For those who would say "yes," it is important to remember that unlike many others in the capital markets -- who can take action on the basis of speculation -- we base our opinions on documented facts.
Right, so if a tree fell in a forest, but no one was around to hear, S&P would still give an AAA rating?
A bond's rating is supposed to be an indication of its future performance, right? So perhaps a little speculation on what the future will bring is called for.
Often an agent hanging around the real-estate office takes an application and forwards it to a mortgage company, which may or may not be his employer. The mortgage company processes the loan, and sells it to an investment bank, or similar entity, that bundles mortgages into bonds. The investment bank sells those securities to hedge funds, pension funds, mutual funds and other investors. As mortgages vary significantly in quality, mortgage-backed bonds are rated by Standard and Poor's and other rating agencies.
At each step along the way, information can be lost and temptations build up to understate the risk of default. The agent gets a big fee for writing the loan only if it is approved, so he puts the prospective homeowner in the best possible light and finds an appraiser who aggressively values homes.
Mortgage companies get their cut from origination fees and are able to push off the risk of default on to the ultimate purchasers of the bonds. Hence they are inclined to be lenient with agents.
Investment banks earn money on the spread between the interest rate charged borrowers and the interest rate on the bonds. The less risky bundles of mortgages that go into bonds appear, the lower the interest rates on the bonds and the more profits the mortgage bankers receive.
As we peel this onion, we are finding many purposeful compromises that look much like the insidious corruption that characterizes commerce in some countries. In essence, Wall Street manufactures bad bonds.
But it gets worse. Large builders established mortgage-writing subsidiaries. When they built more houses than the market could absorb, those subsidiaries exaggerated the incomes and qualifications of buyers on loan applications, and pressed for exaggerated appraisals of their properties to move houses. Those entities operated on lines of credit and resold the loans to investment banks, which bundled them into bonds. Both the large builders and investment bankers had incentive to put lipstick on pigs.
Also, investment banks sought and received collaboration from bond-rating agencies in bundling mortgages of differing quality into bonds. In the process, bond raters such as Standard and Poor's were compromised.
Subprime mortgages are hardly the whole credit market, but the meltdown of their bonds cast a spotlight on the decaying integrity of investment banks and bond-rating agencies. These institutions underwrite and rate all manner of credit, and if they could be corrupted in the subprime-mortgage market, then all commercial paper and bonds become suspect.
Over the past several weeks, creditors have increasingly sensed they can't trust banks or bond-rating agencies, and they have fled to short-term Treasury securities. This was much worse than the collapse of mortgage companies that originated housing loans, because it caused all segments of the credit market to collapse.
Good businesses with sound cash flows couldn't borrow operating capital,
"As part of the ratings process, we do engage in open dialogue with bond issuers. This dialogue helps issuers understand our ratings criteria and helps us understand the securities they are structuring, so we can make informed opinions about creditworthiness. We strive to make sure issuers and investors are fully aware of how we determine creditworthiness and believe that all parties are better served when the process is open and transparent."
I think the translation is: "When we give a rating to a structure lower than that required by the structerer, we explain precisely to him how and why we gave that rating, so he can perturb the structure to get the rating that he wants, thereby allowing him to construct a structure with the desired rating with the maximum possible risk, which is usually several times more than the risk which the rating would indicate."
"This dialogue helps issuers understand our ratings criteria and helps us understand the securities they are structuring, so we can make informed opinions about creditworthiness. "
System gamed. Is there any more need for explanation or need for S&P to exist? GIGO- what comes in is sugarcoated garbage, and what goes out is the same.
Caveat Emptor. That seems to have been forgotten with a ratings stamp issued by these jokers. I fail to see how anyone with a fiduciary responsibility can maintain any benefit in purchasing securities with a given "rating" over any other security, given they can't be objectively rated.
Ah well, they will soon cease to exist if this persists.
But that's the great thing about all this. No one person ever did so much that they feel like they did something wrong. Each link in the chain didn't break the rules, they just bent them a little. It's an example of 'the solution to polution is dilution'/'diffusion of responsibility'. It's not exactly any one person's fault, so it's nobody's fault.
When your paycheck depends on you not fulfilling your full obligation to the market and investors, it should not be a shock that all rating agencies are pretty much obsolete, in this context, and provide a bias rating. Tanta, I agree 100%. My stomach is turning listening to their garbage. If rating agencies are not there to provide an unbiased opinion of the deal and we are not to use their opinions to make any investment decisions, then what in sweet Christmas are they their for. I guess to look pretty and accomplish nothing and be useless. Offcourse, this is all in the context of securitization.
A bond's rating is supposed to be an indication of its future performance, right?
Remember how limited a bond rating is. It makes a judgement about one thing only. What is the risk the issuer will miss a single interest or principal payment on the date it is due. That's it.
a,
I think the translation is: "When we give a rating to a structure lower than that required by the structerer, we explain precisely to him how and why we gave that rating, so he can perturb the structure to get the rating that he wants, thereby allowing him to construct a structure with the desired rating with the maximum possible risk, which is usually several times more than the risk which the rating would indicate."
Except for the very last clause, that is right on the money.
Remember how limited a bond rating is. It makes a judgement about one thing only. What is the risk the issuer will miss a single interest or principal payment on the date it is due. That's it.
With corporate bonds, yes. But with structured cash flows? Really?
Then why has anything been downgraded? Has anyone heard of a bond that actually defaulted yet?
The link that follows is OT, but I wonder if it's worth a separate post? It's an editorial from Prudent Bear's website by Martin Hutchinson titled "Back to Jimmy Stewart!"
