This page has been archived and commenting is disabled.
Shadow Economy Is Back As Citi Prices Upsized $525 Million CLO
The one missing piece for reflating the credit bubble to even beyond its historically monstrous proportions was just put into place, as Citi just priced an upsized $525 million Fraser Sullivan CLO. The CLO focuses on the red-hot High Yield and Bank Loan markets. The terms are as follows (via Loan Connector):
More from Loan Connector:
Earlier, the AAA tranche was at $327 million, the AA at $15 million, the A at $36.5 million and the equity tranche at $102.1 million. All tranches priced at their talked coupons.
The newly priced CLO is a rollover of an existing Fraser Sullivan CLO. The first payment date is Oct. 20, 2010 and quarterly thereafter. The CLO has a settlement date of April 16, 2010 and matures on April 20, 2019. Its reinvestment period ends on April 20, 2012.
The existing Fraser Sullivan deal consisted only of a AAA tranche and an equity tranche, and hence, the pricing of the AA and A notes marks a new price point for those tranches since the credit crisis.
Welsh Carson Anderson & Stowe Fraser Sullivan was formed in 2007 and invests in the high yield credit markets with a focus on bank loans.
The firm manages clients' capital through a variety of investment vehicles and funds including CLOs, levered and unlevered pooled funds and separate accounts. WCAS Fraser Sullivan is the successor to a partnership established in 2005 to focus on the high yield bank loan marketplace.
»
- 4602 reads
- Printer-friendly version
- Send to friend
- advertisements -
It's almost as if the last 3-4 years never happened.
"We have always been at war with Eastasia."
And there was nothing different about the last 3-4 years. They were the same as the last 10 years. Carry on.
http://en.wikipedia.org/wiki/Nineteen_Eighty-Four
Yeah, one could perhaps be forgiven for naively thinking that a big, fat, nearly "world ending" crash would smarten people up.
But so long as Uncle Sugar is supplying the booze and the room it's party on. Woo-Hoo!
This hangover's gonna be a bitch, though.
Madness, sheer madness. - The Bridge on the River Kwai
And I can just imagine that the public pension funds are just gobbling this up.
What a bunch of maroons.
reading this gives me chills - My gut says something is very wrong here
is bond market open on friday for shortened hours?
yes
Take another hit from the bong- party on!
"If I say its safe to surf this beach Captain, then its safe to surf this beach. ... "Charlie don't surf!" Col. Kilgore (Apocalypse Now)
http://www.youtube.com/watch?v=vHjWDCX1Bdw
Great. Sounds like they are adding another 10 floors to this skyscraper of a house of cards...
...built from bricks taken from the foundation, no less.
Next up - The squid will be rolling out CMBS on a 5 year floater I/O.
They have no way to sell this stuff without a middleman fee and the perceived perception of safety (AAA) all for 2% yield. You have to be insane to buy this crap or you have to be using other people’s money (which you should be sued for breach of fiduciary duty). I’d love to know the where this stuff went if in fact it went anywhere expect the banks books…….giving a better mark.
Agree. 10-year AA corporates currently yield slightly more than 5%. Despite being subject to interest rate risk, I would rather own fixed-rate AA corporate paper (credit specific) at 5.0% than the AA-tranche of some POS CLO yielding 3.2%.
FWIW, I would call the AA yield more like 417. The reinvestment period is only 2 years, so the deal is likely to be called in about 3 years (instead of the typical 6).
It is much less liquid than corporate paper. On the other hand, it is backed by a diversified portfolio of loans instead of 1 name.
I'm not a CLO expert but I was interested in how you came up with a 4.17% yield with terms of L +225bp @ 94.252. Implies that this paper is called at roughly 106.5 in year 3.
Wasn't the diversified loan portfolio backing the CLO the selling point when the ritz hit the fan in '08?
2 points.
- I said yield, when I should have said spread. Of course with LIBOR at 29bp, it doesn't make too much difference.
- Not having a calculator, I did a back of the envelope calc. Not an accurate DM. I simply amortized the discount to par over 3 years and treated that as effective spread, which would be an additional 192bp. Not the correct way, I know, but it shouldn't be that far off
- I just checked my math using Excel yield function (40269,41365,0.0254,94.252,100,4,2) and got 4.60%. Backing out 29bp LIBOR gives a spread of 431, so not too far off from my quick and dirty calc of 417
Diversifed portfolio was indeed a selling point back when. And it is valid. Basic portfolio theory says you will have lower volatility with a diversified portfolio. Of course the correlations can get you if not careful.
But the issue in CLOs was more nuanced.
Loan CDS had recently come into play, so there were more players (esp HF) jacking around CDS as they tried to arb it or to make a cap structure play. This introduced new volatility into the loan market that never had been there.
