American Enterprise Institute

June 24, 2008

[Edited transcript from audio tapes]

12:00 p.m.	Registration and Luncheon
12:30	Panelists:	Erik R. Sirri, Securities and Exchange Commission
		Lawrence J. White, New York University
		Joshua Rosner, Graham Fisher & Co.
		Sean Egan, Egan-Jones Ratings Co.
		Sylvain Raynes, R&R Consulting
	Moderator:	Alex J. Pollock, AEI
2:30	Adjournment

Proceedings:

Alex J. Pollock:  All right, if we can come to order, we’ll begin.  I’d like to welcome you all to our conference on how to improve the credit rating agency sector, and thank you for being with us.  We have an outstanding panel today, whom I’ll introduce in a moment.

The great housing and mortgage bubble of the 21st Century, which peaked in 2006, and the painful and costly deflation of that bubble, which continues, make a classic financial pattern.  This time, financial markets applied vast computer power and complex models with reams of data—including the models, of course, of the credit rating agencies which as it turned out helped inflate the bubble.  And whether the siren song of increasing risk and increasing leverage has turned into calculations with computers or with quill pens, the underlying logic and underlying result is the same. 

A part of the classic bubble and bust pattern is an inevitable political reaction.  And the political debate, which always follow in the wake of the big financial bust, is the search for the guilty.  It is clear that among the candidates for this role of the guilty in the current bust are the major credit rating agencies, but our purpose today is not to pursue the search for the guilty but rather to have a dispassionate consideration of how the role of the credit rating agencies can be improved going forward. 

We have recent SEC proposals in this line and numerous other ideas and recommendations ranging from much heavier and stricter regulation of ratings agencies to removing the government from any role in sponsoring certain credit rating agencies or mandating the use of their ratings as has been the case in the past.  And our panel represents a range of views on these issues. 

We will hear first today from Erik Sirri, the director of the Division of Trading and Markets at the SEC, which is responsible for regulation of stock and option exchanges, national securities associations, brokers, dealers, clearing organizations and credit rating agencies.  Erik is on leave from Babson College, where he is a professor of finance.  He has previously served as chief economist of the SEC, a governor of the Boston Stock Exchange, a member of the regulatory board of the Boston Options Exchange and has also worked on planetary missions for NASA.  And we’re very pleased he’s able to join us today.  Of course, that’s one representative of the famous migration of the rocket scientists into financial affairs.

Our next speaker will be Larry White, who is down at the end by the computer, who is the Arthur E. Imperatore Professor of Economics and deputy chairman of the Economics Department at the Stern School at New York University.  During the last really big housing finance bust in the 1980s, Larry served, and then I thought I should write “survived,” as a member of the Federal Home Loan Bank Board. 

He has also served with the Antitrust Division of the Department of Justice and the staff of the President’s Council of Economic Advisers and is the general editor of the Review of Industrial Organization, and in one sense, of course, our topic today is about industrial organization of a certain sector.  I should say that reading Larry’s articles on credit rating agencies several years ago is what introduced me to the issues that we’re going to be discussing today.

Next will be Josh Rosner, who is by Larry on the other end, managing director of the research consultancy at Graham Fisher & Company, where he advises regulators and institutional investors on housing and mortgage finance issues.  Previously, he was with Medley Global Advisors, where he was among the first to identify accounting problems with the housing GSCs, the mortgage bubble, the likelihood of resulting contagion in the credit markets, and issues in the credit ratings of CDO’s.  Josh called me early in 2007 to say that mortgage securities markets were heading for a serious liquidity crunch, a really good call.

Our next speaker will be Sean Egan, a founding partner and managing director of Egan-Jones Ratings Company, which was registered as an NRSRO by the SEC last year.  Egan-Jones is 100 percent paid by investors and its only clients are institutional investors.  It has provided early credit warnings on Bear Stearns, the Monoline Insurers, Countrywide, New Century, California Utilities, Delphi, Enron and Global Crossing, among others.  Sean was a commercial and investment banker before founding Egan-Jones and, I must say, is never shy about sharing his views on the credit rating agency sector.

Sylvain Raynes, who is right in the middle here, will complete the panel.  Sylvain is a founding principal of the New York-based R&R Consulting where he specializes in securitization and project finance.  He is also a faculty member at the New York Institute of Finance and at Baruch College.  Previously, he worked in structured finance at Paine Webber, risk management at Credit Suisse and was a senior analyst at Moody’s and is the author of Elements of Structured Finance and The Analysis of Structured Securities.

Each of our panelists will speak for 12 to 15 minutes and after that, we will give the panel a chance to respond to each other.  And then we will open the floor for your questions.  We will adjourn promptly at 2:30 or whenever you are out of questions, whichever is first.  And with that, Erik, the floor is yours.

Erik R. Sirri:  All right.  Thank you, Alex.  It is a pleasure to be here.  I have to start with the standard disclaimer that anything I say represents my own views and not the views of any of my colleagues at the commission nor any of the commissioners.

So that said, let me talk a little bit about credit rating agencies.  I am not going to use the name “nationally recognized statistical rating organizations.”  I’ll just say credit rating agencies or CRA, so we don’t go crazy here.

Let me start by saying, talking about how credit rating agencies came into being as, if you will, a special kind of organization.  Anyone, of course, can hang out their shingle and just be a credit rating agency and offer their opinion, but to be an NRSRO, to be recognized by the SEC, has an interesting historical origin.  It began way back 20 plus years ago when the SEC essentially crafted a no-action approach.  What they’ve said is that if a broker dealer wished to use a credit rating as a basis for calculating its haircut for a broker dealer’s net capital purposes, then the SEC would not take action against them for enforcement purposes.  So it granted no action letters to particular credit rating agencies over time and that was the genesis of the Nationally Recognized Statistical Rating Organization or NRSRO, sort of, moniker.

So, fast forward to the future, in 2006, Congress passed the Credit Rating Agency Reform Act.  I think that came after a number of years when Congress was interested in getting more competition in this space, getting entry into the space and if you read the legislation, it’s very clear in my opinion what it’s looking for.  It’s looking for more competition, the rise in the space and it’s looking for really market-based solutions to the question of the role of credit rating agencies. 

If the legislation, which is what’s powering our rule-making in this area -- we did rule-making in June of ’07, we did rule-making two weeks ago, and we’ll do some more rule-making tomorrow.  The legislation is very interesting.  It’s simple and that it’s clear.  It’s not hard to read and it lays out in particular, it’s very narrowly tailored, what it has us do.  It says that we shall craft rules to lessen the effects of conflicts of interest in the space, to increase transparency in the space, to resolve certain competitive issues in the space.

But what’s in some ways most interesting about this legislation, in my view, is that it absolutely forbids us from regulating the substance, process or methods by which credit ratings are calculated.  So, that’s notable.  I think it tells us what we should do, what we shouldn’t do and we’ve been working very diligently to operationalize those.

So, what have we done?  Where have we gone with this?  Well, when the rules came into force, it was essentially June 2007, so we registered up seven credit rating agencies.  There had been five before that.  Two other agencies entered in that interim period, two Japanese credit rating agencies.  Since then, we’ve had three other credit rating agencies register up as NRSRO’s, so we now have 10.  The last one came in just a couple of days ago, so we now have 10 of these recognized credit rating agencies. 

Just so you know, these credit rating agencies can achieve their status narrowly or broadly.  We divide the assets base into five groups, so you could be like some of the larger credit rating agencies that essentially get certified across broad classes of assets: corporates, munis, structured, and so on, or you just might pick one: municipals and become an NRSRO in that one space.  Some of the newer entrants have seen fit to just pick their spot and just become an NRSRO in that one space. 

So, we crafted, essentially, enabling rules back in June of 2007.  They led to the registration of a group of credit rating agencies and they led to a basic set of rules about how credit rating agencies should operate, what kind of information they should provide, and really some rules about basic conflicts of interest.  If you own the security, you should not rate the security, those kinds of various straightforward things.  We prohibited certain conflicts of interest and for other kinds of conflicts of interest, we said you should have policies and procedures to manage those conflicts of interest.

Two weeks ago, as I said, we crafted another round of rules.  They are proposed rules, so they have yet to be adopted.  They are out for comment.  So to the extent that maybe you have comments, we’d love to hear them.  Those rules also fall into the same broad two buckets.  Some deal with conflicts of interest, some deal with transparencies.  

I’m not going to go through all of them, but to just mention a couple so you get a sense of the flavor with respect to conflicts of interest.  For example, we prohibited credit rating agencies from structuring products and rating those same products they structured, fairly straightforward.  We prohibited firms from-- we went after the practice of having the individuals who negotiate pricing for credit ratings be involved in the structuring of the rating of the security, again, a basic conflict of interest.  On the transparency side, we asked for more specificity and greater transparency around the performance of credit ratings, so we worked there.  We asked for more information about the frequency with which ratings are reviewed, the so-called -- the process and the resources laid down to the surveillance process, so more transparency there. 

And finally, two weeks ago, when we did this, we proposed something about the symbols that are used for credit ratings.  It’s a little bit complex but it essentially reduces down to a proposal to require credit rating agencies to use different symbols for structured products.  They technically have a choice.  They could produce a report about those structured products or if they choose not to produce a detailed report when they rate, they could elect to use different symbols.  They could use different characters, they could append a suffix, whatever it is.  All of those are out for comment. 

If you read the paper today, you saw an article, I think it was in the Wall Street Journal, that discussed what we may be doing tomorrow, we’ve been fairly public about this.  We talked about -- then the chairman has spoken, I’ve spoken about the need to reduce the securities laws, reliance on the NRSRO designation and credit ratings within our own securities laws.  And what we’ve thought useful to do, and again these will be proposed, the Commission will vote on them tomorrow, so I can’t speak very definitely about them. 