I think this excerpt sums up how (he believes) we arrived at our current inefficient mortgage market. Note: Jimmy Stewart, the small-town lender from It's a Wonderful Life, represents old-style home lending:
"The answer is one word: salesmanship. Anybody who has lived in a US suburb with an economically attractive zipcode and no butler will recognize that excessive salesmanship is the bane of American life. This is even more true in the mortgage business. Homeowners today dont go into their local S&L, save for half a decade and request a mortgage from Jimmy Stewart. Instead they are sold a mortgage product, either directly or over the Internet, by an aggressive salesman who is paid a multiple of what Jimmy Stewart earned, or at least aspires to be. That product is then securitized by an investment bank trader who in good years is paid a LARGE multiple of what Jimmy Stewart earned. With others it is sold to a securitization vehicle of immense complexity that has been set up by Wall Street lawyers paid a HUGE multiple of what Jimmy Stewart earned. Costs have been increased at every point in the process, but aggressive salesmanship has driven Jimmy Stewart out of business."
Yup, same standard. Things have been downgraded because the RA's believe the likelihood of a payment default has increased. I haven't heard of an actual default yet, but you could find them under the D rating. Let me do some digging.
"[L]et me tell you right here and now, I wouldnt trade a high-class Zenith acreage development for the whole length and breadth of Broadway or State Street! . . . its evident to any one with a head for facts that Zenith is the finest example of American life and prosperity to be found anywhere.
"I dont mean to say were perfect. Weve got a lot to do in the way of extending the paving of motor boulevards, for, believe me, its the fellow with four to ten thousand a year, say, and an automobile and a nice little family in a bungalow on the edge of town, that makes the wheels of progress go round!
"Thats the type of fellow thats ruling America to-day; in fact, its the ideal type to which the entire world must tend, if theres to be a decent, well-balanced, Christian, go-ahead future for this little old planet! Once in a while I just naturally sit back and size up this Solid American Citizen, with a whale of a lot of satisfaction.
"Our Ideal CitizenI picture him first and foremost as being busier than a bird-dog, not wasting a lot of good time in day-dreaming or going to sassiety teas or kicking about things that are none of his business, but putting the zip into some store or profession or art. At night he lights up a good cigar, and climbs into the little old bus, and maybe cusses the carburetor, and shoots out home. He mows the lawn, or sneaks in some practice putting, and then hes ready for dinner. After dinner he tells the kiddies a story, or takes the family to the movies, or plays a few fists of bridge, or reads the evening paper, and a chapter or two of some good lively Western novel if he has a taste for literature, and maybe the folks next-door drop in and they sit and visit about their friends and the topics of the day. Then he goes happily to bed, his conscience clear, having contributed his mite to the prosperity of the city and to his own bank-account.
"In politics and religion this Sane Citizen is the canniest man on earth; and in the arts he invariably has a natural taste which makes him pick out the best, every time. In no country in the world will you find so many reproductions of the Old Masters and of well-known paintings on parlor walls as in these United States. No country has anything like our number of phonographs, with not only dance records and comic but also the best operas, such as Verdi, rendered by the worlds highest-paid singers.
"In other countries, art and literature are left to a lot of shabby bums living in attics and feeding on booze and spaghetti, but in America the successful writer or picture-painter is indistinguishable from any other decent business man; and I, for one, am only too glad that the man who has the rare skill to season his message with interesting reading matter and who shows both purpose and pep in handling his literary wares has a chance to drag down his fifty thousand bucks a year, to
Banker, I understand the rating agencies to have claimed that the rating on a structured finance instrument is about the probability of loss of invested principal, not the probability of an issuer failing to make a payment. I do think there's quite a difference there.
Corporates aren't backed by assets. That changes things. It's an important part of the reality that the RAs have to deal with: the corporate bond rating model doesn't so easily jibe with structured finance.
mingle with the biggest executives on terms of perfect equality, and to show as big a house and as swell a car as any Captain of Industry! But, mind you, its the appreciation of the Regular Guy who I have been depicting which has made this possible, and you got to hand as much credit to him as to the authors themselves.
"Finally, but most important, our Standardized Citizen, even if he is a bachelor, is a lover of the Little Ones, a supporter of the hearthstone which is the basic foundation of our civilization, first, last, and all the time, and the thing that most distinguishes us from the decayed nations of Europe.
"I have never yet toured Europeand as a matter of fact, I dont know that I care to such an awful lot, as long as theres our own mighty cities and mountains to be seenbut, the way I figure it out, there must be a good many of our own sort of folks abroad. Indeed, one of the most enthusiastic Rotarians I ever met boosted the tenets of one-hundred-per-cent pep in a burr that smacked o bonny Scutlond and all ye bonny braes o Bobby Burns. But same time, one thing that distinguishes us from our good brothers, the hustlers over there, is that theyre willing to take a lot off the snobs and journalists and politicians, while the modern American business man knows how to talk right up for himself, knows how to make it good and plenty clear that he intends to run the works. He doesnt have to call in some highbrow hired-man when its necessary for him to answer the crooked critics of the sane and efficient life. Hes not dumb, like the old-fashioned merchant. Hes got a vocabulary and a punch."
The above criteria are for residential mortgage backs. The summary reads as follows
Issue ratings are an assessment of default risk, but may incorporate an assessment of relative seniority or ultimate recovery in the event of default. Junior obligations are typically rated lower than senior obligations, to reflect the lower priority in bankruptcy, as noted above.
So it is not an assessment of whether the issuer makes the monthly distributions on the monthly distribution dates. The issuers of an MBS are just passing through collected cash flows. It's not like GM being the issuer of GM bonds.
And it is not just about BK.
The issue is the defaut risk of the obligors on the underlying collateral.
