Spreads on AAA CLO tranches not budging

Top rated investment grade US corporate bonds now trade at or even below pre-recession levels. Depending on the maturity and the type of issuer, new issue paper clears the market at spreads (to treasuries) of 30-80bp for AAA bonds.

Securitized corporate paper for similar maturities on the other hand continues to trade at a significant premium. AAA CLO bonds clear the market at 145-155 basis points spread to LIBOR - unchanged from two years ago. The two (typical) CLO deals below show that while lower rated tranches have tightened significantly (BBB for example tightened about 200 basis points over the same period), AAA spreads for standard tenor deals have not budged (and in fact are higher in some instances).

DM means effective spread
(some tranches  are issued at discount; L="LIBOR"; source: LCD/S&P)

Some attribute this premium to higher risk of structured credit relative to single name bonds. However it is important to note that not a single AAA CLO bond lost principal through the financial crisis. The elevated spread is primarily driven by regulatory pressures and funding markets. The new FDIC rules for example penalize banks for holding these bonds (this is in addition to the Basel rules). Ironically these same rules may encourage banks to move toward riskier CLO tranches with higher yields in order to compensate for the increased FDIC charges.

Normally when there is a market dislocation such as this one, hedge funds find a way to take advantage of it. But LIBOR+150 is not yieldy enough for hedge funds and these bonds are nearly impossible to leverage in order to boost yields. So hedge funds and others in search of yield stick to the lower-rated tranches. Non-US participants, in particular yield-starved Japanese institutional investors, have been the only consistent buyers of AAA CLO paper recently. This makes the primary market vulnerable to disruptions if these investors decide to exit.

SoberLook.com

From our sponsor:

The CLO market - then and now

After a conversation with Ed Grebeck (Tempus Advisors), I thought it may be helpful to do a quick overview of the evolution of Collateralized Loan Obligations (CLOs) from the "bubble" years to the current environment. The CLO market is the only securitization survivor of the financial crisis that has retained the full capital structure - from "AAA" down to equity (unrated tranche) and has fairly long maturities. The AAA tranches of the pre-crisis years were severely mispriced - paying as low as LIBOR + 23bp in 2007 for example - but came out of the crisis mostly unscathed (except in a couple of fraud cases).

Unlike their brethren in the CDO space that securitized sub-prime resi loans, the CLO collateral - portfolios of non-investment-grade corporate loans - experienced fairly modest default rates. The US corporate sector did reasonably well through the recession and was able to tap the HY bond market to refinance its debt or extend maturities (even for many weak credits). Some companies of course struggled and were too leveraged (like Clear Channel), sometimes undergoing debt restructuring (Tribune, Dynegy, etc.). Some firms, such as TXU - one of the largest LBO transactions to date - are yet to be restructured.

But even more importantly it was the diversification assumptions for the corporate sector that generally worked. Different industries were not impacted equally by the downturn. In the subprime mortgage space on the other hand, diversification was based on geography, a highly flawed approach. The rating agencies applied similar diversification assumptions/logic to California and Florida properties that they would to telecom and energy corporate sectors. The lesson was that corporate diversification, which has been employed by banks for centuries, works reasonably well but can not be translated into residential property markets based on geography.

In spite of its relative success, the CLO market has changed markedly since the pre-crisis era. With no ability to use bank-backed commercial paper (ABCP) to fund the AAA tranches and no monoline (such as Ambac) guarantees, the AAA needed real investors. That means volumes, pricing, and leverage all needed to adjust. Volumes are of course a fraction of the bubble years these days, but are beginning to pick up.

Source: JPMorgan

Leverage has also changed dramatically. The pre-crisis leverage kept increasing through 2007 as the equity tranche became "thinner". 2007 leverage got as high as 15:1 (15x), averaging 12x (assets to equity).

Source: LSTA

That leverage dropped to 7.5x in 2010 as the rating agencies swung to the other extreme, taking the ultra-conservative approach.