But it’s probably fair to say very generally that we’re seeking to reduce reliance, so we’re asking the question, does it make sense instead of using a credit rating in a particular place, in a rule, to instead replace that by something that refers to, say, the characteristic for which the credit rating was proxying: liquidity, volatility, probability of loss, those sorts of things?  Can we find a way to replace the credit rating with something along those lines?  So, we’re going to be exploring that tomorrow.

Let me close by just talking about the other work that we’ve been doing, so you have some context.  Besides what I’ve told you about now, in our rule-making efforts, we have examination authority over the credit rating agencies.  So we’ve been going into the three largest credit rating agencies that rate structured product: S&P, Moody’s, and Fitch, and examining their processes and their procedures and such over CDO’s and mortgage-backed securities. 

We expect those exams to finish up in the coming month or so and that process has informed the current rule-making that we’ve done.  We’re not completely done with those exams, but nonetheless, we’ve learned a number of things from that.  And I think that’s been a useful process that will be continuing.  We will always have examination authority over these, so that will be an ongoing set of procedures we have.  I think it’s useful because it will inform our rule-making, I think, in an intelligent way.

I would say, before I leave, before I stop talking, I would like to point out one thing that I think is particularly difficult here, and I think it comports with the legislation.  As I think about the legislation, what it sought to attain, it really looked, as I said, at market-based solutions and in many ways that requires investors to lift their share of the load here.  Amongst the puzzles that we’ve seen when we go out into this space, we talk to investors, we ask them, how are you using credit ratings, how are you relying on these credit ratings? 

And, of course, you get a wide range of answers.  It depends on the investor to whom you speak.  If you speak to large institution investors that have a raft of credit rating analysts, you get different answers, because within that pool of credit rating analysts, you will have people who used to work for the largest ratings organizations and they will understand, exactly, the internal procedures. 

So, we’ve been informed that -- we have also talked to other types of investors, smaller investors, advisers, middle-size investors, each of which will give you a different set of answers but I will tell you that amongst the puzzles that we all see here, at least I’m speaking for myself, that I see here, is a way to craft the solution that causes individual investors to take more cognitive burden, more responsibility for the investment decision on their own. 

Reducing reliance on credit rating agencies within the securities laws is a step, but is not the only thing that has to be done.  In the end, there has got to be a lot of people working together to reach that solution.  You can naturally ask the question, I think, why is it that credit ratings don’t improve themselves, why doesn’t competition solve the key issue here? 

And we spend a lot of time thinking about that.  Let me try and explain that to you this way, at least, in my view, in the structured space.  Usually, most products and most products bases, the best product in the end gains market share.  That’s the trick and it works for peanut butter, it works for bleach, it works for all these things.  Quality intends, over the long run, to cause you to sell more of your product. 

In the credit rating space, that’s a particularly difficult issue.  We could have a discussion of what does it mean to have a high-quality rating.  We might agree for the moment that it may mean something like you’re accurate.  That is, that if you rate something triple A, it has a much, much lower probability of default than you rate it triple C and so on.  That it accurately comports with real default probabilities.  The question is, if that’s how you define a better product, is that how people purchase?  Is that the basis by which people purchase credit ratings? 

For the most part, in an issuer-pay model, people do not purchase credit ratings according to their quality.  They purchase them with an eye toward their commercial interests and the commercial interests are, of course, in selling or distributing the product they have.  So, to make the fine point, if there are two entities, one of which delivers a high credit rating to a product and another entity that has a better mouse trap and that delivers a medium credit rating to that product, the lower credit rating. 

Then it comes to the issuer to decide which one to buy and let’s say, everybody understands that the medium credit rating is more accurate, let’s say there’s an objective basis for this, there will be a tendency for the issuer to buy the higher credit rating, to pay the credit rating agency to grant that.  That is what it is, commercially, that we understand why that happens. 

But of course, two things occur.  That, one, you have to ask, well, there’s not a mechanism for normal competitive forces to work in that market.  That’s a bit of a puzzle, and as regulators standing in this, we spend a lot of time thinking, what can we do with our rule-making authority to cause competitive forces to be more effective.  I think us asking that question is exactly what Congress wants us to do and we’ve thought very carefully about that.  That’s reflected in the rules that we have done and in the rules that we have not done on both sides. 

The second thing that it points to is if this is the case, why isn’t it that investors are paying more attention to the quality of ratings ex post?  Why aren’t they asking those same probing questions?  Why is it up to us or others to come in and say, hey, tell me about these resources you’re dedicating to surveillance, tell me how often you update those rating, tell me what happens when you make a change to your rating models, do you backtrack it into your surveillance, and does it hit those already cast ratings or not?  Those are issues that we are, of course, asking, that we’re writing rules about and asking for comment on, but I firmly believe these are issues that investors should be diligently working on as well.

So, I’ll stop there and turn it over to the next speaker.

Alex J. Pollock:  Thanks very much.  Larry.

Lawrence J. White:  All right.  I am not going to give you -- Sean is going to give you the detailed critique of -- I must confess--

Alex J. Pollock:  That that comes very close to a 2-minute penalty at AEI.

Lawrence J. White:  I couldn’t resist, I just couldn’t resist.

The other thing, if you haven’t read this morning’s Wall Street Journal -- Disclaimer, I own no stock in the Wall Street and Dow Jones, except as part of a diversified, I guess it’s now News Corp., portfolio.  None of my relatives, to my knowledge, work for the Wall Street Journal or any of its relatives.  But you really ought to look at this story in Section C over here.  In fact, one of the authors of the story, Erin [indiscernible] is here, and it’s quite an interesting story.

All right.  What I wanted to do is you’ve got a copy of my slides and so I don’t have to spend a lot of time.  As a good business school professor, I’ve got to tell you what I’m going to say, then I’m going to say it and then I’m going to tell you what I said.  Also as a good New Yorker, I have to speak quickly.

All right.  Why bond-rating firms?  Because they are part of the efforts of any credit market anywhere to deal with what economists call the problem of asymmetric information.  In plainer terms, the fundamental issue that any lender always has to ask is:  Am I going to get paid back or not?  And who is going to more likely, who is less likely to pay me back and I want to adjust the terms of my lending or perhaps not lend at all, and so, having somebody offer judgment, opinion can be valuable in that work.

All right.  We know a lot about the market structure already.  I don’t want to spend a lot of time on this.  We have these three dominant firms.  This is not going to be the wheat market.  We wouldn’t expect thousands or tens of thousands on --  there are economies of scale, there are network effects in the economics language but surely the SEC regulation that Erik mentioned -- that goes back to 1975 -- has exacerbated the dominance of these three. 

Let’s remind ourselves that prior to the early 1970’s, the basic model was an investor-pay model.  That’s what John Moody basically established back in 1909 and it changed in the early 1970’s to an issuer-pay model.  In the corporate world, you know the corporate guys have gotten very accustomed to this, almost everybody would pay, they didn’t have to but they would, and all of the structured financed ratings are solicited [sounds like]. 

Why did they change the model?  Best I can tell, it was a combination of the fear of the photocopying machine and the realization that issuers needed ratings in order to get their bonds into the portfolio of regulated entities.  I’ll talk about that a bit more in just a minute.  And that’s I think what drove the change.  Sean will tell you what his firm does and there are other smaller firms who still manage to make a living by selling their ratings to investors. 

The potential conflicts, and I would argue, actual conflicts in the issuer-pay model are manifest and the ratings firms themselves have acknowledged the potential for the conflicts.  They would tell you that they wouldn’t want to risk their reputations.  They manage around those conflicts.  Their long-run reputations are the most important things to them, but of course, that’s what Arthur Andersen would have told you in 2001.  Again, there’s the early origins, and as of the 1920’s, there’s no question these guys were meeting a market test. 

How did we get to where we are today?  It all started in 1931, when the Office of the Comptroller of the Currency, the OCC, the regulator of national banks, tells banks that if they have bonds in their portfolios that are below investment grade, they’re going to have to be marked to market.  If they are at investment grade or above, then they can hold them at original purchase price depreciated.  Whose ratings, whose determination of investment grade?  Well, the comptroller was sort of vague, talked about recognized ratings manuals.  In ’36, the OCC went further and basically told banks, and this is still true for all depository institutions in the year 2008, that they could not hold in their portfolios a bond that was below investment grade.  The other financial regulators followed in the ‘30’s and the ‘40’s, insurance regulators, other financial regulators, and it’s important to realize what was being done. 

In essence, the financial regulators -- as Alex mentioned, I come from a bank regulatory background and so I understand the idea of safety and soundness.  I’m a big believer in safety and soundness for depository institutions by telling banks in 1989, S&Ls were told the same thing, that they had to look to bond ratings for a determination of what was safe to hold in a portfolio, what could not be held.  In essence, the regulator had outsourced a safety decision.  They had delegated that decision to third parties to this up until 1975, a vaguely defined group of raters.  In ’75, as Erik indicated, the SEC wanted to establish capital requirements for broker dealers, wanted those capital requirements to be sensitive to the quality of the assets, wanted, essentially, to use ratings as an indicator of the quality, again, outsourcing that decision. 

To its credit, the SEC realized that the “whose” rating issue had really not been well defined up to that point.  And the problem was what happens if a bogus rating company comes along and starts handing out triple A ratings to anybody who puts enough money in the pocket of that rating firm.  The SEC wanted to try to deal with that problem and so created the NRSRO category, sort of formalizing this outsourcing process.  The SEC designated a few more NRSRO’s over the next 25 years, but by the year 2000, they had all merged into Fitch, and so as of the year 2000, there were only three NRSRO’s.  The SEC had never formally defined what it took to be an NRSRO.  They proposed a horrible set of definitions in 1997.  Fortunately, those never got past the proposal stage.