But the RAs say they don't re-underwrite the loans, they just model the cash flows to predict how likely you are to take a principal write-down.
There are a lot of these things out there that are taking write-downs. But the issuers aren't defaulting.
What is the risk the issuer will miss a single interest or principal payment on the date it is due.
But when you do that you are speculating about the future, the future cashflow of the bond issuer. To come up with an accurate rating I would think they would have to do some speculating about the future state of the housing market, the general economy, and even the odds of a cat 5 hurricane hitting Florida.
Considering that the downgradings seem to have comeabout once it was discovered that refinancing and/or selling for a profit were no longer easy options, what documented facts were they working off of when they originally rated them? Appreciating housing prices and low interest rates? Even then you would have to speculate that those trends would continue into the future.
Gah, losing my train of thought. Basically it just seemed like the S&P quote was saying 'we read some papers saying housing was hot, don't blame us for not considering that that might not be the case going forward'.
Bust isn't the problem lie with the investment professional who use the rating agency output as gosphel and substitute the rating with their own judgement? I can form a company and put out a rating on every CDO out there. But no one will care a bit and I won't be able to affect the market in any way, shape or form. So the question is why would S&P and Moody (and may be Fitch) carry so much weight? I think the rating agencies already admitted that they are looking at rear view mirror with existing facts not projecting things..
"But they can and do change -- either as a result of fundamental adjustments to the risk profile of a bond or the emergence of new information"
the biggest advance that allow all these CDO to furish is becuase of the rating system that allow fund manager to be lazy and not doing their job. They may all wake up and start doing their own rating on each individual offering (or pay for a real rating service that will do a good job in rating the offering that they are interest in buying). That should slow down the CDO market substantially. The repricing of the risk that we are going through now is just the first step to move from a rating agency is "god' mentality to we better do our "DD" or paid someone that we trust ro the DD...
Tanta, may be you should start a business with CR and provide those evaluation for pension fund and the like(that is how much a CDO is really worth..).. I bet it should be fun...
I don't disagree that there should be differences about how the ratings are done for corporates verses structureds, only that S&P doesn't seem to differentiate much.
Davenyc,
The RA's claim to run economic cycles and that the ratings, if done right, should stay constant over an economic cycle.
at the end of the day, you can whine at the ratings agencies, but caveat emptor, if you bought something that you didn't understand, you can't blame someone else if it doesn't perform how you wanted. the ratings agencies aren't perfect, but if some moron hedge fund bought a mezz tranche of a CDO with a high composition of sub-prime debt in it, he should have figured out his own view of how it would perform. I would never buy ANY structured product without knowing EXACTLY what is in it and EXACTLY what it takes for that asset to perform/not perform.
I think this applies to structured cash-flows even more so than regular corporate risk, where the ratings agencies typically have more information than the market.
and yes, the ratings agencies will rate structured bonds on the probability of the principal being repaid, which is why you can structure "principal-protected" notes by using a AAA entity to pay back the principal (say using a 10 year zero-coupon bond issued by a AAA issuer, or Treasury strips etc) and then risk the coupons on CDO equity or such like. yep, AAA for getting your principal back, although pretty decent chance you lose all your coupons...
"our recent downgrades affected approximately 1% of the $565.3 billion in first-lien subprime residential mortgage-backed securities (RMBS) that Standard & Poor's rated between the fourth quarter of 2005 and the end of 2006."
Seems like they're cherry-picking their stats here - only looking at first-lien subprime that is less than two years old. As they said, "Ratings are designed to be stable." So, even these ratings didn't quite make it to "stable." I wonder what the numbers would look like if they said how much total RMBS they've rated since the beginning of 2005 has been downgraded.
100ea $100,000 mortgages of differing risks due to differing credit histories, proven or unproven incomes and differing down payments, and differing rates, etc, become a single:
$10,000,000 total principal MBS
Some of the mortgages are 20% down, proven income, 5.5% fixed 30 years
Some are flakey 1.5% for two years, reset to 2% plus prime for remaining 28 years.
And many other mixes.
Who is to say which group of mortgages will produce the best return over say the next ten years. Let alone say what the MBS will return in total, except for some very complicated computer program.
If real estate prices continue to rise and interest rates maybe decline a little, there is a nice profit to be made with little risk on the total MBS.
But if housing prices decline and rates go up, the flakey mortgages produce a loss of principal and a loss of some of the higher income streams.
It seems to this somewhat ignorant observer, that it would not take much to wipe out short term profits, result in panic fire sales, and provide an excellent opportunity for those investment banks that personally know the details of particular MBS to buy back profitably what they once over rated and sold?
One of the interesting aspects of the RAs, for me at least, is the frequency that specific RA agent names from particular agencies are associated with any given entity looking for a rating.
My guess is that the people who pay for ratings liked the S&P was doing ratings. I know little about who pays for ratings but I do know that customers usually get what they want. I'm further guessing risky CDOs got underrated because some institutional investors couldn't hold them if they were rated risker than the ratings S&P gave them.
Or look at this another way, I seriously doubt there will be any upstart rating agencies that gain traction even after this spectacular failure of S&P and Moodys. If their customers wanted something else, they would be able to find it.
Moody's discusses ratings migration from 1920 to 1996 on pp. 7-9 (with migration matrices) of their July 1997 Special Comment, "Moodys Rating Migration and Credit
Quality Correlation, 1920-1996":
I haven't looked deeply, but my recollection is that S&P and Fitch produced similar studies with similar results.