Source: LSTA

Leverage has increased somewhat since 2010. One of the deals that printed this month, managed by Symphony Asset Management and structured/distributed by Morgan Stanley, is leveraged just under 10x (417.75M total deal size over 43M equity tranche size = 9.7x).

Source: S&P

Note that the X tranche (78bp) represents Morgan Stanley's fees as well as closing expenses (mostly legal and "warehousing" costs) which the bank financed for Symphony via the "super-senior" tranche. Pricing for the AAA tranche is now around LIBOR+150, making it a bit more interesting for institutional investors like insurance firms. Here are a couple of other recent CLO deal structures - both with leverage under 10x.

Source: JPMorgan

Other changes include permissible collateral. In 2007 CLOs permitted the inclusion of 5-10% of unsecured HY bonds and 5-7.5% of tranches of other CLOs. Up to 10% of second lien (as opposed to standard first lien) loans could also be included. Current CLOs generally do not allow any of this. Maybe a couple percent could be in bonds, but they would all have to be senior secured.

Legal maturities went from 12-14 years in 2007 to about 10 years these days. Perhaps the biggest change has to do with the "reinvestment period" - the period during which the manager is permitted to replace loans that prepay (partially or fully). It went from 6-7 years in 2007 to 2 years these days. Investors want the manager to build a portfolio and after a couple of years let the deal begin amortizing. That makes the AAA durations considerably shorter, reducing mark-to-market volatility.

CLO managers continue to be plagued by difficult markets. Low volumes of new institutional loans make it harder to ramp collateral quickly (a sufficient amount of collateral has to be invested in order to close the deal). At the same time banks do not finance loan "warehousing" for too long in fear of getting stuck with the collateral if the market shuts down and the deal does not close - which happened in 2008. Equity returns are considerably lower (because of lower leverage) than they used to be, ranging from 10% to 15% (vs. in the 20s during the pre-crisis era). New regulations pertaining to bank capital and risk retention rules will put a damper on how much the banks will be able to structure and distribute. And of course Europe could quickly bring this market to a grinding halt. Nevertheless as demand for fixed income stays strong (these days investors are chasing anything with a coupon), CLO managers are cautiously optimistic.

SoberLook.com

Why CLO managers continue to struggle

By all accounts the CLO business should be going well. Some $3.3 billion of new CLO deals have been done this year already as compared to $4.4 billion for the whole of Q4 of 2011. Seems like a significant amount but this volume is actually quite light relative to historical levels (see chart).

In fact CLO managers are finding it tough to do business these days. Here are some reasons why they continue to struggle:

1. In order to execute a CLO deal, a manager needs to have a portfolio of loans ready to go. Most stock traders would say, so what, just buy it in the market. But corporate loan liquidity is quite poor, and to accumulate (“ramp”) say $200mm of loans in order to launch a new deal could take months (weeks if you are lucky). This process is made particularly difficult because loan “issuance” (origination of new corporate loans by banks) has been rather slow (see chart). There is practically no LBO activity and the merger activity has slowed down. These are the typical transactions that require corporate loans - the rest is mostly refinancing of existing debt. There simply isn’t enough of new supply hitting the market, making it tough for CLO managers to build portfolios.

New syndicated corporate loan volume  (source: LCD)

2. Prior to the issuance of the CLO tranches, managers need banks to provide a “warehouse” – a facility to finance the “seed” portfolio. The problem is that such facilities these days require 10-15% first loss commitment from the manager. Smaller managers may be required to post that amount as cash in order to obtain the warehouse facility. And that commitment is nearly impossible for some managers. That’s why most CLO issuers in recent years have been the bigger firms (such as private equity).

3. The “equity” tranche for new CLOs has prospective returns of 10-14%. This is due to the fact that senior tranches now require significant subordination by the rating agencies – who are being overly conservative to compensate for their "sins" of the past. Higher subordination makes the equity tranche larger and the returns lower. At these modest returns and given the negative connotation associated with any structured credit, equity tranches are hard to place (particularly with pensions and other institutional investors).