Erik, I thought, you know, was being very kind to the SEC when he talked about the process of designating an NRSRO.  It is a no-action letter.  It was an incredibly opaque process.  Sean can certainly go at greater length and probably show you all the scars on various parts of his body that were related to the opacity of this process.

So, what’s the consequence?  We’ve now got this NRSRO designation being used by virtually all financial regulators.  And because the major players in the bond markets must heed those ratings, in essence, the bond markets must heed the ratings.  And Erik asked, why is competition not working?  Competition is not working because there’s a small handful, now two handfuls, of guys who have a guaranteed market, who the markets must pay attention to.  Gee, when you got a guaranteed market, would it be surprising that pricing isn’t very vigorous, that innovation isn’t very vigorous; that you get distortions in that market.  Again, the markets have to pay attention to these guys, even if the markets think they don’t have a very good idea of default probabilities.  Those guys’ opinion must be heeded. 

Do they meet a market test?  Well, obviously, we can’t tell, we don’t know if these guys are really the best that could be done there because the markets have to pay attention to them.  Now, you might say, yeah, but when they change the ratings, markets react.  Yeah, that’s right, but it may well be that the markets are reacting just to the fact that, say, I’ll just pick Moody’s -- Moody’s as it changes a rating on a bond from, say, single A triple B where triple B is, of course, the cutoff for investment grade, the cliff.  They’ve moved that bond closer to the cliff.  They decide to downgrade it again and it has fallen off the cliff. 

Of course, the markets would react, but not necessarily because they think Moody’s has any good ideas about default probabilities but just because Moody’s has moved that bond closer to a cliff and one more move and it falls off the cliff.  This all most recently came to light, because the NRSRO’s were excessively optimistic about subprime-based indices [sounds like] and CDO’s.  Josh may tell us a bit more about that.  There are lots of cultural reasons why they were slow.  There is a listing of some --

All right.  How did we get that new legislation that got passed in 2006, that Erik mentioned.  Well, what I’ve just described to you, this NRSRO structure was the best kept secret in Washington D.C.  I spent a lot of time in Washington, until I started doing research I hadn’t heard of the whole NRSRO structure.  I would ask very knowledgeable friends.  They had never heard of it, and probably, most people here hadn’t heard of it until the Enron debacle and the press and the Hill discovering that up until five days before Enron’s bankruptcy, Moody’s and Standard & Poor’s and Fitch were all maintaining investment grade on Enron’s debt. 

You got the congressional attention, you got hearings, you got a clause in Sarbannes-Oxley that said, give us a report.  They issued a report, didn’t say a whole lot.  They issued a concept release in 2003.  They designated two more NRSRO’s in 2003 and 2005, bringing it up to five, so now we have a handful.  They proposed another set of definitions.  Fortunately, that didn’t see the light of day. 

In the meantime, the Congress passed a new legislation and President Bush signed it in September of 2006.  As Erik indicated, there is the basic intent of the Congress to open things up.  To tell the SEC to stop being a barrier entry to see if more competition is going to work and to be more transparent in what it does.  Just before it issued those final regs, the SEC designated two more NRSRO’s, bringing this up to seven, and then it issued the final regs in June of 2007.  As Erik indicated, the legislation is clear.  You’re not supposed to dictate the business model, they designated two more NRSRO’s.  Sean finally got his designation at the end of 2007, another firm in February, and then, as Erik just indicated -- tells us just a few days ago, we got up to 10, so this is now out of date.  There are now 10.

Alex J. Pollock:  One minute left.

Lawrence J. White:  All right.  So, the current situation, as everybody knows, Congress is very unhappy.  We’ve had the attorney-general of the State of New York come into an agreement with the three large NRSRO’s, more disclosure stage, payments, due diligence, representation and warranties.  We’ve got that pile of stuff that I just mentioned where again more disclosure, more limits on conflicts, new procedures.  They’re not supposed to dictate a business model.  This stuff influences the business model.  It influences how they do things.  That’s part of a business model, so I think it is a pipe dream to think you can talk about these things and not be influencing what they do. 

I’m really interested in what happens tomorrow, because as you’re going to see, I think that’s the direction to go.  I think this kind of regulatory approach is unlikely to think -- make things better and I think is likely to make things worse because it will rigidify, it will solidify, it will discourage innovation and it may make life harder for some of the competitors in the market and I guess we’re going to hear from Sean about that.  It is a big mistake. 

And so, what is the better course?  It’s the course that I hope we’re going to hear the SEC talk about tomorrow, where that delegation of safety decisions has to cease.  Rather than telling banks or telling broker dealers or money market mutual funds or insurance companies that they’ve got to pay attention to some select few group of credit raters.  Pull the delegation back.  Stop outsourcing it.  Make the same kind of safety and soundness determinations for bonds that the bank regulators, as a matter of course, do for garden variety loans and for other assets and liabilities of a depository institution.  After all, the judgments haven’t been all that good recently and better methods can be developed. 

Once they stopped doing that, we can eliminate the NRSRO category.  You don’t need it anymore.  Remember, it was their -- because of this outsourcing decision.  Once you stop outsourcing, you don’t need that designation and you don’t need to regulate them because the participants in the financial markets once they are not restricted to who to pay the attention to can make up their own minds and figure out who has the conflicts and should be avoided, who doesn’t, who has good track record, I want to pay attention them.  That’s where competition -- that’s how competition is going to work.

And so, in summary, think how we got here.  We got here because of that outsourcing decision.  The way they get out of this is to stop the outsourcing.  Once we stop it, we can end that category and end all this stuff about business models and transparency.  Let the markets make up their mind.  Ask yourself, do we really want the New York AG or the SEC dictating business models, dictating how the bond rating business works, or should we let markets do it? 

Thank you very much, Alex.

Alex J. Pollock:  Thank you, Josh.

Lawrence J. White:  You want me to put your slides up.

Joshua Rosner:  Sure.  Thank you.  So, precisely because I couldn’t agree with Larry’s views more, I’m going to recommend or suggest that we should actually have some regulation, further regulation, or potentially legislation.

[Microphone adjusted]

I said precisely because I couldn’t agree with Larry’s view any more, I’m going to actually make a few recommendations which actually call for greater regulation or legislation to the end of, actually, eliminating the need for the rating agencies as outsourced gatekeepers.  I’m agnostic in the sense that to my mind, you should ideally have a world in which no one is required to rely on rating agencies.  Unfortunately, given the fact that it will be almost impossible in any short or coordinated manner to get global bank regulator safety and soundness regulators, treasury for pension investors, and 50 state national associations of insurance commissioners to all agree to do away with precisely those regulations that require those investors to rely on rating agencies. 

I think we have to actually chart another course to that end, and so to some degree, I’m a little bit concerned.  I applaud the SEC for some of their proposals.  I think some of them moved the ball forward.  I do worry that too much information actually may just create a sense of a safe harbor, a sense of if you meet this standard you have met your obligation.  Unfortunately, I don’t think that’s necessarily going to work.

I’m primarily going to focus on the structured market.  I’m probably one of the only people at this table who actually does not have a specific problem with the issuer-pays model.  It seems to me that the issuer-pay model needs to have appropriate firewalls to manage conflicts of interest but so too does the investor-pays model.  You know, I think, it’s just as likely that an investor may pressure you to find a conclusion that they would appreciate as an issuer might. 

I’m not going to go through all of them, but historically, the rating agencies argued that their approach was, in policy and practice, to provide a consistent framework.  In other words, a triple A as a triple A is a triple A.  There should be some comparability across asset class and consistency across sectors and geographies. 

Over the past decade, as we see in structured products, newer structured products emerged.  We’ve seen the variants between triple A for, let’s say, a sovereign and a triple A for a CDO cube grow and a triple A is no longer a triple A in any standard format; which created a great opportunity for the rating agencies to generate incredible amounts of income by just doing large volumes of securities that if they were actually using consistent definitions, they never would have been able to generate.

So, one of the questions that I think needs to be considered is: does this make sense?  Is it really sensible, given what we now know that there would be that proportion of CDO’s rated triple A to sovereigns, especially if you consider the fact that the sovereign issuer has the ability to tax.  So, I think when you start thinking about things in the structured world versus the single-issuer world and one other point that I would make is when we talk about competition and all the new entrants and that we now have 10 NRSRO’s, I think it’s important to consider how many of those 10 really do anything, and do anything credibly, in the structured markets.  In the structured markets, unfortunately, we really are still at 3.  And I think that’s something that’s very important to consider in this discussion.  But some of the key difference single-issuer, broad base of issuers concentration on structured side, less complex legal structure, complex legal structure and we can go on. 

Down here, I think, it’s important -- the corporation in the single-issuer world exists prior to the rating.  That is to say that GM, that IBM, they exist and there’s very little that they can do in anticipation of a rating to gussy themselves up for that rating.  On the structured side, the corporation does not exist but for the rating.  And the rating agencies, at the end of the day, have the ability therefore not only to define what the rating is but to define the structure that needs to be met to garner the rating. 

So, you end up on the corporate side with a dynamic structuring world, a heterogeneous structure of different corporations set up differently with different management goals even within the same sector and different business practices.  And on the structured product side, definitionally, you have a great deal of homogeneity and that actually was one of the factors that actually exacerbated the crisis that we’ve seen. 