Basically, according to their research -- and I haven't seen anything to the contrary -- a AAA security is much less likely to default, or even suffer a significant downgrade, than a BB security. That's what you're paying them for: not a guarantee of your particular security, but a statistically-based assessment of the likelihood of default by the issuer (which, in the case of a structured finance transaction, is the likelihood that your particular tranche will not pay according to its contractual terms, which almost always means the interest at the stated rate on time or principal at final maturity). Whether the rating of an NIG security should reflect estimated recoveries (loss given default) was a matter of debate in my day; I think it was resolved by creating private ratings, the sense being that public ratings should consider only the likelihood of default.
Of course they've been co-opted to a certain extent: competition breeds rating-shopping and a race to the bottom. It will be interesting to see the ratings migration matrices for originations in the past year or two. As many have said, no sane investor should rely blindly on a rating.
at the end of the day, you can whine at the ratings agencies, but caveat emptor, if you bought something that you didn't understand, you can't blame someone else if it doesn't perform how you wanted.
Right. Caveat emptor.
Which explains why nobody with half a brain is buying this stuff now and won't until there is sufficient transparency to asses the risks.
If the instruments are sold with misrepresentations and intensional obfuscations it then quickly evolves into caveat venditor .
This is hilarious. I was thinking of blogging it but you've done such a good dissection, I have nothing to add.
However, I'd like to see more discussion and analysis on the idea of regulating the rating agencies. NPR briefly broached the topic this morning, drawing a parallel with accounting auditors and Enron. (Inject here slam from Tanta on MSM and Enron parallels).
Anyway if the rating agencies grossly misrepresented the expected performance of securities, then isn't an SEC investigation more appropriate than government writing new rules?
Forgot to mention that the RAs don't do due diligence. They rely on the information they're given. It's a big gap; they recognize it. So they disclose the absence of due diligence upfront. No one is willing to pay them to do due diligence, and I'm not sure they'd want the increased responsibility even if the market would pay for it.
The rating agencies' problem is that they've got no good model for predicting how RMBSs will behave in a down market because RMBSs barely existed (and the "innovative" loans that back them didn't exist at all) in the last down cycle.
Mortgages as financial instruments carry two basic types of risk: prepayment risk and default risk, neither of which is particularly well understood. The rating agencies essentially guessed as to what levels of prepayments we'd see as interest rates fell and the market heated up, and all through the past decade they missed wildly. Alas, that fact seems to have alerted no-one that they also might not have such a good handle on defaults as the market cooled.
When you're talking about a CDO that's backed by residual and net-interest bonds from a dozen different RMB securitizations, it's simply not humanly possible to predict the likely value of the resulting securities in various market conditions. The rating agencies simply assumed that the most senior slice of such a structure should get a high rating. If it was something they'd never seen before, they might insist on fatter mezzanine and "Z" tranches, or even an interest rate hedge, but how big a junior tranche or how big a slice they required might as well have been determined by voodoo.
IMHO, rating agencies would be more reliable if they had some skin in the game--that is, if they took a loss if their rating didn't accurately describe a security's performance and enjoy a gain if it did.
Perhaps the future wave of "ratings" will look something like interest rate hedges, with the agency getting paid if the security performs in a certain range and paying off it it falls outside of that range.
IMHO, rating agencies would be more reliable if they had some skin in the game--that is, if they took a loss if their rating didn't accurately describe a security's performance and enjoy a gain if it did.
The RAs had tons of skin in the game... they got lucrative consulting contracts & fees on how to 'construct' offerings from the very same securitizers who were then going to place the offering... All with the unspoken quid pro quo that the RAs then give those offerings high ratings.
The only way RAs will worry about performance would be if the buyers (not the sellers) paid the bill. Banker suggested that a while back - couple weeks though it already seems like years ago.
Having skin in the game and realizing a profit from the game are two different things. In a Vegas poker game, the dealer gets paid regardless of who wins. That's the position the rating agencies are currently in. I'm suggesting they should bet with the bondholders--like a side bettor on a blackjack table. That would give them an incentive to be more fastidious about counting the cards.
I don't know why you think structured products don't work with corporate rating model - if anything, the model better suits structured products. There's a heck of a lot more judgment involved in rating debentures (note that they are still backed by the company's assets) or even corporate-sponsored asset-backed securities.
Nor do I think most ratings were particularly awful - they are gamable and most rating agencies are fully aware of that, but given the extent to which most investment banks fought to get higher ratings, I think rating agencies held their ground reasonably well (and in many cases, kept things from getting much worse).
And credit ratings are supposed to change over time - if not, they are not doing their job. Default probability changes, expected loss severity changes and given that the housing market lately has been close to the bottom end of their forecast range, it's not suprising that most rating changes are downgrades.
Bullshiet
For those who would say "yes," it is important to remember that unlike many others in the capital markets -- who can take action on the basis of speculation -- we base our opinions on documented facts.
Right, so if a tree fell in a forest, but no one was around to hear, S&P would still give an AAA rating?
A bond's rating is supposed to be an indication of its future performance, right? So perhaps a little speculation on what the future will bring is called for.
Often an agent hanging around the real-estate office takes an application and forwards it to a mortgage company, which may or may not be his employer. The mortgage company processes the loan, and sells it to an investment bank, or similar entity, that bundles mortgages into bonds. The investment bank sells those securities to hedge funds, pension funds, mutual funds and other investors. As mortgages vary significantly in quality, mortgage-backed bonds are rated by Standard and Poor's and other rating agencies.
At each step along the way, information can be lost and temptations build up to understate the risk of default. The agent gets a big fee for writing the loan only if it is approved, so he puts the prospective homeowner in the best possible light and finds an appraiser who aggressively values homes.
Mortgage companies get their cut from origination fees and are able to push off the risk of default on to the ultimate purchasers of the bonds. Hence they are inclined to be lenient with agents.