4. It may also become difficult to place the AAA tranches soon. The buyers of these tranches tend to be banks and the new regulations may make holding this paper uneconomical.

Because of these uncertainties, analysts are divided as to the expected size of total CLO issuance this year. Forecasts range from $12 to $25bn for the whole year. Back in 2007 that would have been a bad quarter.

SoberLook.com

3 observations about the AAA CLO market

Here is followup to a recent post on the proposed draconian capital rules for banks holding AAA CLO paper.  Any regulation, whether it's pharmaceuticals or financial products, should be based on data and other empirical evidence rather than political motivations and career advancements.  Over-regulating something just because a regulator does not fully understand the market can result in unintended consequences with adverse impact on the economy.

Let's consider the data available for AAA CLO tranches to see how it reconciles with the proposal to increase capital requirements by a multiple. Here are 3 key empirical observations about this market.

1.  Current CLO deals typically have a 35% subordination for the AAA tranches.  The collateral represents a diversified pool of senior secured corporate loans with 90% or higher in first lien.  In order to impair a AAA tranche, 70% of corporate loans have to default, assuming a 50% recovery rate (which is generally much higher for first lien senior secured loans). Even during times of severe credit crunch, loan default rates did not exceed 10% in a year. 

Institutional loans default rate (CS)

2. As conservative as the rating agencies have become in downgrading everything within their sights, their analysis continues to show that the bulk of AAA tranches actually remain "AAA".  Some may argue that they have seen this movie before, but given the guns pointed at the rating agencies, they have all the incentives in the world to downgrade CLOs.  They have aggressively done so for the more junior tranches but not the AAA.

Downgrades of CLO tranches by Moody's (CS)

3. If these assets were more risky than investors initially believed, a year like 2011 would have proven it out and the risk would manifest itself in price volatility (as it was with other assets). Again, that was definitely the case for the more junior tranches, but the AAA paper held up well.

So before the Basel rules are adjusted as proposed, bank regulators need to look at this data.  Rather than overcompensating for the RMBS secularization fiasco in 08, regulators need to start from scratch when developing capital rules for this market.

SoberLook.com

Rating agencies under fire

The rating agencies are now taking hits from every direction. The Big Three are becoming increasingly focused on trying to defend themselves against law suits, investigations, and pending regulation. The most critical blow recently came from the highly publicized case involving Cheyne Finance. This was an SIV that issued medium term notes which were rated "investment grade" by the agencies. The notes defaulted recently and the holders took the agencies (as well as Morgan Stanley) to court.

Dow Jones: Two credit-rating agencies must defend a lawsuit over the collapse of a $5.86 billion structured investment vehicle in 2007, a federal judge has ruled, rejecting an argument that their ratings are protected free speech.

In an order Wednesday, U.S. District Judge Shira Scheindlin in Manhattan rejected an argument by rating firms Moody's Investors Service and Standard & Poor's that their ratings of SIV Portfolio PLC were protected by the First Amendment.

The rating agencies' traditional defense had always been the "freedom of speech" concept. The argument is that the ratings they issue are simply their opinion and therefore should be protected under the US constitution. This case however shook things up because the judge ruled otherwise (at least in a couple of specific instances). With evidence like this IM exchange between two S&P employees, it's a bit difficult to argue freedom of speech.

source: CNBC

The agencies are confident they will be able to overturn this ruling, but nevertheless this may set a precedent and open a floodgate of litigation.

But the Big Three's problems are not limited to investor lawsuits. As people begin to fully appreciate their role in the financial crisis, more players are coming out of the woodwork to start investigations, file complaints, or propose regulation. Now the California Attorney General is jumping on the bandwagon.