On the single-issuer side, you’ve got little principal risk.  On the structured product side, you’ve got complete principal risk to the investor.  So, now that we’ve established some of those points where the credit rating agencies always lag markets in non-credit risk instruments and that I think that’s actually sort of funny, we use the word “credit rating agency” on the one side.  On the other side, we’ll use the words Nationally Recognized Statistical Rating Organization. 

On the structured world, these instruments tend not to be primarily credit-risk exposed.  We’ve seen a lot of instruments where the spreads would indicate you’ve got a problem, liquidity would indicate that you’ve got a problem but the cash flows are still coming and the credit is still okay.  I wouldn’t say the majority, but there is still quite a bit of that.  And so, the rating agencies are sort of put in a position where their ratings are being used for something other than what they were intended as well. 

And so, I think that really needs to be considered, liquidity and market risk is a primary feature here.  And so, I think that is part of the reason that they have lagged and will continue to lag.  If you start going through it and look at it in some of the structured instruments, markets begin to widen spreads on the RNBS due to early payment defaults in the fall of ’06.  We didn’t see the rating agencies take any significant re-rating actions until the summer of ’07.  The markets had already done that. 

Rating agencies continue to claim active managements of CDO’s would result in low correlation between the RNBS that they downgraded and the CDO’s ending up having problems; that was quickly disavowed.  Markets drove the spreads wider on asset-backed commercial paper, on CIV’s [phonetic], on CLO’s, the rating agencies lagged.

We are not at the end of what unfortunately, I think, has been miscast, and frankly even in the SEC proposed rules, the word “subprime” really is the prominent feature when describing the assets.  And unfortunately, we are still at the very front end of this, and if the economy continues to weaken, we will see it spread rapidly into CMBS and I can’t imagine that we won’t rapidly see it expand into an CMBS and we are seeing an acceleration in prime. 

So, it’s not just they got their models wrong for subprime assets.  The models just aren’t really the appropriate measures to be using is really the argument here.  And actually, when you really start looking at competition and what are they competing for, I’m not going to go through all of them, it’s sort of interesting the IASCO code of conduct, which was a voluntary code that the International Organization of Securities Commissions put together, seems to have been violated pretty consistently, although even in the audit report by IASCO or by the committee of European securities regulators for IASCO, they didn’t really call it out as such. 

But I think if you look here, the IOSCO Code says that the rating agencies should adopt, implement, and enforce written procedures to enforce that the opinions it disseminates are based on a thorough analysis of all information known to the CRA that’s relevant to its analysis.  And then they put in, unfortunately, according to the CRA’s published methodology, which becomes a safe harbor, right? 

And then when you look at the concept that we’ve gotten out of -- that was Fitch, we’ve had similar ones out at Moody’s -- they claim no obligation to verify or audit any information provided to it from any source or conduct any investigation or review, or to take any action to obtain any information that the issuer has not otherwise provided.  I think that’s sort of ludicrous to hang our regulatory capital on.  And I think as much as we talk about the SEC should not be looking at the methods or the models, you can’t help but wonder how they can actually achieve an effective market for rating agencies without considering that. 

Even in terms of the questions of the involvement or review of models and methods, I think one has to consider that you can’t really fix a problem that has a systemic model failure problem without fixing either the methodologies that brought those models about or the models themselves.  And I think there’s a little bit of a misunderstanding. 

As an example, we have housing codes.  It doesn’t mean that we over-regulate.  You can de-regulate but you have to enforce sensible regulations.  Housing codes don’t limit the number of builders; it doesn’t limit the number of types of buildings that can be built.  And it seems to me, until we can effectively do away with this conferral of authority to these institutions, we need to have a transition period to get us out from under the dangerous grip of them.

So, to that end, we’re making some recommendations, we have some very specific ideas.  You can get the detail and more in the packet.  But bottom-line, it seems that charter-constrained investors should only be allowed to buy exchange-traded structured instruments.  That would require that any issuer who’s going to actually issue securities, that are going to be held, they have to be exchange-traded.  At least there, you will end up with a specific, minimum amount of disclosure and the ability for the investors to do a reasonable amount of due diligence on their own.  We’ll go through all of these. 

On that, it would also reduce the burden on safety and soundness examination staff.  It would create reputation-based issuer incentives to build better pools because of the transparency.  It would create a market in which both price and value could be discoverable.  And I think importantly, you could in this have some sort of a safe harbor where issuers choose not to list securities, chartered investors could still purchase securities but would be required to demonstrate to examiners that their own internal analysis independently supported holdings.  And that would further reduce the reliance on credit ratings.  So, I’m trying to use regulation legislation judiciously with the intent of decreasing the impact and the powers of the rating agencies for the purposes of market structure. 

It seems to me when the rating agencies change their models -- it's worth stepping back for a quick moment -- the rating agencies, I don’t have the same complaint in the single-issuer world that other people here do, I think in the single-issuer world, actually, if you look at the large failures of ratings over the past hundred years, they have not been that dramatic or that many.  I know Sean will probably disagree and that could be a discussion but by and large their models have actually been used in a relatively unchanged form for most of the past hundred years.  And they’ve got a hundred years of empirical data; in some cases, on specific issuers, on, certainly, specific sectors, on macro economic trends and cyclical trends, and that is what informs those models in the single-issuer world. 

Unfortunately, in the structured world, we’ve got a different problem.  You have new structures that never existed, in which all of the data that they’re using is not data at all; it’s statistical assumptions.  And unfortunately, what those statistical assumptions are being applied to are largely statistical assumptions because much of the collateral that’s being included in those structures also did not exist.  And so what we’ve done is we’ve actually allowed the replacement of empirical data with statistical data and the building of model flaw into structured model flaws and we’ve created a homogeneity of the structures so that if those models actually are flawed, we all lose.  And unfortunately, that’s where we’ve ended up. 

Again, I think this IOSCO Code is also important, to think in terms of the rating agencies actions.  When the rating agencies -– and this is the second recommendation –- when the rating agencies change their model in the world of structured finance because they are learning by doing, their models change constantly.  It seems that they should be required on a timely basis to go back and re-rate all existing securities that were rated with the prior model.  Currently, they typically, when they make a model change, use it prospectively only for new issues.  And then when they’ve got a problem with a prior or secondary market security, then they go back and they’ll use the current model. 

This actually creates ratings migration between vintages and this approach would prevent, as example, stale ratings of RMBS from being used to structure newer vintage CDOs.  As example, if you’re creating a 2008 vintage CDO, which we know can’t happen because no one would buy it, and you were to put 2005, 2006, 2007 vintage triple B mezz tranche RMBS into it, there’s strong possibility, maybe even a probability, that that 2006 triple B has not been re-rated since 2006.  And so the issuer has the incentive–-

Alex J. Pollock:  Josh, you’re on your second recommendation but you’re out of time.

Joshua Rosner:  Well, we’ll just run through them.

Alex J. Pollock:  Wrap it up in another minute.

Joshua Rosner:  Similarly, they require the automation and utilization of full original assumptions and life-loss [sounds like] curve coupled with monthly remittance data for regular automated secondary market ratings.  I think we should reduce some of the liability exemptions for certain structured finance rating practices which we detail in the package; create minimum indices standards for analyst professional trading and structured finance, there are none; prohibit revolving door practices as we’ve done for auditors in Section 206 of Sarbanes-Oxley. 

I don’t think you’re seeing analysts being bought off for chocolate cake.  I think part of the reason that they may be actually favoring issuers is the possibility of getting hired away to an issuer who they’ve done a good job in rating and we’ve seen that time and again.  And require an independent office of chief statistician; it’s great to try and firewall and protect the analyst from conflicts, but unless you actually make sure that the models are kept out of the revenue line management’s hands, it doesn’t matter how good an analyst you have.  So with that I’ll --.

Alex J. Pollock:  Thank you.  Sean?

Sean Egan:  Thank you very much, Alex.  There is one rule for the fourth speaker and that is you’re not supposed to be boring so if I am boring, I’ve asked Sylvain to poke me and if anybody else finds me boring, just raise their hand.

As I was traveling down on the train this morning, I was looking out on the Delaware Bay and it felt like I was in Northern Italy, in Lake Como, and it just struck me that there is a proposal that I’d like to make that’ll get rid of all these nonsense that we’ve been dealing with.  And I’d like to name -– it’s a modest proposal –- I’d like to name the proposal for somebody who has benefited terrifically over the last ten years from these skewed incentives and it’s Angelo Mozilo and so I’m calling this the “Triple A proposal.” 

And for those who are fluent in Italian, I’m calling it the “The Angelo Arrivederci Al Presto proposal.”  And it goes like this -- stick with me.  It goes like this:  the losses in the whole mortgage area had been about a trillion dollars -- that’s the IMF, I have a feeling we’re going to be bouncing off from that.  So, it’s a trillion dollars.  Now some people say that the major rating firms are a key part of it, so do the math, let’s say it’s between 10 and 20 percent, let’s say 15 percent, so you’re talking about a $150 billion that the major rating firms cost.  What I would like to suggest is that we get our inspiration from the farm subsidies. 

What you want to do is rating firms - if you pay them and ask them not to issue the ratings, they’ll do that.  So the federal government –- hold on, hold on, this makes sense –- the federal government could take a portion of that $150 billion and give it to the rating firms, S&P, Moody’s and Fitch; they’ll accept it gladly, they’ll love it because revenues are down, of course, but a portion of that $150 billion and they won’t issue any ratings and it works.  It really does.  Because if you look at the Fed, the Fed has had to contribute, just for one failure, that’s Bear Stearns -- trust me there are going to be other failures, there already are, it’s just in the backdoor -- but Bear Stearns they had to support $29 billion so if you take it, let’s say at 30 percent haircut, that’s $10 billion just on one issuer. 