Investment banks earn money on the spread between the interest rate charged borrowers and the interest rate on the bonds. The less risky bundles of mortgages that go into bonds appear, the lower the interest rates on the bonds and the more profits the mortgage bankers receive.
As we peel this onion, we are finding many purposeful compromises that look much like the insidious corruption that characterizes commerce in some countries. In essence, Wall Street manufactures bad bonds.
But it gets worse. Large builders established mortgage-writing subsidiaries. When they built more houses than the market could absorb, those subsidiaries exaggerated the incomes and qualifications of buyers on loan applications, and pressed for exaggerated appraisals of their properties to move houses. Those entities operated on lines of credit and resold the loans to investment banks, which bundled them into bonds. Both the large builders and investment bankers had incentive to put lipstick on pigs.
Also, investment banks sought and received collaboration from bond-rating agencies in bundling mortgages of differing quality into bonds. In the process, bond raters such as Standard and Poor's were compromised.
Subprime mortgages are hardly the whole credit market, but the meltdown of their bonds cast a spotlight on the decaying integrity of investment banks and bond-rating agencies. These institutions underwrite and rate all manner of credit, and if they could be corrupted in the subprime-mortgage market, then all commercial paper and bonds become suspect.
Over the past several weeks, creditors have increasingly sensed they can't trust banks or bond-rating agencies, and they have fled to short-term Treasury securities. This was much worse than the collapse of mortgage companies that originated housing loans, because it caused all segments of the credit market to collapse.
Good businesses with sound cash flows couldn't borrow operating capital,
"As part of the ratings process, we do engage in open dialogue with bond issuers. This dialogue helps issuers understand our ratings criteria and helps us understand the securities they are structuring, so we can make informed opinions about creditworthiness. We strive to make sure issuers and investors are fully aware of how we determine creditworthiness and believe that all parties are better served when the process is open and transparent."
I think the translation is: "When we give a rating to a structure lower than that required by the structerer, we explain precisely to him how and why we gave that rating, so he can perturb the structure to get the rating that he wants, thereby allowing him to construct a structure with the desired rating with the maximum possible risk, which is usually several times more than the risk which the rating would indicate."
"This dialogue helps issuers understand our ratings criteria and helps us understand the securities they are structuring, so we can make informed opinions about creditworthiness. "
System gamed. Is there any more need for explanation or need for S&P to exist? GIGO- what comes in is sugarcoated garbage, and what goes out is the same.
Caveat Emptor. That seems to have been forgotten with a ratings stamp issued by these jokers. I fail to see how anyone with a fiduciary responsibility can maintain any benefit in purchasing securities with a given "rating" over any other security, given they can't be objectively rated.
Ah well, they will soon cease to exist if this persists.
Someday this war's gonna end...
In essence, Wall Street manufactures bad bonds.
But that's the great thing about all this. No one person ever did so much that they feel like they did something wrong. Each link in the chain didn't break the rules, they just bent them a little. It's an example of 'the solution to polution is dilution'/'diffusion of responsibility'. It's not exactly any one person's fault, so it's nobody's fault.
The whole housing bubble is my fault. I apologize profusely. Now that is settled, let's move on to the future ramifications.
When your paycheck depends on you not fulfilling your full obligation to the market and investors, it should not be a shock that all rating agencies are pretty much obsolete, in this context, and provide a bias rating. Tanta, I agree 100%. My stomach is turning listening to their garbage. If rating agencies are not there to provide an unbiased opinion of the deal and we are not to use their opinions to make any investment decisions, then what in sweet Christmas are they their for. I guess to look pretty and accomplish nothing and be useless. Offcourse, this is all in the context of securitization.
Davenyc,
A bond's rating is supposed to be an indication of its future performance, right?
Remember how limited a bond rating is. It makes a judgement about one thing only. What is the risk the issuer will miss a single interest or principal payment on the date it is due. That's it.
a,
I think the translation is: "When we give a rating to a structure lower than that required by the structerer, we explain precisely to him how and why we gave that rating, so he can perturb the structure to get the rating that he wants, thereby allowing him to construct a structure with the desired rating with the maximum possible risk, which is usually several times more than the risk which the rating would indicate."
Except for the very last clause, that is right on the money.
UPDATE 2-Federal Home Loan bank mortgage 'advances' surge
| Reuters
"Federal Home Loan bank mortgage 'advances' surge"
the more they try to defend themselves, the worse their reputation gets...
moody is hosting a call on SIV ratings tomorry...fun!
Remember how limited a bond rating is. It makes a judgement about one thing only. What is the risk the issuer will miss a single interest or principal payment on the date it is due. That's it.
With corporate bonds, yes. But with structured cash flows? Really?
Then why has anything been downgraded? Has anyone heard of a bond that actually defaulted yet?
Tanta,
The link that follows is OT, but I wonder if it's worth a separate post? It's an editorial from Prudent Bear's website by Martin Hutchinson titled "Back to Jimmy Stewart!"
404 - Error: 404
I think this excerpt sums up how (he believes) we arrived at our current inefficient mortgage market. Note: Jimmy Stewart, the small-town lender from It's a Wonderful Life, represents old-style home lending:
"The answer is one word: salesmanship. Anybody who has lived in a US suburb with an economically attractive zipcode and no butler will recognize that excessive salesmanship is the bane of American life. This is even more true in the mortgage business. Homeowners today dont go into their local S&L, save for half a decade and request a mortgage from Jimmy Stewart. Instead they are sold a mortgage product, either directly or over the Internet, by an aggressive salesman who is paid a multiple of what Jimmy Stewart earned, or at least aspires to be. That product is then securitized by an investment bank trader who in good years is paid a LARGE multiple of what Jimmy Stewart earned. With others it is sold to a securitization vehicle of immense complexity that has been set up by Wall Street lawyers paid a HUGE multiple of what Jimmy Stewart earned. Costs have been increased at every point in the process, but aggressive salesmanship has driven Jimmy Stewart out of business."
