John Thain comes clean

By Felix Salmon
October 7, 2009
Financial News has a quote from John Thain, at "a speech this month":

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Financial News has a quote from John Thain, at “a speech this month”:

“To model correctly one tranche of one CDO took about three hours on one of the fastest computers in the United States. There is no chance that pretty much anybody understood what they were doing with these securities. Creating things that you don’t understand is really not a good idea no matter who owns it.”

This is the same man, of course, who, during his tenure as CEO of Merrill Lynch, would repeatedly take a large writedown, raise capital, and say that his marks were conservative and that there would be no further writedowns or capital-raisings. And then of course he’d be back with more a couple of weeks later. Certainly Thain tried his darndest as CEO to give the impression that he knew exactly what his CDO holdings were worth, and that Merrill understood them very well. So I guess he’s now saying that he was lying?

Still, I’d like a bit more color on the “correct model” that Thain is talking about. Which computer does he have in mind? And how much faith does he have in the results? It’s hilarious to me that even after everything he’s been through, Thain thinks that if only we had more computing power, we might not have been in this mess. More likely, we’d be in an even bigger mess.

Update: The Q&A session at Wharton from which the quote was taken can be found in PDF form here. (Thanks, Cardiff!)

Update 2: John Carney points out that in January 2008, Thain seemed a bit less sure of himself when it came to how much all those CDOs were worth. But that call, on January 17, came in between a statement on January 15 saying that he was “certain that Merrill is well-capitalized”, and a statement on January 18 that he was “very confident that we have the capital base now that we need”. (Of course, any further writedowns in the CDO book would directly hit Merrill’s capital base, so statements that Merrill’s capital was safe were tantamount to statements that there wouldn’t be any further CDO writedowns.)

Later on in the year, Thain continued in this vein:

“…Today I can say that we will not need additional funds. These problems are behind us. We will not return to the market.” (March 8 )

“We have more capital than we need, so we can say to the market that we don’t need more injections. (March 16)

“We have plenty of capital going forward, and we don’t need to come back into the equity market… No more capital raising; I’m sure we have enough capital.” (April 4)

“We deliberately raised more capital than we lost last year … we believe that will allow us to not have to go back to the equity market.” (April 8 )

“We are well-capitalized. We’re comfortable with our capital position.” (June 11)

“We are in a very comfortable spot in terms of our capital.” (July 17)

How could he be so sure, if no one had a clue what Merrill’s enormous CDO book was really worth?

Given the size of Merrill’s CDO exposure, and the degree of uncertainty which Thain now claims was endemic to that exposure, it seems incredible that Thain could have been so certain about the amount of capital that Merrill needed. Of course, in retrospect all his claims about needing no more capital were wrong: the losses would never end, even after Merrill was taken over by Bank of America. As he now admits, the CDO book was so opaque that Thain had no grounds for being so cocky.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

it’s not hilarious, just sad, and it’s called hubris

Posted by yr | Report as abusive

Model – computer – Maybe he was using some 8086.

Thanks for the investment banking industry, it “pop” the technology industry some time earlier in this decade and he should not blame for the computing power problem. Blame himself!

Posted by Rose Eli | Report as abusive

Really? You really read this as him saying that more computing power would have been helpful. I don’t see how you get to that at all

He is saying that no one understood these things precisely because it took a supercomputer three hours to churn them out – implying that it would take a human doing it by hand god knows how long, certainly too long to make ABS CDOs a big business.

As for modeling correctly – a poor choice of words on his part. I am going to assume that he means a model that was considered correct at the time. But who knows.

Posted by Murray | Report as abusive

Hi Felix, that quote was part of the exchange below. The transcript is here: Thain isn’t saying that these securities could have been better understood with more powerful computers, but simply that they were too complex and incomprehensible to begin with. So to the extent that as CEO he gave the impression that “Merrill understood them very well”, as you write above–yeah, that could be charitably called misleading at best (my personal view, not that of Financial News). Anyway, here is the exchange:

Siegel: Related to that, I think you really hit on this when you said there was a lot of money creating these bundled securities and commissions in selling them. I think the problem was they were also being held. Maybe for two reasons — maybe they couldn’t sell them at the end, and maybe they just thought they were good investments. It was very different than the tech bubble where the brokerage firms basically … got them all out into the system and didn’t hold them. The actual loss of stock market wealth in that bear market was really very much similar to this, but it didn’t have the impact on the financial firms because the financial firms didn’t hold these Internet stocks that tanked the way this did. Should they have just gotten rid of them instead of having them concentrated in their own portfolios?

Thain: Well, whoever owned them still is going to lose money.

Siegel: But it wouldn’t have tanked the firms. I think the tanking of the firms is really a major source of the recession.