So you only have to give them let’s say $2 billion to S&P, $2 billion to Moody’s and you’re home free.  And also I’d just wanted to -- so the Fed is on board with it; S&P and Moody’s, no problem there; Fitch, the same thing.  And if you talk to a lobbyist, they’ll be cheering, S&P and Moody’s lobbyists will be cheering because their job of course is to maintain and that’s what’s really preventing anything from being cleaned up.  So, I think –- and then what will happen is that you default either to non-conflicted ratings or else ratings generated by the investors themselves.  I think that that really works so think about it carefully as you’re going home tonight and I think that’s a path to get out of this debacle.

Now, if you disagree with that approach, I’d like to refer you to this piece of paper, actually, ten pieces of paper.  This doesn’t look like much but our attorneys charged us $25,000 for this so I would be mortified if you don’t take the time to read it.  He’s poking me.  No but really, this is the core here and we hit it on it earlier is that the people don’t understand borrower behavior, that’s what I learned in MBA class 20 years ago, is that to really be successful in the market, you have to understand borrower behavior. 

The rating firms -- S&P, Moody’s, and Fitch are not in the business of issuing timely accurate ratings.  If you do that, you’re lost.  In fact, there’s an articles where Moody’s announced that when they tightened up the standards for their commercial mortgage-backed securities in Bloomberg -- they even issued a press release on this, I couldn’t believe it, they provided me with the material -– but they said that when they tightened up their standards in the commercial mortgage-backed securities area, they lost market share.  Well, you know what?  They’re not in the business of issuing timely accurate ratings.  That’s our business.  We lose clients when we don’t issue timely accurate ratings. 

In fact, business hasn’t been extremely good for us lately just because of that, because we’ve been on the other side of some things.  We’ve been bearish on the monoline, bearish on the auto companies, bearish on the homebuilders and almost everything else except for a few things and people have made money from it.  It’s very easy.  You normally say that this credit is falling apart, it’s at $85 in the case of MBIA and then it drops to $4.  You have to be stupid not to make money on it, and that’s what’s happened. 

And so that what’s important, that’s what our business is all about: timely accurate ratings.  And you’re delusional if you think that S&P and Moody’s are in that business.  So that’s the hurdle that most people have to get over.  In fact, everything is -– and don’t moralize S&P and Moody’s.  They’re businessmen.  It’s like Truman Capote.  What do they expect?  The money is there, they have to get it.  There’s an incentive to do that.  I’ll tell you the Truman Capote story later, I don’t have time now.  Anyway, they need to do that. 

Let me just run through some key points here.  Timely accurate ratings, all these proposals and there’s proposals from the New York AG, there’s proposal from the SEC, there’s proposal from Europe, there’ll be another one from Europe, they’re all erroneous.  They’re erroneous because they precede on the false premise that rating firms are in the business of issuing timely accurate ratings.  That’s wrong.  They’re in the business of facilitating the issuance of securities, completely different business.  So if you try to somehow reform S&P and Moody’s, if you don’t take away the incentive, you’re doing nothing; it’s window dressing.  So, get over that hurdle. 

Also, if we don’t address this problem, our view is that any proposal is a miserable proposal if it doesn’t address the core problem.  It just creates a false sense of security, that’s post-Enron.  I’ve learned that “Oh, we’ll change some analyst, we’ll get rid of some models, we’ll do this, we’ll do that, get some forensic accounts in there.”  Nothing happened because they didn’t address the underlying problem.  That is the core incentive.

Before I forget, some people say, “Well, we should get rid of all rating firms.  Why not get rid of them?”  I put that in the category of why don’t I grow my own crops.  I’m trying to.  No really, I’m trying to.  I’ve planted some tomato plants and because of disease and everything, whatever it is, and I’m trying.  The problem is I have to wait until the end of August before I eat; minor consideration.  I think it’s great on paper, sort of like saying, “Well, you should fix your own car.  What’s wrong with you?  Why can’t you fix your own car?”  Well, I’m sorry, I don’t have the time.  I used to do that about 20 years ago and that set the car on fire, but it’s not very good. 

And in fact, if you want to broaden the market for securities, which is what our job should be, broaden the market securities, you want a natural third party so that the third party can do some due diligence.  It’s just like doctor.  If I feel terrible, I’d go to a doctor.  If the doctor can’t fix it, hopefully, he refers me to somebody else. 

It’s the same thing with rating firms.  If I want to rely on a rating firm, I want to make sure that rating firm has my interest at heart and don’t kid around with me.  Don’t get paid gobs of money by the drug company who suggests to you a drug that means nothing, the guy will lose his license.  Same thing with the rating firm, they should have the investor’s interest at heart.  Don’t worry about the issuers, don’t worry about the investment bankers, don’t worry about the lobbyists, they all can take care of themselves.  It’s the investors. 

I always think of the guy who heads up the French insurance company, he was told to invest in only investment-grade securities so he was smart and he said, “Okay, I’m not going to lose my job.  I’m not going to lose my job.  I’m going to invest only in triple A-rated securities.”  Unfortunately, it had the name of Ryan Bridge [phonetic] in front of it so it went from triple A down to single D in one day and so he came back from lunch and all of a sudden, he found that it wasn’t just double A, single A.  No, it was D and you’re not going to get any of your money back.  That’s the reason why the markets are frozen.

Let me run through this expensive paper before I forget.  Okay, the synopsis, you really should read that.  That’s helpful.  Let me point out a few other things.  On page three, there’s something interesting.  Read that segment especially the –- here it is -- they have something from the NCRC, they know what they’re doing in this area.  They say they have false securities.  Do you have something from coffee [sounds like] there?  I think that’s further on. 

Well, the major rating firms –- at number four, the major rating agencies have accumulated so much dominance that they actually impede performance.  This is great.  MBIA and Ambac, they weren’t triple A.  They never were triple A.  They couldn’t be triple A.  It’s just that the major rating firms couldn’t say that they didn’t have any clothes [sounds like] because if they did, then Eric Dinallo would’ve gone after them.  So they had to delay that and that just cost a couple of billion dollars. 

We got flagged because we said that their monolines needed $200 billion to maintain their triple A -- about seven or eight firms and some people said you’re crazy.  Now, I think we’re probably low in the $200 billion.  Remember, you can’t just cover your losses; you have to do much more than that.  But when your S&P and Moody’s [sounds like] aren’t getting paid by the insurance, you really have to be very careful about taking a major negative action.  I’m on page four and that is the issuer versus the pay. 

Here’s the coffee quote, that’s interesting, here it is.  It says, that the issuers and underwriters may fear independent rating firms because there’ll be credit call.  And then it goes on to that reference about Moody’s saying they lost market shares.  On to page five, these are some misconceptions.  This is worthwhile.  Take your time looking at this:  Problems are limited to structured finance.  I completely disagree with the notion that somehow structured finance is different than others.  We rate everything, basically. 

We’ll be introducing some more information on the structured finance area, but we don’t care if it’s in a corporate form, it’s in a partnership form, it’s in a coop form, it’s a new company, if it’s in the grocery industry, the finance, it doesn’t matter.  Credit analysis is credit analysis.  You have to be good at it.  The core question is whether or not you’re going to get your money back on time and in full.  And it doesn’t really matter.  And in fact, it creates a disservice to the investment community that somehow you’d say, “This isn’t in that category therefore you make this adjustment with that form.”  Forget it.  You go to the doctor and you just say, “What the heck am I supposed to do?  Give me a prescription and I’ll take care of it.  If I need surgery, tell me what’s involved.”

Alex J. Pollock:  Sean, one minute.

Sean Egan:  Okay, good.  That’s all I need.  The issuer-supported rating firms distribute theirs for free -– oh it’s not for free, you can go into that.  Government can issue ratings, well good luck.  I’m sorry but you had Chairman Cox saying three days before Bear Stearns’ bankruptcy that Bear Stearns was in terrific shape, it wasn’t.  And we had them way below everybody else and the reasons why that higher Chinese walls, forget it. 

Chinese walls don’t work and higher Chinese walls will never work.  More rating firms will open up the market, in fact, quite the opposite is true.  And you can go through -- some reasons, I argue that because Fitch is involved, that has made life more difficult for S&P and Moody’s.  Now they have to do what Arthur Levitt feared which is to have rating inflation.  They lie to us, they always lie to us.  It just happens and you have to get through it. 

Separate consulting from ratings, forget it, it’s a nonstarter.  You can’t separate them.  Good luck.  Four phone calls is consulting whereas three is okay, forget that. Investor-supported model is not viable.  It only worked for 70 years, what do you mean it’s not viable?  That’s crazy.  Investor-supported rating firms have conflicts.  How in the world do you have conflicts?  We have like 500 investors.  We don’t know whether they’re long or short.  We don’t care if they’re long or short.  We can’t poll everybody, what -- “Hey, guys, should we go long on this, go short there?”  Forget it.  We don’t have the time and another than that, maybe if they are long, maybe they shouldn’t be long.  So that’s again a nonstarter and investors are at fault.  There are some others that I’ll add but I can address that later. 

The last thing, I think the last thing is I want to point you to is on page eight, the analysis, basically if this and that isn’t a good recommendation then don’t bother doing it.  Our view is that it’s back on the investors.  The investors have to check out what they’re using, that it’s foolish to use a rating firm that doesn’t have their interest at heart.

Alex J. Pollock:  Thanks, Sean.  Sylvain?

Sylvain Raynes:  Hi, I’m Sylvain Raynes.  First, I would like to thank the American Enterprises Institute and Alex in particular for this invitation to be on this panel.  Because I’m the last speaker, I would like to make this remark as short as possible.  I think everybody would like to move to the next section where we go at each other.