Tanta,
Yup, same standard. Things have been downgraded because the RA's believe the likelihood of a payment default has increased. I haven't heard of an actual default yet, but you could find them under the D rating. Let me do some digging.
Do I understand them correctly?
"We only underestimated the riskiness of the most risky sh**.(Which was the reason you paid us for our opinion in the first place.)
But we nailed the risk evaluation of the stuff that everyone knew was not risky".
-- Hiding Out
OK, so here's S&P' criteria
http://www2.standardandpoors.com/portal/site/sp/en/us/page.article/2,1,9,4,1148445953418.html
Babbitt has Jimmy Stewart beat every time.
"[L]et me tell you right here and now, I wouldnt trade a high-class Zenith acreage development for the whole length and breadth of Broadway or State Street! . . . its evident to any one with a head for facts that Zenith is the finest example of American life and prosperity to be found anywhere.
"I dont mean to say were perfect. Weve got a lot to do in the way of extending the paving of motor boulevards, for, believe me, its the fellow with four to ten thousand a year, say, and an automobile and a nice little family in a bungalow on the edge of town, that makes the wheels of progress go round!
"Thats the type of fellow thats ruling America to-day; in fact, its the ideal type to which the entire world must tend, if theres to be a decent, well-balanced, Christian, go-ahead future for this little old planet! Once in a while I just naturally sit back and size up this Solid American Citizen, with a whale of a lot of satisfaction.
"Our Ideal CitizenI picture him first and foremost as being busier than a bird-dog, not wasting a lot of good time in day-dreaming or going to sassiety teas or kicking about things that are none of his business, but putting the zip into some store or profession or art. At night he lights up a good cigar, and climbs into the little old bus, and maybe cusses the carburetor, and shoots out home. He mows the lawn, or sneaks in some practice putting, and then hes ready for dinner. After dinner he tells the kiddies a story, or takes the family to the movies, or plays a few fists of bridge, or reads the evening paper, and a chapter or two of some good lively Western novel if he has a taste for literature, and maybe the folks next-door drop in and they sit and visit about their friends and the topics of the day. Then he goes happily to bed, his conscience clear, having contributed his mite to the prosperity of the city and to his own bank-account.
"In politics and religion this Sane Citizen is the canniest man on earth; and in the arts he invariably has a natural taste which makes him pick out the best, every time. In no country in the world will you find so many reproductions of the Old Masters and of well-known paintings on parlor walls as in these United States. No country has anything like our number of phonographs, with not only dance records and comic but also the best operas, such as Verdi, rendered by the worlds highest-paid singers.
"In other countries, art and literature are left to a lot of shabby bums living in attics and feeding on booze and spaghetti, but in America the successful writer or picture-painter is indistinguishable from any other decent business man; and I, for one, am only too glad that the man who has the rare skill to season his message with interesting reading matter and who shows both purpose and pep in handling his literary wares has a chance to drag down his fifty thousand bucks a year, to
"you not guilty plea has been filed and we shall proceed to trial. . . " Judge Judy.
Banker, I understand the rating agencies to have claimed that the rating on a structured finance instrument is about the probability of loss of invested principal, not the probability of an issuer failing to make a payment. I do think there's quite a difference there.
Corporates aren't backed by assets. That changes things. It's an important part of the reality that the RAs have to deal with: the corporate bond rating model doesn't so easily jibe with structured finance.
mingle with the biggest executives on terms of perfect equality, and to show as big a house and as swell a car as any Captain of Industry! But, mind you, its the appreciation of the Regular Guy who I have been depicting which has made this possible, and you got to hand as much credit to him as to the authors themselves.
"Finally, but most important, our Standardized Citizen, even if he is a bachelor, is a lover of the Little Ones, a supporter of the hearthstone which is the basic foundation of our civilization, first, last, and all the time, and the thing that most distinguishes us from the decayed nations of Europe.
"I have never yet toured Europeand as a matter of fact, I dont know that I care to such an awful lot, as long as theres our own mighty cities and mountains to be seenbut, the way I figure it out, there must be a good many of our own sort of folks abroad. Indeed, one of the most enthusiastic Rotarians I ever met boosted the tenets of one-hundred-per-cent pep in a burr that smacked o bonny Scutlond and all ye bonny braes o Bobby Burns. But same time, one thing that distinguishes us from our good brothers, the hustlers over there, is that theyre willing to take a lot off the snobs and journalists and politicians, while the modern American business man knows how to talk right up for himself, knows how to make it good and plenty clear that he intends to run the works. He doesnt have to call in some highbrow hired-man when its necessary for him to answer the crooked critics of the sane and efficient life. Hes not dumb, like the old-fashioned merchant. Hes got a vocabulary and a punch."
Remind you of anybody?
Tanta,
The above criteria are for residential mortgage backs. The summary reads as follows
Issue ratings are an assessment of default risk, but may incorporate an assessment of relative seniority or ultimate recovery in the event of default. Junior obligations are typically rated lower than senior obligations, to reflect the lower priority in bankruptcy, as noted above.
Right, Banker.
So it is not an assessment of whether the issuer makes the monthly distributions on the monthly distribution dates. The issuers of an MBS are just passing through collected cash flows. It's not like GM being the issuer of GM bonds.
And it is not just about BK.
The issue is the defaut risk of the obligors on the underlying collateral.
But the RAs say they don't re-underwrite the loans, they just model the cash flows to predict how likely you are to take a principal write-down.
