? But today, since the levels reflect the pessimistic distress you predicted, now suddenly the levels are "correct" as they validate your beliefs.

As for the above CMBS loss levels by Anon 8:47 - while they may be a baseline (maybe a little light, but we aren’t smart enough to predict at this point), if I was an investor, I would not buy based on baselines, I would buy on worst case scenarios and then ask for a 10 to 15% return loss adjusted. No need to mess around when there are safe alternatives out there that offer these kinds of returns.

I agree with all of daddyo’s comments as well.

Btw, I am long CMBS in my personal accounts at very attractive risk adjusted returns. Happily, I am a hold to maturity buyer and not concerned with mark to market.

vp3434 said...

Anon@3:11: 99% loss severity to impair the AAA tranche? Doesn't a 60% default rate with a 60% loss severity impair a super senior AAA tranche with 30% credit support? What are your other assumptions.

Btw, I think the poster who was suggesting baseline scenarios of 1 to 3% loss rates is smoking crack. He must work in risk management at an investment bank.

Anonymous said...

Yes, if you know these structures, and know what an A1 bond is. Within the super seniors, there are various bonds tied to expected loan amortization and maturities. If you buy the front pay (short est term) A1 bond, then any loan that defaults with a severity less than 100% will cause your tranche to be paid down first (generally). So to generate 30% losses, unless all those losses were generated with 100% severity loan losses, you are already paid off well before losses reach 30%. thats why the loss severity needs to be so large, so that you dont get paid down.
So in your example of 60% def and 60% severity, that would generate 24% of principal paydown to the deal. that would pay off a lot of bonds, far more than is needed for the A1

it is safer than anything i can imagine,..yet offers 7-8% yields. (A1 yields are no longer at 10%, they have come in some since the last I looked). If you tell anyone that trades these or RMBS you are making 8% returns and excited, they will tell you that return sucks, because there are far better opportunities available. i only offer it as an example because its so obvious and is a good example of how messed up these markets are.

Anonymous said...

Anon 8:47
curious on your CMBX.2 with a higher cumm loss at 3.6% vs the CMBX.3 at 3.2%.

was that a missprint? if not, any thoughts on why the relative strength of 3 vs 2?

vp3434 said...

I'm not an expert on this, but I don't think you're right. Even A1 bonds have credit support at a maximum of around 30%. Yes, the A1 bonds are first to amortize under normal repayment scenarios, but in the event of default, proceeds from liquidation of the collateral are paid pari passu to bonds of the same credit support levels. Otherwise, an A1 bond would have a credit support of over 76% per your example of a 24% principal paydown from default making all of A1 whole. Did you work in CMBS?

Anonymous said...

yes A1's have 30% credit support but super seniors are not paid pro rata with respect to principal payments. it doesnt matter how the principal is created, its paid out as described in the prospectus.

"Otherwise, an A1 bond would have a credit support of over 76% per your example of a 24% principal paydown from default making all of A1 whole. "
Not correct.

if you have bloomberg, run an A1 in the SYT page and try to get it to lose money.


vp3434 said...

I stand corrected. I checked with a friend in CMBS and he said you're right. Seems like if A1 is small relative to the deal size, then it's pretty indestructible. I think commercial RE is going to get ugly, but not overnight. You're saying you can get 7-8% yields on A1s now? Can you post some CUSIPs? I want to look into buying some. Thanks.

Also, do you know where I could get my hands on some prospectuses?

Anonymous said...

last i looked, yields were in that range, not sure after today.

not easy to find CMBS bonds. best way is probably to find a dealer that has a CMBS desk, and ask for their offerings.

once you find a bond you are interested in, the prospectuses are filed with the SEC, which you can view from their site.

good luck

stockman3434 said...

Thanks anon, I found some prospectuses on the sec website. Will check with my broker tomorrow to see what he has.

Just to be clear, I think there are scenarios in which A1 isn't solely paid first. For example, assume the following:

-$5B deal
-$100mm A1 with 20% credit support.

Let's say two large loans of $1.5B in principal default and are written down to 10 cents on the dollar. At that point, class A-M will be written down completely and the $150mm in recovered principal will be paid prorata to outstanding super senior clases (A1-A4).

Something like this is unlikely, but do you agree that the mechanics check out?

In actuality, even in the worst case scenario in which A-M is eventually completely written down, I think the intial defaults won't happen in such mass that A1 doesn't get a chance to be completely repaid before things really hit the fan.

You mentioned you bought some A1s. What deal and at what price, if you don't mind sharing?

Anonymous said...

yes, i think you have the basics of a scenario in which it could happen. but its not 20% credit support, its 30%. effectively you would need 30%+ losses to happen prior to getting paid down.
so your scenario of 10 cents on the dollar of 1.5b (90% severity loan loss) would not create a loss on the super seniors, but would pay off the A1. Your scenario would have to happen without any other prior significant defaults which prior would have paid you down.
another factor making your scenario difficult, loan sizes are typically limited in size relative to the size of the deal, so it would be hard to find a deal with two loans that are 30% of a deal. Maybe a better way to put this, its easy to find deals where it would take 5 or more loans that combined would be 30% of the deal.

last thing to help you out, many A1 bonds have already amortized down maybe in half, so if you look at a 1 million notional bond, its probably down to 0.5 million current now since principal has paid down. this effectively reduces the size of the A1 relative to the size of the deal. making them even safer.

I have not bought A1's, in my view not the best risk adjusted return in CMBS space for a hold to maturity buyer. I used it above because it was easy to explain the relatively risk free nature of the asset and the corresponding large return. I own similar bonds and also other seasoned CMBS bonds, which are not as easy to explain why i like them better...other than the obvious that the return is far better.

Anonymous said...

lastly, make sure you go over your conclusions with your buddy who is in CMBS before you buy,...investing based on advice from an idiot on a blog is bad practice.

vp3434 said...

Got it, thanks anon.