First, I would -–

Erik Sirri:  I feel like we’re already in the next section.

Sylvain Raynes:  I would like to say that if you were a lawyer, you’ll want to sue; if you’re a regulator, you want to regulate; if you’re a politician, you want to politic.  So the proposed SEC regulation, which is a disclosure, is based on disclosures and other things, which are centered on it is understandable and it’s quite in keeping with the history of the SEC, which is centered, on disclosure.  The securities laws are not preventing people from investing in bad deals.  They’re supposed to prevent people unknowingly investing in bad deals.  What is disclosed is really, in my mind, the issue. 

Regulation AB which, at least in my humble opinion, is a very forward-looking regulation because it discloses something that is very meaningful, static pool analysis, which is a centerpiece of the analysis of structured securities, at least the good old fashioned ones that I used to know when I was at Moody’s.  We’ve moved on the mandate of structured finance has extended way beyond its original intentions.  Market value, CDOs are not structured securities.  When we talk about credit risks, when we should really make sure that rating agencies stick to their guns and stick to something they know.  Once they extend their franchise beyond credit risk analysis to market risk something they know nothing about, then we are heading for trouble. 

So the first time that a rating agency analyst said yes to Bear Stearns, which was the first firm to propose the market value CDO, then it was stepping outside the bounds.  This was not known to anybody inside the SEC.  If they had been known, if this had been explained properly, then we would not be in the mess that we’re in today.  So whenever you cross the line, only you know that.  Isn’t there a law that prevents you from doing it?  The first time a CFO manager’s earning is just a little too much, he or she is the only one to know that.  So regulation can only go so far.  I believe education is much more in keeping with the long-term survival of the system.  So if the disclosure requirements of the current proposition can go in the same direction as Regulation AB which, as I said, very forward-looking, then I think we will accomplish a lot. 

You can shower people with data, when I was at Moody’s, people would send me boxes and boxes -- I couldn’t even get into my office -- of so-called information and they would still call me and say, “What’s your problem?  I gave you a lot of information.”  And I said, “No, you’ve given me no information.  What you’ve given me is data.  Information is meaningful.  Data is not meaningful.”  So we need more information and less data.  If you disclosed the expected loss, which is a single number which is what, four bytes of information, of data?  That is all you’ll ever need to effectively do a back of the envelope analysis and get the rating yourself on your blackberry.  So that would be a step forward.  You don’t need all the rest.  There is too much data already. 

Now Erik put his finger on something very important.  When you want to eliminate something, an institution like a rating agency, when you want to throw the captain overboard in the middle of the storm?  You must ask yourself the question, if you’re a responsible regulator, who’s going to be the captain?  The question I think does not have an answer.  Okay, the SEC will be the captain.  No way, I would not want the SEC in charge of credit ratings and perhaps the SEC also doesn’t want that.  So giving out driver’s licenses and regulating traffic is not the same thing.  So how do we move forward if not by having something that looks suspiciously like a rating agency?  We do not have to have Moody’s and S&P, these are just names in the air.  So if you whack them, which is going to happen pretty soon, then who is going to replace them? 

Is there a mechanism -– and this is what Erik was talking about.  Can we do away with the notion of the broker?  Now that speaks to the primary market valuation.  Do we know how to value securities in the primary market?  What that means in the primary market, it doesn’t mean the security is not in the secondary market.  It means that you use the deal itself, you use the service [indiscernible] reports, which come in every month, to value the security.  Is this possible?  Absolutely.  In fact, this is the promise of structured finance, that it is uniquely in position to value deals by themselves. 

You don’t need to value deals in structured finance.  Why?  Because deals can value themselves.  You can have an automated nonhuman intervention system in the secondary market that actually updates ratings every month.  This is amazing because there are about 25,000 deals out there.  There is not enough people and talent to monitor these deals and to do a job that is even half of what it needs to be done.  You should not have human beings -- if you’ve ever worked in a monitoring group at a rating agency, you know how miserable you are.  You feel that you are a Maytag repairman.  Nobody wants to know what you think and you wished that you were an analyst, you want to do deals.  Therefore, the monitoring should be automated.  Is this possible?  Yes.  Is this possible in corporate finance?  No.  Structured finance is a nonlinear problem. 

Once the security enters the secondary market, it is linearized therefore a linear problem has one solution but a nonlinear problem has multiple solutions.  That’s why we need rating agencies.  If the primary market valuation were a linear problem, then there would be no need for rating agencies because there would be a unique solution.  So, uniqueness then is the essence of value. 

Supposed you have 25 rating agencies and they each rate a deal.  There will be, in theory, 25 different answers.  What is the market going to do with that?  The answer?  Nothing.  What will emerge is uniqueness.  One of those 25 or the average of them, the so-called mean, will then become the rating.  The market cannot accept -– it’s a visceral reaction.  You don’t have to tell anybody that.  You don’t have to pass laws about this. 

This will happen to Fitch.  Why does Fitch exist?  Because in the good old days, before Fitch came into prominence, you would need -- first of all, everybody knows you would need two ratings, therefore by definition, Moody’s and S&P would rate the deal.  By definition almost, they would have two different answers -- unless they would talk to each other, which was and still is, I guess, illegal, at least unethical.  So the investment banker faced with this dilemma and wanting to do the best job for their clients would tell one of them, could you move your enhancement down? 

They would say, “Of course not.  We are a responsible firm,” which is exactly the answer they should give.  What you would then have to do if you’re a responsible investment banker, knowing the market does not like split ratings, you would simply raise the enhancement of the deal so that both rating agencies would give the same rating.  You have then cost your customer, the issuer, potentially millions of dollars in increased interest cost. 

Now comes Fitch.  You go to Moody’s and S&P.  You have the same thing as before.  What do you do?  You pick up the phone, you call Fitch and you say, “What do you say?”  I’ve never seen Fitch walk away from a deal like that.  And now you have saved your clients millions of dollars, you are a genius.  So, is this what we want?  A bargaining -- this is called rating shopping.  That’s not going to disappear if we have 25 rating firms; it’s going to be even worse. 

So, what is the solution to this?  There is no solution.  What is the solution to the healthcare crisis?  It’s obvious, nobody should get sick.  But that’s not what we want.  That's not what we mean when we say, what’s the solution?  So I think Erik has it right.  We need to involve the investors in the rating process.  So now, the world is very simple.  The investor, when he or she is involved in the rating, he or she has implicitly signed off on the deal.  If something happens, the SEC can simply walk away and say, “It’s your mess.  You pick it up.”  That’s what’s going to happen. 

So by involving the four parties-- now one party is missing, the investors -– now I have to say this: the average institutional investor is really not very clever.  They are more interested in going to bars and going to play golf than doing their job.  Now, it’s not as simple as that to be in finance, guys.  It really is not as simple as reading the Wall Street Journal, no offense here.  There is more to do in finance than to read the first page of a prospectus and ask your broker, “Can you give me three beefs more?”  There is such a thing as finance.  It used to be the same in other fields. 

A hundred years ago, airplanes were not designed.  You would have some nut on top of a barn, jumping down, killing himself, his wife will collect the insurance and hopefully marries someone else.  Then, because people wanted to fly and I guess it was better than to swim across the ocean or to take a boat, they said, “You know, people can die in these airplanes.  Maybe there is such a thing aerospace engineering.”  They looked it up and it was done 200 years ago. 

Now, here’s the good news.  The things that need to be done in structured finance were not done 200 years ago.  They were done 2,000 years ago.  This is nothing new, guys.  These problems have been faced and solved.  Now, is there such a thing as a regulation already on the books in some country that gets rid of the need for rating agencies in corporate finance?  Yes.  Where is that?  Brazil.  What’s the name of this regulation?  Resolution 2682 and unfortunately, it has not been translated into English but it’s close enough to French that I could read it.  This regulation is eliminating the need for rating agencies in corporate finance in Brazil. 

It has worked so far.  If you remember Brazil, well, you cannot say -- whatever you say about Brazil, you can’t say they don’t know anything about credit risks.  I mean, the average shoeshine boy on Copacabana Avenue -- I can tell you this from personal experience -- knows more about credit risk than a managing director at JPMorgan.  It’s very sad.  JPMorgan did not have a credit culture.  JPMorgan had zero losses.  They specifically knew nothing about credit while we all believed they knew everything.  Of course, the obvious solution if you have zero losses, you must be a great credit analyst because you never lose any money.  No, you don’t lose money because the people that you give money to don’t really want it.  They have too much anyway. 

Where do we go to now?  Like Alex said, there’s not really a point here to blame anyone.  There’s enough blame to go around.  We want to blame the rating agencies because they’re the cops.  They should have stopped the bank robbers from robbing banks.  Bank robbers will tell you, “Hey, I rob banks.  What’s your problem?”  It’s the rating agencies that said I’m the cop and then stopped being cops; maybe they were like aiding and abetting. 

But in the end, this is the impossibility of absence of conflicts, no matter where you turn, and this is what Josh said, there is always a conflict.  There’re conflicts of interest between me and the other guy crossing Fifth Avenue.  The janitors in my building have conflicts of interests.  So buyers and sellers have an inherent conflict, which will not disappear, therefore we need ratings.  We need a third party, which is objective and this will not disappear because that third party is paid by one party or the other. 

So I’m suggesting a world where there are two brokers: one on behalf of the investors and one on behalf of the issuers.  The one on behalf of the issuers is obviously going to be the investment bank and the one on behalf of the investors is obviously going to be the rating agency.  So if you want, one part of the fee will be paid by the investor, and the other part of the fee will be paid by the issuer.  They can split down the middle and I’m hoping that I can collect a lot of those fees.  So, what do we do now to move forward?