There are a lot of these things out there that are taking write-downs. But the issuers aren't defaulting.
My argument, Banker, is that a corporate bond is more like a mortgage loan, conceptually, than it is like a bond backed by mortgage loans.
What is the risk the issuer will miss a single interest or principal payment on the date it is due.
But when you do that you are speculating about the future, the future cashflow of the bond issuer. To come up with an accurate rating I would think they would have to do some speculating about the future state of the housing market, the general economy, and even the odds of a cat 5 hurricane hitting Florida.
Considering that the downgradings seem to have comeabout once it was discovered that refinancing and/or selling for a profit were no longer easy options, what documented facts were they working off of when they originally rated them? Appreciating housing prices and low interest rates? Even then you would have to speculate that those trends would continue into the future.
Gah, losing my train of thought. Basically it just seemed like the S&P quote was saying 'we read some papers saying housing was hot, don't blame us for not considering that that might not be the case going forward'.
Bust isn't the problem lie with the investment professional who use the rating agency output as gosphel and substitute the rating with their own judgement? I can form a company and put out a rating on every CDO out there. But no one will care a bit and I won't be able to affect the market in any way, shape or form. So the question is why would S&P and Moody (and may be Fitch) carry so much weight? I think the rating agencies already admitted that they are looking at rear view mirror with existing facts not projecting things..
"But they can and do change -- either as a result of fundamental adjustments to the risk profile of a bond or the emergence of new information"
the biggest advance that allow all these CDO to furish is becuase of the rating system that allow fund manager to be lazy and not doing their job. They may all wake up and start doing their own rating on each individual offering (or pay for a real rating service that will do a good job in rating the offering that they are interest in buying). That should slow down the CDO market substantially. The repricing of the risk that we are going through now is just the first step to move from a rating agency is "god' mentality to we better do our "DD" or paid someone that we trust ro the DD...
Tanta, may be you should start a business with CR and provide those evaluation for pension fund and the like(that is how much a CDO is really worth..).. I bet it should be fun...
FT: heart attack markets are in for a test...
FT.com / Capital Markets - ‘Heart attack’ markets face test
Tanta,
I don't disagree that there should be differences about how the ratings are done for corporates verses structureds, only that S&P doesn't seem to differentiate much.
Davenyc,
The RA's claim to run economic cycles and that the ratings, if done right, should stay constant over an economic cycle.
You should start a service that rates an RA's chance of changing its ratings. I wonder how Moody's would compare to S&P...
...but that was in another country; And besides, the wench is dead.
at the end of the day, you can whine at the ratings agencies, but caveat emptor, if you bought something that you didn't understand, you can't blame someone else if it doesn't perform how you wanted. the ratings agencies aren't perfect, but if some moron hedge fund bought a mezz tranche of a CDO with a high composition of sub-prime debt in it, he should have figured out his own view of how it would perform. I would never buy ANY structured product without knowing EXACTLY what is in it and EXACTLY what it takes for that asset to perform/not perform.
I think this applies to structured cash-flows even more so than regular corporate risk, where the ratings agencies typically have more information than the market.
and yes, the ratings agencies will rate structured bonds on the probability of the principal being repaid, which is why you can structure "principal-protected" notes by using a AAA entity to pay back the principal (say using a 10 year zero-coupon bond issued by a AAA issuer, or Treasury strips etc) and then risk the coupons on CDO equity or such like. yep, AAA for getting your principal back, although pretty decent chance you lose all your coupons...
Banker, so I exaggerated slightly in the last clause, shoot me.
"our recent downgrades affected approximately 1% of the $565.3 billion in first-lien subprime residential mortgage-backed securities (RMBS) that Standard & Poor's rated between the fourth quarter of 2005 and the end of 2006."
Seems like they're cherry-picking their stats here - only looking at first-lien subprime that is less than two years old. As they said, "Ratings are designed to be stable." So, even these ratings didn't quite make it to "stable." I wonder what the numbers would look like if they said how much total RMBS they've rated since the beginning of 2005 has been downgraded.
Trying to keep it simple but make a point.
100ea $100,000 mortgages of differing risks due to differing credit histories, proven or unproven incomes and differing down payments, and differing rates, etc, become a single:
$10,000,000 total principal MBS
Some of the mortgages are 20% down, proven income, 5.5% fixed 30 years
Some are flakey 1.5% for two years, reset to 2% plus prime for remaining 28 years.
And many other mixes.
Who is to say which group of mortgages will produce the best return over say the next ten years. Let alone say what the MBS will return in total, except for some very complicated computer program.
If real estate prices continue to rise and interest rates maybe decline a little, there is a nice profit to be made with little risk on the total MBS.
But if housing prices decline and rates go up, the flakey mortgages produce a loss of principal and a loss of some of the higher income streams.
It seems to this somewhat ignorant observer, that it would not take much to wipe out short term profits, result in panic fire sales, and provide an excellent opportunity for those investment banks that personally know the details of particular MBS to buy back profitably what they once over rated and sold?
Totally pathetic. Sounds like the lawyers wrote the article...
One of the interesting aspects of the RAs, for me at least, is the frequency that specific RA agent names from particular agencies are associated with any given entity looking for a rating.
My guess is that the people who pay for ratings liked the S&P was doing ratings. I know little about who pays for ratings but I do know that customers usually get what they want. I'm further guessing risky CDOs got underrated because some institutional investors couldn't hold them if they were rated risker than the ratings S&P gave them.
Or look at this another way, I seriously doubt there will be any upstart rating agencies that gain traction even after this spectacular failure of S&P and Moodys. If their customers wanted something else, they would be able to find it.