Thain: I agree with that. But these instruments were still bad. These instruments were so complicated. One of these — I’m talking now about ABS CDOs, and actually the CDOs-squared are even worse. Merrill created one. I picked one particular one. To actually model out the things that are inside an ABS CDO [is difficult] because derivatives are already inside the CDOs. They’re already derivatives created out of derivatives. So you have to go way underneath to get to the actual pools of mortgages. To model correctly one traunch of one CDO took about three hours on one of the fastest computers in the United States. There is no chance that pretty much anybody understood what they were doing with these securities.
Creating things that you don’t understand — that the buyer doesn’t understand, that the rating agency doesn’t understand, that the regulator doesn’t understand — is really not a good idea no matter who owns it. I think that the degree of complexity that was created in the securities, and the lack of anybody’s ability to really understand how they were going to perform, was simply an error and a bad thing. The fact that the firms that created them were stupid enough to own them doesn’t make me feel any better.

Posted by Cardiff | Report as abusive

re: computational complexity and ‘bad’.

may god help you all if you are ever thirsty and have to drink a glass of water.

Posted by q | Report as abusive


All you really need for huge computing power these days is to harness the unused cycles of volunteers. Join BOINC, describe your project as “folding proteins” or “finding signal through the noise of intergalactic radiation” and compute away.

Murray and Cardiff are right. The quote is interesting to say the least. I always get a kick out of hearing Thain’s post-collapse excuses. I hate to defend him, but Felix, your criticism doesn’t match the quote’s content at all.

Posted by Eileen | Report as abusive

There is a quote regarding debugging computer programs by Kernighan that goes:

“Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.”

I think “writing the code” could be replaced with “building financial instruments” and “debugging” with “assessing risk”

Posted by dmo | Report as abusive

My comment was poorly worded and vague. To clarify, I meant that Thain’s behavior as CEO was misleading to the extent that he expressed confidence in Merrill’s understanding of these securities. I didn’t mean to imply that Felix’s post was misleading–it wasn’t.

Posted by Cardiff | Report as abusive

I don’t think Thain was lying when he kept saying “this is the last write-down”, I think he had no clue about the depth of the problem, just like his predecessor and many others in the industry. Computer models exist mainly to justify what those execs wanted to do, which was to ride that gravy train as long as possible, and if it crashed, be able to escape with a outsized package (like Stan O’Neal, the guy before Thain).

@dmo, thanks for bringing Kernighan into this discussion. It’s nice to be reminded of a world ruled by logic and reality.

Posted by KenG | Report as abusive

It’s not the computer. It’s the model. It’s who designed the model and how it’s designed. The rationale, the mathematics, the algorithms, the assumptions.

But one does not need any model to see the conceptual and ethical flaws of CDO, the slicing and dicing.

If a bunch of PhDs present a CEO with broad sense of business and ethics, with financial products based on such a concept, it would be rejected within 2 seconds. But none of the big investment bank brass did. Because they don’t operate using such senses or ethics. When they believe they are master of all, emperors of empires, then nothing else matter.

Is not surprising that John Thain, even now, finds it impossible to understand basic values of business, humanity, ethics, responsibility.

Just a G W Bush finds it impossible to comprehend anything beyond his idiotic fundamentalism of black and white.

Absolute power within primitive brains corrupt and destroy absolutely.

Posted by The Real Deal | Report as abusive

If John Thain is as clueless as he now freely admits; what does that say about the talents and smarts of everyone else in the financial industry?

Posted by martin braun | Report as abusive

No CEO or any executive staff should say they did not know or have an idea about how bad these ‘assets’ were. If it’s commonplace in the daily news to read about people with undocumented incomes of getting $500K loans, well, you don’t have to be very smart to figure it out. You just need to read and not listen to the majority’s opinion when you make up your mind. John Thain knew. But why blow it if you’re making millions?

Posted by blackbean | Report as abusive

What a shame that there was ‘no chance that pretty much anybody understood what they getting paid’ to market those CDOs. In contrast, those numbers were very tangible.
Bring on the civil suits and criminal charges.

Posted by OSMR | Report as abusive

“But one does not need any model to see the conceptual and ethical flaws of CDO, the slicing and dicing. ” posted by The Real Deal

Real Deal, could you please explain the ethical flaws of CDO?

Posted by jkingqm | Report as abusive

I have nothing against the logic of the argument. But I think it is unforgiving. Surely how one judges Thain’s recent comments should depend on what one thinks he was trying to say. Of course, you could just see it (as Felix does) as spurious wisdom after the event, but you could also see it as a mea culpa. And in that case, surely the correct response would be that of Luke 15.7: “Joy shall be in heaven over one sinner that repenteth, more than over ninety and nine just persons, which need no repentance”.

Posted by Jonathan Ford | Report as abusive

It certainly boggled Thain’s brain!

As for OSMR, the word is not just shame, but sham!

As for not having a clue, KenG, Thain was head of the mortgage desk from 1985 to 1999 at Goldman Sachs, and president and co-chief operating officer there from 1999 to 2004, and CEO of the New York Stock Exchange from January 2004 to December 2007, where he laid off THOUSANDS of people, THOUSANDS. Including a lot of people who might have had a clue….

So when we look at all the bad debt due to defaulted mortgages, let’s also thank Thain, who was responsible for the free-for-all environment that ‘valued’ these things as worth something, a notional number on paper only, and which we then tried to ‘prop up’ with REAL dollars, not notional ones, from the hardworking American tax payers.

The guy lies like a rug, AND is incompetent too.