Alex J. Pollock:  And you have one minute to tell us what to do.

Sylvain Raynes:  My suggestion is first what Josh said, the cooling off period.  We have to stop this revolving door.  If you work at a rating agency, you should have to wait a little bit, more than two days to work on the sell side.  The secondary market, we already talked about that, how to monitor deals automatically using the reverse engineering of service reports.  Is this possible?  Yes.  Is this easy?  No.  You have examples of this for a countrywide deal in your pack.  And primary market standards, again, these standards are about 2,000 years old so the big names here are people that are dead; so-called [indiscernible] count Aristotle and Plato.  So this is nothing new guys.  And I hope we can get there with everybody working together.

Alex J. Pollock:  Thanks Sylvain and thanks all for a very lively and interesting panel.  I’m going to give each panelist a chance to respond or elaborate on comments, if he wants to.  I’ll just say first, Sean said something about, of course, they’re lying.  I remember one of my mentors in credit who was a hard-nosed prime-rate plus-six percent commercial lender said, if the credit is in trouble they’re lying.  That was his rule.  And I think it’s actually quite a good rule to remember.  I’m going to go down the panel and give everybody a chance to comment.  Erik, we’ll start with you.

Erik Sirri:  Hard to know where to start so let me pick one or two points and leave the rest for others.

We’ve, sort of, had to maintain assumptions as we talked that the losses that we saw were somehow a function of agency conflicts in the credit rating agencies.  I don’t think the data support that.  So let me give you a concrete example or two. 

The securities firms themselves, the commercial banks that were involved in structuring and underwriting also lost money.  They lost money, they were part of the problem, they understood the nature of the relationship that a number of people here have talked about between the credit rating agency and the sponsor, and yet, they held these securities on their books.  And so as you read the paper, you’ve seen the losses. 

In a world where it was fully believed that these things were wrong, there is an interesting question about why they would have held this paper on their books.  Some have said that it just got stuck there, but I would say in my experience from looking at these firms and working with them, it means, I believe there is an affirmative decision of these firms to hold this paper on their books because they did not believe the credits would deteriorate that way.  So that’s observation one. 

Closely related to that, you have the monoline insurance companies.  Now the monolines had skin [sounds like] in the game.  They were using their own capital to guarantee when they wrapped these derivative products, the CDOs, they were using their own capital to guarantee their performance.  They too lost money, billions of dollars if you’ve read in the paper.  So their models also were wrong yet they didn’t have any agency conflicts because they were the people who had no upside and only downside. 

Point is, by counter example, there exists sophisticated players who made the same mistake who are not relying on the credit rating agencies to do their analyses.  So, you then are left with saying, how much can we say is due to the credit rating agencies?  I don’t know.  I can’t make that judgment.  But I can make the judgment that it’s not all due to that because there are at least two sophisticated groups that were out there that did something else. 

Let me make one other comment and then I’ll stop.  There is a question about what your expectations are for a credit rating agency and the quality of that rating.  Think about a simple situation in where you have someone who is right about a credit.  Now, take a bond that’s riskless, a treasury and compare a risky bond to it.  The only thing that separates them is their probability of default.  So the key information you have in this simple world, ignoring liquidity and everything else, is the thing that you care about in trading bonds is the differential probabilities of default or the differential probabilities of loss. 

If you have true information, if you’re good and you know something, where are you most likely to be located with that information and what’s the value of that information?  The ultimate value of the information is the most you can make by trading on it.  No one would pay more for the information than your ability to capitalize on it.  That’s the maximum value.  And who is in the position to pay the most for that?  A pure agent who stands out there and sells it in some way or someone who can trade on it as principal?  And I would suggest it’s the entity that can trade as principal because you don’t have the problem of separating good from bad sellers of information.  Result is, the best credit work is not going to be done in the public markets; it’s going to be done privately by proprietary investors. 

I’m not editorializing or critiquing credit rating agencies.  This was observed by Jack Hirshleifer in the ‘60’s, it’s a well-known tenet in economics.  So you have to step back and say, “All right.  What is my expectation for this credit rating agency?”  And I think we have to be limited in what we can expect.  Again, I’ll come back to the investors.  With that limitation, I’m going to look at the investors and say, “You investors have to understand that and temper your reliance accordingly.”

Alex J. Pollock:  Thank you.

Lawrence J. White:  Could I take a pass?

Alex J. Pollock:  You can absolutely pass.

Lawrence J. White:  Could I come back [indiscernible]

Alex J. Pollock:  Well, we’ll see.

Lawrence J. White:  Okay.

Alex J. Pollock:  Sylvain, any further thoughts?

Sylvain Raynes:  All I have to say is very briefly, the world we have today in structured finance, structured rating are inherently dynamic meaning because of nonlinearity, when you wish you [audio glitch] you have a trust with a finite life, the finitude of the life of the trust creates then the dynamic aspect of the rating.  It’s a static pool for that very reason.  Corporation, on the other hand, are infinite.  If you don’t know anything about corporate finance, the corporations were invented to last forever because as everybody knows, people die and this is very inconvenient that people die because they cannot accumulate asset --

Erik Sirri:  For some people, it’s very convenient.

Sylvain Raynes:  So you can have two worlds in deal-making, in structured finance.  You can have a dynamic structure with a therefore static rating or you can have the opposite, a static structure and therefore a dynamic rating.  This is the world we have now.  I would prefer the other world.  In the first world, the one we have now, the investor or the hedge fund manager can then benefit from the arbitrage if as Erik just pointed out, he or she knows what the situation is, so objectivity of information is the key here, which is what we’re all talking about. 

The other world, the world where we have a dynamic structure and therefore static rating, is the world where the issuer, the company, those who give jobs to people, guys, they would benefit because they would then release the enhancement that they have in the deal in order for -- because the investor doesn’t seem to care that these ratings are dynamic.  Most of the deals, believe it or not, even today, most deals work out, guys, otherwise, there’d be a revolution in this country.  Most deals work out, meaning most people get their money back.  Yes, it’s a bubble and all these. 

We don’t have a liquidity crisis.  A liquidity crisis in the United States is impossible since the 27th of December 1912.  So, there is no such thing as a liquidity crisis.  What there is, is evaluation crisis.  Now that’s a different animal.  The Federal Reserve cannot do anything about that because the Federal Reserve is a liquidity management tool for the United States banking system.

Alex J. Pollock:  1913.

Male Voice:  I thought you were referring to that, that’s 1913.

Sylvain Raynes:  Thank you.

Alex J. Pollock:  Josh, a comment or two?

Joshua Rosner:  Yes, I mean, I would like to first point out that I just disagree a little bit that the banks ended up taking massive losses and holding these securities, most of which actually were triple A rated, most of them were actually the super senior tranches.  And then also taking losses on assets and warehouse lines that got stranded and couldn’t be moved, doesn’t mean that they actually were not relying on ratings.  It just means that they too believed in the mania and that the markets were going to go up forever and that’s a classic cyclical greed response that we’ve seen. 

And similarly, with the monolines, here they were actually smarter than the rest of us relying on the ratings.  In fact, they had to believe that they were taking counterparty position to the investment banks and the investment banks, they’re not very smart on certainly on the CDS side, right?  And so I’m not sure that I actually would equate the fact that they got caught with their pants down with the fact that they weren’t relying on ratings or that they actually have good risk systems.  I’m not sure I can draw any conclusion tying those two together. 

On the other side, I guess one of the things that still -- I’ve got to go back to the issuer-pay versus investor-pay for a moment, because it still seems to me that there are four questions that can be asked of a ratings firm, that it doesn’t matter whether it’s investor paid or not:  Do you ever get asked by your client to rate specific securities whether it’s issuer or investor?  The outcome or the answer to that question is important.  Do you ever know whether the client has a financial interest in the rating?  How will your organization manage those clients’ requests?  And what percentage of your new ratings are unsolicited? 

And I think that those questions actually need to be asked of every firm, issuer-pays or investor-pays, and there needs to be -- back to my point -- the ability to manage the conflicts.  Sean had made a comment that they aren’t going away and shouldn’t go away.  If you want to rely on a rating agency, you should.  And I think that’s right.  I think the problem that most of us here agree is there’s a big difference between choosing to rely and being forced to rely and currently we’re forced to rely.

Alex J. Pollock:  Good.  Now let me get going.  And we’d need to keep these comments fairly brief, then we’re going to get our question period -- but Sean and then we’ll get to it.

Sean Egan:  Sure.  Thank you.  Why don’t I pick up on those four questions?  I think they’re very good.  Are we asked to perform ratings?  Yes, we are asked.  My view is that we’re in the market to serve investors and if investors have a particular concern, then we’ll be happy to give our opinion.  Do we know what the investor’s position is?  No, we don’t know with that investors or any of our other investors.  They just say, “If you’ll please generate a rating on it.”  So we don’t know if they’re long or don’t know if they’re short and we don’t want to know.  And they can go [audio glitch] short very easily now unlike they could before. 

How will it be managed?  We get to the truth quickly.  That’s how we manage or process.  Portion of unsolicited ratings?  I would say that from the issuers perspective, we have a 100 percent unsolicited ratings.  From an investor’s perspective, maybe we have five percent solicited ratings and they get 95 percent unsolicited.

Alex J. Pollock:  Thank you.  Larry?

Lawrence J. White:  All right.  The issue -- what’s the right model?  How to think about this thing?  And Sean mentioned about going to your doctor –- I don’t think that’s quite the right model.  Think more about you’re interested in buying an automobile.  Well, you could go to Consumer Reports and see what they have to say.  You could go to the auto show in Detroit or in New York or wherever they hold auto shows.  You could ask your neighbor.  You could get the advice of your mechanic. 