But the real issue was on the demand side. CLO investors are worldwide institutions (especially banks and insurance companies, but also high net worth and pensions, and of course CDO^2).
These investors were also investors in CDO^2 and mortgage CDOs. When the mortgage CDOs tanked, the investors lost bucks big time. That dried up their available capital, and in many cases their willingness to buy CDOs of any flavor. Usually, that was due to management freaking out moreso than the investors (portfolio managers) themselves. I know this because we got hit with the 2007 downturn while we were marketing a deal.
With no demand for CLOs, there was no demand for loans, without that demand, the prices plummeted (9 sigma based on historical volatility). At the same time, the banks had huge loan positions on their books they had intended to sell to the CLO market. So supply goes up, demand goes down, prices fall.
Loan defaults were not the problems in CLOs. While they went up, it was more or less in line with the 2002 recession, which the CLOs weathered quite well.
The issue was a dramatic decline in demand due to the investors' portfolio stresses caused by mortgage CDOs.
So, yes, diversification was and is a selling point. It's not a cure-all, but it helps. I'm not faulting the idea of buying a corporate bond, instead of a AA CLO tranche, just pointing out the differences
While your astute observations seem to be from a professional level, allow me to add a bit more from an academic perspective.
Since you brought up portfolio theory, the most basic portfolio theory, going back to Fama & French with the whole introduction of optimal portfolios, we can safely say never meant for any sort of CDO (or debt for that matter, since their work relied around equity) to be present in a portfolio. However, their lesser known work still focused on debt financing and how to create synthetic positions across different asset classes.
What i want to add to your discussion, is the notion of creating synthetic positions in your CDOs comprised of straight debt (or corporate CLOs in this case). If one can purchase some treasuries as collateral, then write some CDS on any large entity with enough depth and of equivalent value/maturity, youve effectively created a synthetic position in your debt. Plus, you get a premium on your holdings of whatever the bp spread is over your treasuries. And since youre (presumabely) not holding the underlying asset of the CDS, you dont lose if your entity goes down either.
This creation of anything synthetic is exactly the theme of securitzation and structured finance. While i wouldn't say loan defaults were NOT the problem for the CDO market literally dying overnight, you can't say it was demand, and just demand. Look behind demand, what drove the decrease? Yes there were stresses in banks not wanting to hold these things anymore, PMs not buying those things anymore, and of course the retail ABCP (here in Canada, dont think it hit the US) market going down at the same time; but the stresses were again from the very start of the chain which begins with defaults.
Fama and French studies may be on equity but the diversification benefits still apply to loan portfolio. I know that from experience (though academically, that would be anecdotal I know). But, as I think you are alluding to, CDOs are very complicated and cannot be completely modeled, especially if they are actively managed.
Your point on a synthetic position is well taken. However, a Cash CLO (as opposed to a Synthetic CLO) has a strict limit on CDS or other synthetics (varies, but as I recall it was 10-20% range). We never did get comfortable with CDS as an asset because of the counterparty risk (rightly so, it turned out) as well as liquidity and documentation issues (e.g., default definition)
I disagree that defaults were at the start of the chain. At least not loan defaults. The default rate was still around 1% in 2007 when it hit the fan. The demand for CLOs dried up because investors had no cash. They had no cash because they took it on the chin in RMBS CDO.
Because we managed Cash Flow CLOs (as opposed to Market Value CLOs) our structures had the luxury of never having forced sales, so if we like a credit, we can work it out in bankruptcy. Defaults stress the cash flows but never force a sale. And the stresses are at the bottom of the waterfall (the equity and BBs) and should not hit the AAA unless you eat through 25% of the structure. With a hypothetical 10% default rate and 50% recovery, it would still take 5 years at that stress levels.
But AAAs widened dramatically from 25bp in early 2007 to this new deal at 190. Defaults didn't do that.
If you say mortgage defaults started the chain reaction, ok, but it clearly was not loan defaults.
I like your synthetic loan idea. Unfortunately, rating agency models did not allow for that.
Does anyone know if Ben's writing the check?
"...the pricing of the AA and A notes marks a new price point for those tranches since the credit crisis."
New prices to mark all the "AA" and "A" toxic crap in inventory.
Why does my head suddenly hurt?
Someone please explain why this is such a bad thing? Structured finance was never going to dissapear, and it looks as though there is plenty of equity to cover any potential downside. I'm not trying to make an arument for this, as much as I am trying to understand why the hostility and what are the bad mechanics about this deal.
I don't know this deal, but in general I would assume to hostility is due to a return to failed models.
1. Depending on rating agencies yet again for something that they utterly misrated before.
2. Ambac just BKed. What other monolines are on the chopping block, and who is involved in insuring the issue.
3. Performance triggers in the past left a lot to be desired re: creditworthiness of a so-called AAA tranche.
If you know they deal, I'd like to know how these issues aren't a problem now.