Moody's discusses ratings migration from 1920 to 1996 on pp. 7-9 (with migration matrices) of their July 1997 Special Comment, "Moodys Rating Migration and Credit
Quality Correlation, 1920-1996":
http://www.moodyskmv.com/research/files/wp/25097.pdf
I haven't looked deeply, but my recollection is that S&P and Fitch produced similar studies with similar results.
Basically, according to their research -- and I haven't seen anything to the contrary -- a AAA security is much less likely to default, or even suffer a significant downgrade, than a BB security. That's what you're paying them for: not a guarantee of your particular security, but a statistically-based assessment of the likelihood of default by the issuer (which, in the case of a structured finance transaction, is the likelihood that your particular tranche will not pay according to its contractual terms, which almost always means the interest at the stated rate on time or principal at final maturity). Whether the rating of an NIG security should reflect estimated recoveries (loss given default) was a matter of debate in my day; I think it was resolved by creating private ratings, the sense being that public ratings should consider only the likelihood of default.
Of course they've been co-opted to a certain extent: competition breeds rating-shopping and a race to the bottom. It will be interesting to see the ratings migration matrices for originations in the past year or two. As many have said, no sane investor should rely blindly on a rating.
If you want a guarantee, pay a bond insurer.
at the end of the day, you can whine at the ratings agencies, but caveat emptor, if you bought something that you didn't understand, you can't blame someone else if it doesn't perform how you wanted.
Right. Caveat emptor.
Which explains why nobody with half a brain is buying this stuff now and won't until there is sufficient transparency to asses the risks.
If the instruments are sold with misrepresentations and intensional obfuscations it then quickly evolves into caveat venditor .
Neither worries are good for a healthy market.
This is hilarious. I was thinking of blogging it but you've done such a good dissection, I have nothing to add.
However, I'd like to see more discussion and analysis on the idea of regulating the rating agencies. NPR briefly broached the topic this morning, drawing a parallel with accounting auditors and Enron. (Inject here slam from Tanta on MSM and Enron parallels).
Anyway if the rating agencies grossly misrepresented the expected performance of securities, then isn't an SEC investigation more appropriate than government writing new rules?
Forgot to mention that the RAs don't do due diligence. They rely on the information they're given. It's a big gap; they recognize it. So they disclose the absence of due diligence upfront. No one is willing to pay them to do due diligence, and I'm not sure they'd want the increased responsibility even if the market would pay for it.
ayn rand is alive and well.
we are not waiting for gadot, rather, john galt.
Damn,it was the butler,after all.and moody's,fitch and s&p all looked so guilty.
we are not waiting for gadot, rather, john galt.
Waiting for John Galt? What a comical paradox... lol. I guess one could say that's proof Ayn Rand never lived.
The rating agencies' problem is that they've got no good model for predicting how RMBSs will behave in a down market because RMBSs barely existed (and the "innovative" loans that back them didn't exist at all) in the last down cycle.
Mortgages as financial instruments carry two basic types of risk: prepayment risk and default risk, neither of which is particularly well understood. The rating agencies essentially guessed as to what levels of prepayments we'd see as interest rates fell and the market heated up, and all through the past decade they missed wildly. Alas, that fact seems to have alerted no-one that they also might not have such a good handle on defaults as the market cooled.
When you're talking about a CDO that's backed by residual and net-interest bonds from a dozen different RMB securitizations, it's simply not humanly possible to predict the likely value of the resulting securities in various market conditions. The rating agencies simply assumed that the most senior slice of such a structure should get a high rating. If it was something they'd never seen before, they might insist on fatter mezzanine and "Z" tranches, or even an interest rate hedge, but how big a junior tranche or how big a slice they required might as well have been determined by voodoo.
IMHO, rating agencies would be more reliable if they had some skin in the game--that is, if they took a loss if their rating didn't accurately describe a security's performance and enjoy a gain if it did.
Perhaps the future wave of "ratings" will look something like interest rate hedges, with the agency getting paid if the security performs in a certain range and paying off it it falls outside of that range.
IMHO, rating agencies would be more reliable if they had some skin in the game--that is, if they took a loss if their rating didn't accurately describe a security's performance and enjoy a gain if it did.
The RAs had tons of skin in the game... they got lucrative consulting contracts & fees on how to 'construct' offerings from the very same securitizers who were then going to place the offering... All with the unspoken quid pro quo that the RAs then give those offerings high ratings.
The only way RAs will worry about performance would be if the buyers (not the sellers) paid the bill. Banker suggested that a while back - couple weeks though it already seems like years ago.
Good luck getting buyers to pony up.
MarketWatch
Merrill slashes banks' profit forecasts - 65% chance of recession
Merrill Lynch trims banks' profit estimates on recession risk - MarketWatch
Having skin in the game and realizing a profit from the game are two different things. In a Vegas poker game, the dealer gets paid regardless of who wins. That's the position the rating agencies are currently in. I'm suggesting they should bet with the bondholders--like a side bettor on a blackjack table. That would give them an incentive to be more fastidious about counting the cards.
Tanta,
I don't know why you think structured products don't work with corporate rating model - if anything, the model better suits structured products. There's a heck of a lot more judgment involved in rating debentures (note that they are still backed by the company's assets) or even corporate-sponsored asset-backed securities.
Nor do I think most ratings were particularly awful - they are gamable and most rating agencies are fully aware of that, but given the extent to which most investment banks fought to get higher ratings, I think rating agencies held their ground reasonably well (and in many cases, kept things from getting much worse).
And credit ratings are supposed to change over time - if not, they are not doing their job. Default probability changes, expected loss severity changes and given that the housing market lately has been close to the bottom end of their forecast range, it's not suprising that most rating changes are downgrades.