Reuters: California Attorney General Jerry Brown issued subpoenas on Thursday to Standard & Poor's, Moody's Investors Service, and Fitch Ratings as he launched an investigation of whether they broke state law with the ratings they provided mortgage-backed securities.
…
The investigation focuses on whether the agencies broke consumer protection and unfair business practice laws, in the most populous U.S. state, which give the state broad authority to bring suit in cases of false advertising and unfair competition.

The insurance industry is also getting ready to pound on the rating agencies.

NAIC: The National Association of Insurance Commissioners (NAIC) will hold a public hearing on September 24 to discuss the past and future roles of Nationally Recognized Statistical Ratings Organizations (NRSRO). The hearing will examine the role of these credit rating agencies in the insurance regulatory system and what changes may be needed in light of the financial crisis.

Insurance companies hold nearly $3 trillion in rated bonds and the insurance industry constitutes the largest sector of the financial services industry to rely on credit ratings to supervise capital asset adequacy. Insurance regulators currently mandate the use of credit ratings to determine capital reserves and other regulatory requirements for insurance companies.

And here comes the SEC with what seems to be a fairly constructive disclosure rule that may cause the agencies and banks some pain:

Reuters: The SEC may require banks and other issuers to disclose preliminary ratings, to prevent them from shopping around for the better ratings, the people familiar said. They requested anonymity because the discussions are private.

The SEC may also require all credit agencies to reveal more information about past ratings so investors could compare their relative performance, with perhaps a one- or two-year time lag, the people said.

The structural problem with the rating agencies and their role in the crisis is finally getting the much needed attention. Here is a simple fact: if the rating agencies' subordination requirement for AAA subprime paper was higher when the deals were rated, the subprime market would never have grown the way it did. A typical subprime deal had 21% of collateral below AAA, while Alt-A subordination was about 7%. If the rating agencies were just a bit more conservative, requiring a higher cushion, the mortgage rates for these loans would have been much higher to make the securitization work. And that simply means most such mortgages would not have been possible.

The question is where do we go from here. Securities ratings process is so entrenched in the financial services industry, it's hard to imagine the system functioning without some form of ratings. It's part of the lexicon: when we discuss "investment grade" market or the "high yield" market, there is an implied rating assumption - whether people use it or not. The lesson of 2008 was of course to rely less on the agencies, but the system depends on them being there. And the government is doing a great deal to increase (rather than reduce) reliance on the ratings process: from new money market regulation to TALF - all depend on some form of agency ratings.

That means that the rating agencies are here to stay, and when the dust settles, the surviving agencies (and it's not clear if all of the Big Three will survive) will have a great deal of work to do. It's critical that we fix the ratings system: from agencies' internal structure to the ratings process to models and transparency to fees/conflicts - all must be revamped in order for the system to function properly. Because not getting this right will set us up for another fiasco down the road.

CMBS TALF is starting to ramp up

In our previous discussion on CMBS called CMBS balloon risk looming, we covered the refinancing risk for commercial mortgages. These mortgages form the collateral pools for CMBS bonds, creating significant risk that these securities will default on part or all of their principal. That risk however varies dramatically based on the seniority of CMBS tranches.

Here we refer to CMBS securities in terms of ratings, which represent the level of seniority (the actual ratings themselves are fairly meaningless for CMBS at this point). The chart below shows spreads (from Morgan Stanley) for the highest seniority tranche (super-senior AAA) and a junior tranche (BBB).

Super-senior AAA and BBB spreads over US Treasuries (spreads are on two different scales - see left and right y-axis)


Since the start of the financial crisis, the spreads have blown out on these securities in part because of increasing delinquencies, but mostly due to refinancing risk. However as the credit markets stated rallying, the senor tranche spread has narrowed, while the junior tranche stayed at distressed levels. The junior tranche is priced based on coupons it may pay, assuming that it will not pay any of it's principal. In fact in some instances these tranches are expected to pay only a portion of their expected coupons.