BTW, the 7Y swap spread went negative.
I don't know this deal specifically, but I know CLOs (9 years experience - all job offers welcome ;-) )
1. Rating agencies are far more conservative than they were a few years ago. Also, Fraser-Sullivan is an experienced manager (though not the most experienced)
2. I don't know if any notes are insured. Frankly I doubt it given the state of the monolines. Plus the insurance on the old deals was to create a negative basis trade, which I doubt can be done anymore.
3. No info on the collateral quality tests, so I can't comment specifically, but note the high equity % and low AAA % in the deal, compared to pre-crash vintage deals as noted in my comment below)
This isn't an attack on your comments at all. Just some questioning.
Rating agencies may be more conservative than they were, but what is the level of external validation especially aggregation of tranche risk exposures? I'm not an expert, just assuming that is the core where transparency needs to be improved.
Securitization is a great vehicle and it will come back. Just seems like we are trying to reboot something before we get some needed transparency.
Questioning is good. No offense taken at all (not even sure of anything remotely offensive, so no worries).
I'm not sure I understood your point though. Are you asking about what validation of the rating agency models has occurred? Not sure as I have been away from the CLO market for a while (job hunting....). However, the models aren't that complex for a CLO, IMHO.
Not sure what you meant about aggregation of tranche risk exposures. It sounds as if you are thinking of a CDO2. Those models probably need a lot of work. They are much more complex and cannot realistically model the real world as there are too many variables (especially portfolio manager discretion). Aggregating tranches of CDOs into a portfolio is very complex. Lack of transparency into the underlying is a key problem.
But aggregating bank loans (corporates, not mortgages) is pretty simple. It's not too much different than modeling a portfolio of bonds.
Anyway, I may have misunderstood your point, but I don't think you are describing a CLO.
I am talking about the validation of rating agency models that they use to assign a credit score. Seems so black-box.
About aggregating bank loans or any defaultable revenue stream for that matter. The whole copula methodology didn't seem adequate to handle the stresses of the CLO structure in what was an extreme environment of late 2008. Specifically, the model used to quantify the probability of correlated default didn't provide a good enough estimates of loss.
Thanks, man.
I see. I don't believe the rating agencies ever used a copula model for rating. I could be wrong about it, but I never heard of them doing so.
The only time I heard of copulas being used was in pricing for use by traders. I tried to read up on it, but I never saw an example of how the math was to be used, just theoretical discussion, so I can't claim to know too much there.
I think you are right in that the models (copula or otherwise) did not properly quantify the correlations. The problem, I suspect, was that the correlations were based on history which was relatively benign. Unfortunately, as proved to be the case, default correlations do increase in times of economic stress, and that was not modeled well, as you say.
But I still think the main stress on CLOs was due to demand falling off a cliff, and not to defaults of the underlying. And no model of the structure would have considered demand except in modeling price performance. The loan pricing had been very stable before that. Even in the 2002 recession, loan pricing bottomed out at about 90 on average, rather than the 50 it hit in 2008.
So to create a CLO tranche, do folks just score loans with a risk/reward score generated through logistic regression, and then tranching is done via score cuts?
Indeed, structured fin never will disappear, but youd assume the next level of financial engineering would at least bring into account some of the problems which cause the last mess in 06/07.
This CLO will no doubt be packaged up many a time until it becomes CLO^4 and your supposed AAA paper is actually the bottom tranche of a completely diff security.
.
At first glance I would go ranting too....but truthfully I cannot comment enough on the CLO market. I personally know the CMBS market, and fully expect hooting and hollering when the first new issue CMBS comes out though it is undeserved.
The TBTF are also busy for a while repackaging the FDIC's
seized assets, CDOs and MBSs. Securitisation,
wash, rinse and repeat.Blythe Masters at Jamie's at work on
some new synthetics repored the Financial Times Deutschland a few weeks
ago , she gets paid in diamonds, lol
Bennie to kick the can and extend the program ?
The risk premiums look a bit light to my untrained eye, but there shall be no worries in the financial sector. In the Nanny State of America, only one bad outcome is permitted - enslavement.
I wouldn't go so far as to say that the shadow banking system is back based on this deal. The average spread as printed was 197bp. Factoring in the $4.5MM discount (and assuming the deal is called 1 year after reinvestment period, that works out to an effective spread of 235bp.
In late 2007 (after the first downleg in the economy), my old shop printed a CLO at about 118bp with an effective spread after discounts of about 150bp.
In early 2007 (about as hot as the market ever was), we printed at 43bp average spread with no discounts.
Also consider that the new Fraser-Sullivan CLO has 22.5% equity. That compares to 10.3% equity in the late 2007 deal mentioned above and 7.6% equity in the early 2007 deal.