There’re lots of places you can look and no one forces you to -- and you can make mistakes and Consumer Reports may get it wrong sometimes but over time you’ll learn who -- and especially for things that you buy frequently –- cars, you may buy only once every three or four years, but there are alternative sources.  You don’t have to do the research yourself, you just can rely on others and figure out over time who is worth relying on and who is not. 

We shouldn’t be trying to force perfection.  Perfection is a chimera.  It isn’t going to happen.  People are going to make mistakes, institutions are going to make mistakes but markets will figure out who makes the fewer mistakes, who should be avoided because they have conflicts, because whatever.  And so, like Josh, I’m agnostic about what the right model is.  An investor-pay model works for Sean.  I don’t know if it would work for Moody’s.  Maybe it would, maybe it wouldn’t.  Remember, we are now in the world of the photocopy machine and the internet.  It isn’t working very well.  The buyer-pays model isn’t working very well for the record labels in the world of the internet.  Would it work for Moody’s?  I don’t know.  But we ought to have the opportunity to find out. 

One last point, and I agree with much of what Josh was saying especially about this larger issue of the model.  Whoever [sounds like], he brought up housing codes as a defense for imposing some regulation on these firms and he says, that hasn’t interfered with competition.  That’s right, however, first we have housing codes because you have ill-informed buyers who come into the housing market, maybe only once every seven years and aren’t very well informed about the details and unfortunately, the history of housing codes, is they may not have discouraged competition but they have discouraged innovation. 

Plastic pipe took a long time to get established because the plumbers unions wanted to install only metal pipes.  Housing codes have discouraged less expensive pre-manufactured type housing because the local carpenters and the local builders didn’t like pre-fab stuff.  So I don’t think the housing code analogy, the housing code model is a very good one. 

Again, just think in terms of this kind of Consumer Reports versus the local hardware store versus your neighbor versus your mechanic.  That’s the kind of model to be thinking about.

Alex J. Pollock:  Thank you.  Now we have come to the time where we’re going to have the questions from the audience.  We have a -– where’s our microphone right here?  So please wait for the microphone, identify yourself please and your affiliation and if the question verges into a statement, I’ll be limiting your time strictly.  The first hand I saw was back here.

Raghubir Goyal:  Thank you, sir.  Raghubir Goyal from India Globe.  There were several hearings on the Capitol Hill as far as credit ratings are concerned or credit cards and many people complained about the misleading credit reports on their credit ratings.  Where would you fit credit ratings today in this market where many people are losing jobs and also foreclosures and factories are closing down and so forth and their credit reporting agencies also making mistakes?  And what suggestion or advice do you have for people who has a bad credit, how to repair it or they will not get sick in the future as far as credit rating is concerned?

Alex J. Pollock:  I think you’re thinking about credit rating agencies of consumers.  Anybody want to comment on that?  It’s a little -- I think you have our panel stumped here.  Bert?

Bert Ely:  Thank you.  Bert Ely, banking consultant.  My question is primarily for Josh but also for the rest of the panel.  In one of your slides, you talked about reduced liability exemptions for certain structured finance rating practices, I’d like to ask a much more basic question: why should the rating agencies have any exemption from liability, why should their First Amendment exemption continue particularly in an issuer-pay model? 

We’d had lots of other professionals in finance, lawyers, accountants and so forth who’ve gone bankrupt, in some cases, because they’ve been sued for mistakes that they have made.  It seems to me that’s a perfectly reasonable disciplining tool.  Why should rating agencies have this exemption?  Hasn’t this First Amendment exemptions perhaps outlived whatever usefulness it had?  Just another question, Mr. Raynes, you mentioned a certain Brazilian resolution but didn’t tell us about it -–

Alex J. Pollock:  Hang on.  Let’s hang on.  Let’s see if Josh or somebody wants to pick up the First Amendment issue.

Lawrence J. White:  It seems to me again, there is some value, it seems to the exemptions as long as they’re protected where -- it is just an opinion and because you want to incent them to feel comfortable offering an unbiased opinion.  My concern is again, single-issuer versus structured, where they are defining the structures, where they are actually helping to advise by nature -– And that’s another point that I would make to the SEC is they don’t define a bright line of where you cross from the iterative and interactive discussions in the pre-ratings process where that’s distinguished from advice that the SEC is currently trying to propose rules on.  It seems to me that where they cross that line, however we ultimately define it, they should actually be seen as no longer journalists but underwriters, statutory underwriters, and that liability should go away. 

The other point and it’s the reason I drew it narrowly because I do see some value, is when they know that there’s fraud in the deal, when they’ve got reason to know that they’ve limited the amount of information that they’ve gathered so they don’t have to “know” there’s fraud in the deal, it seems to me again the liability exemption should be reduced.  I think there should be a burden but once that’s met, they should have liability because it is an effective performance tool.

Alex J. Pollock:  Sean, First Amendment question?

Sean Egan:  Yes, without the First Amendment protection, rating fees would probably increase by a factor of ten.  And even before you got to it, I think the rating industry would shut down for the simple reason that you won’t be able to get insurance.

Bert Ely:  So is that -- maybe that’s instead of the Fed giving them the $2 million.

Sean Egan:  Great idea.

Alex J. Pollock:  I’m going to come to this question back here in a minute, but before we do, Sylvain, briefly what does the Brazilian resolution say that you referenced?

Sylvain Raynes:  In the interest of time, I didn’t elaborate because it’s the whole thing, it’s called Resolution 2682, you can look it up on the internet.  Basically it bases the capital charge of financial institutions in Brazil on the delinquency structure of their assets.  The intent is to punish institutions that make risky loans but don’t collect them.  So the idea is if you make risky loans and you collect them, you’re a genius, you should be encouraged and you should take over your rivals. 

It’s only people who make risky loans that can’t collect them.  Therefore, if you make very safe loans and you always collect them, you’re no better than the guy who makes risky loans and collects them most of the time and sometimes he doesn’t.  Those two institutions have the same risk profile.  So you can leverage treasuries 50 times but you can only leverage double B exposures three or four times.

Alex J. Pollock:  Okay, thank you.  I have the question back here.

Neil Roland:  Neil Roland, reporter with Crain’s Financial Week.  First, for Erik then Sean, please.  How do you expect the SEC proposal, if adopted, to affect the competitive landscape of the credit rating agency industry?

Erik Sirri:  Well, I can’t speak about the proposal specifically because we haven’t proposed it yet, but I think –- I’m sorry.

Alex J. Pollock:  You mean the ones that are already out.

Neil Roland:  [inaudible]

Erik Sirri:  Fine.  I think what we’re trying to do is both increase transparency and reduce conflicts.  That and broadly, in this whole space, encourage entry.  I think for instance, on a transparency point, we are encouraging the production of information and the release of information by credit rating agencies when they rate a particular credit so that another credit rating agency had enough information so they could come in and perform an unsolicited credit rating on that same instrument.  So we took a significant step toward that.  So I think, to that extent, we’re trying to increase competition in that space through the lowering of cost for unsolicited ratings.

Sean Egan:  Our view is that if you ignore the borrower behavior, increased competition will exacerbate the problem and Arthur Levitt had a comment on that, I think it’s probably about four years ago, he worried about the race to the bottom.  Basically, when he saw Fitch getting stronger via the combination with Duff & Phelps, IBCA, and Thompson BankWatch, his concern was very well placed, it’s exactly what we have now, that is that there would be more rating shopping and so if there is more competitors in the market, if there is more shopping, then you’ll have bigger problems and more frequent.

Alex J. Pollock:  Larry and I want a lot more shopping by investors.

Sean Egan:  It’s fine if you have some mechanism for tying into the timely accurate ratings.  Right now, 90 percent of the compensation in this market comes from the issuers.  The issuers want the most attractive rating possible and most times, that’s the highest rating.  There’s some odd instances where they want a lower rating but most times they want a high rating which means that in this market, that’s an inaccurate rating and that’s a big cause for the shutdown in the structured finance market and other markets.

Lawrence J. White:  That’s because they have a captive market and if we can get away from the captive market so banks don’t have to pay attention -- and banks are freer to pay attention to you, maybe we’ll see some changes here.

Sylvain Raynes:  I disagree with that.

Alex J. Pollock:  That’s good.  That’s what we like at AEI panels.  I’m going to go Chuck Muckenfuss and then we’ll get you next and then I have a hand over here.

Chuck Muckenfuss:  Chuck Muckenfuss of Gibson, Dunn & Crutcher, I have, kind of a gratuitous comment that most of you won’t get which is that if you’re interested in this topic, you should look at the history of KMV, most of you in finance will know about KMV, you should just study the history from inception to purchase of KMV and it’s an interesting thing to look at. 

The question is really in the last dialogue, it was exactly what I was going to say.  I was going to ask the panel what would the world look like if Larry’s argument about no captive market were to obtain, that is to say, you would strip out of any SEC regulation a requirement of a reliance on rating agencies and if you stripped out all other reliance on regulation, agree or disagree and what would the world look like if you just did that?

Alex J. Pollock:  Who wants to try?

Lawrence J. White:  I’ll start.

Alex J. Pollock:  Larry, you try first.

Lawrence J. White:  It would look like the world of the 20s where guys who are providing value will survive and those who aren’t –- and again, I don’t know what business model they would have but they provide value to the market, they will survive, they don’t provide value, they don’t survive. 

And separately, the regulators will have to make their own safety and soundness decisions about what’s an appropriate bond portfolio for a bank.  What’s the appropriate set of assets for a money market mutual fund?  What is the appropriate capital for a broker-dealer?  That can be done. 

Bank regulators do that right now for the loan portfolio, it isn’t