The super-senior AAA is different. Even in an environment of highly depressed real estate values, the tranche (for many CMBS deals) is expected to pay a significant portion of it's principal due to substantial subordination "beneath" the tranche. When properties are liquidated (because mortgages can not be refinanced), there may be enough to pay down the AAA, but the market does not expect there to be much left to pay the junior tranches. The diagram below illustrates how the market views the risk on these bonds:



Some view the most senior tranches of CMBS as a potential investment opportunity. Even if not all the principal will be recovered (though some market participants think many of the senior tranches are money good), the discount (or effective spread) may justify the investment. But 500 basis points over Treasuries for securities that are 5-10 years in maturity is not a great return for many investors who are taking this risk.

But the Fed came to the rescue to bump up the return, by providing leverage on the senior CMBS tranches via TALF. This is the riskiest component of the TALF program for the Fed (and that's why the leverage on CMBS is lower) Thus CMBS TALF has started doing some volume.



However, this is still a drop in the bucket, given the $700 billion CMBS market. Also these TALF transactions are secondary CMBS only. The big question remains, is whether TALF will stimulate any private financing of new properties. So far new CMBS activity is nearly non-existent, but there are some deals in the works to take advantage of TALF. From the WSJ:

Vornado Realty Trust, one of the U.S.'s largest real-estate investment trusts, is planning on raising between $550 million and $600 million through a bond sale that would qualify for a key government program aimed at resuscitating the commercial-property market, according to people familiar with the matter.

Participants are banking on a larger TALF operation in September to move some of the massive existing inventory and possibly do some primary deals.

CLO tranches follow the leveraged loan rally

The recent rally in leveraged loans is making it's way into the secondary CLO market. In the dark days of late 08 - early 09, secondary CLOs got hit from several sides. On one hand the natural buyers for senior paper have all but disappeared due to the collapse of the ABCP markets. On the other hand concerns about the collateral quality continued as leveraged loan default rates escalated rapidly. The junior tranches were trading cents on the dollar as participants anticipated cash flows being cutoff from due to expectations that deals will hit their OC triggers (see Moody's pounding CLOs with downgrades ).

There were a few buyers of AAA CLO tranches who felt that AAA CLO paper has what's called "positive convexity". If credit deteriorates, OC triggers get hit and AAA starts getting paid down. Even though collateral gets worse, the principal is getting reduced. And unlike other types of CDOs, the cushion for AAA in CLOs has generally been sufficient - so far. If credit does not deteriorate enough to hit OC triggers, the principal would not get reduced, but spreads will tighten. A few saw this positive convexity as an opportunity to get long secondary AAA CLOs (the primary market is virtually shut down) and at 60-70 cents on the dollar some took a chance. Anything below AAA was viewed as highly speculative and traded at massive discounts.

How the times have changed. As the collateral rallied (see Leveraged loan rally continues), the securitized product is starting to rally as well. The projected collateral default rates have dropped in part due to all the amend-and-extend activity. There is clearly an arbitrage opportunity when the senior most tranche of a loan portfolio is yielding more than the portfolio itself. The AAA rally in turn is pulling with it the lower tranches.

From Wells Fargo:

CLO bull market is raging on. After a big rally in May and June many customers expected a pullback in July, and while we had a couple of weeks of summer lull, the last four weeks prices trended straight up, aided by low supply. A lot of activity was on top of cap structure, with most 1st pay AAA improving from high 70s to mid 80s and AAs from low 50s to mid 60s.

.....

What lies ahead? CLOs are STILL cheap to loans and HY, fundamentals are clearly improving, there's negative net supply of paper, and rating agencies haven't gone overboard with downgrades. Unless there is a meaningful pullback in credit and stocks (SP500 back to 900), we don't see much weakness coming up.

It's certainly impressive to see a rally in a structured credit market, given the dirth of natural buyers. However with the recent equity markets' correction, it may be time for this market to take a pause.