This doesn't strike me as egregious at all.
Also the new CLO has AAA spread of 190 (vs 87 and 25 for the two deals I mentioned). Not only that, but the AAA in the F-S deal is 67% vs 75% for the 2007 deals.
There is nothing fundamentally wrong with CLOs. They are leveraged investments, but the leverage is way down here from what it used to be. Senior secured loans are relatively stable (much lower price volatility than other instruments), they are senior secured and so have lower default rates and higher recoveries.
And perhaps most importantly, CLOs are managed. The portfolio manager chooses the loans that go in the portfolio, as opposed to garbage being stuffed in by a bank trying to clean its balance sheet. (Actually, the F-S CLO is only lightly managed, so this doesn't hold quite as well, but this is dramatically different from the mortgage CDOs that were the real problem.
Interesting commentary. I think everybody agrees that leverage magnifies returns - both positively and negatively. A higher level of equity obviously reduces risk. However, isn't that offset by the dramatic spread tightening in corporate loans and bonds?
Your comments got me thinking about the LBO cycle. During the peak of the late '80s (Drexel, RJR Nabisco), LBO transactions could be financed with as little as 10-15% of equity. The HY market collapsed in 1989 and with the demise of Drexel, the LBO business basically ground to a halt. When LBOs came back in '92-'93, the equity component increased to 30% + of the capital structure. Still, the transactions that were done in the early '90s generated excellent returns despite this higher equity component because the purchase multiples (EV / EBITDA) were so much lower than the '80s peak (5-6x vs. 9-10x).
So with the revival of CLOs, I guess more equity is a positive, but aren't historically tight credit spreads on the underlying loans / bonds a negative? This transaction is probably good for Frasier-Sullivan, but I'm not sure it will be good for the investors who buy this paper from a risk / return standpoint.
Without knowing what's in the portfolio underlying that deal, I cannot comment too much. But the spread tightening in the loan market really just gets us back to where we were in early 2008, and just inside where they were at the peak of the 2002 recession. They are still about 200bp wider than they were in 2006 and early 2007. Loans now are at a spread of about L+400. They peaked at L+2200 in 2008, and averaged about L+1700 in late 2008 / early 2009. But they had been about L+175 in 2007, and back in the 1990s were about L+325.
In 2007, it was getting very difficult to manage deals, spreads were so tight that early redemption of loans accelerated dramatically. New issue loans could not support the spread on the CLOs very well. So the spread tightening had a dramatic vintage effect. If you had low cost liabilities (say 2006 and 2007 vintage) you weren't too bad off. If your liabilities were higher (say 2003 or 2004) it could be very painful.
This is offset some by gains if you sold away your positions that were purchased at a discount, but those were usually minimal as non-stressed loan paper didn't often trade below 95 (the old level that indicated stress back in the day). Loans usually don't get much above par since they are callable at par at any time.
So spread tightening hurts, but it still is at a respectable level. Figure on 60bp of expenses on the portfolio. At L+400, and assuming no defaults, there would be about 110bp available for equity since debt cost for F-S is about L+240 (doing this in my head, so numbers are rough). At 4.5x, equity would get only about 5%, assuming no defaults.
If the portfolio is full of higher spread loans with call protection (assuming it was available in 2008 and 2009), performance might be ok. It may be they are just lightening the damage from a bad structure in the prior deal.
You hit on an important point wrt leverage on LBOs. Leverage got so high in mid-2000s that the risk of default went up just as spreads on loans were incredibly tight and loan structures were weak. This should show up in the default and recovery stats for loans of that vintage.
That was where CLOs did their part to damage the world. They made debt cheap for companies, so there was a lot of economic expansion as a result. But there were so many CLO buyers of loans, that the PE firms were able to place ridiculously cheap debt, pay themselves a dividend, and at that point had little incentive to make the companies profitable (as the PEs had been paid off in full in about 18 months in many cases). They basically raped many companies. That could not have happened without the bubble in CLOs.
(Sorry for all the rambling)
Thanks Anonymouse for your many posts on this thread. No rambling, but a lot of good data and market insight.
Thank you very much.
I have read the definition of what a CLO is a number of times before tonight.
Until your post, I did not understand them at all. If used correctly they can be beneficial to all parties.
If used.....
Thank you again
Thank you. I happen to be a fan of the asset class, though they did play a part in the financial crisis (due to creating overly cheap corporate debt, not to fraud).
Good linux hosting option package offered by ucvhost which not only provides the best in terms of hosting packages but also believes in truly being there for the customer, 24x7. windows vps Moreover , they offer unlimited bandwidth as well as nearly 1GB storage along with database maintenance, email facility along with storage, availability of sub domain and many other important features for a very low price. ucvhost