Bonus culture datapoint of the day

By Felix Salmon
December 22, 2010
Jake Bernstein and Jesse Eisinger have a great story today about how Merrill Lynch persuaded its internal fund managers to buy the toxic nuclear waste spewing from its out-of-control CDO factory:

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

Jake Bernstein and Jesse Eisinger have a great story today about how Merrill Lynch persuaded its internal fund managers to buy the toxic nuclear waste spewing from its out-of-control CDO factory:

Bank executives came up with a fix that had short-term benefits and long-term consequences. They formed a new group within Merrill, which took on the bank’s money-losing securities. But how to get the group to accept deals that were otherwise unprofitable? They paid them. The division creating the securities passed portions of their bonuses to the new group, according to two former Merrill executives with detailed knowledge of the arrangement.

The executives said this group, which earned millions in bonuses, played a crucial role in keeping the money machine moving long after it should have ground to a halt.

“It was uneconomic for the traders” — that is, buyers at Merrill — “to take these things,” says one former Merrill executive with knowledge of how it worked.

Within Merrill Lynch, some traders called it a “million for a billion” — meaning a million dollars in bonus money for every billion taken on in Merrill mortgage securities. Others referred to it as “the subsidy.” One former executive called it bribery.

The money went straight to a small group informally known as the “Super Senior Facilitation desk,” headed by an Asia hand named Ranodeb Roy:

The agreement, according to a former executive with direct knowledge of it, generally worked like this: Each time Merrill’s CDO salesmen created a deal, they shared part of the fee they generated with the special group that had been created to “buy” some of the CDO. A billion-dollar CDO generated about $7 million in fees for Merrill’s CDO sales group. The new group that bought the CDO would usually be credited with a profit between $2 million and $3 million — despite the fact that the trade often lost money…

Roy made about $6 million for 2006, according to former Merrill executives.

It would be fun to dredge up some contemporaneous Merrill statements about their culture of risk management. But my worry is that the culture of risk management at BAML, or whatever Merrill Lynch is calling itself these days, is unlikely to be much better.

Merrill’s antics are the reductio ad absurdum of bonus culture, and show why it’s so silly for investment banks to pay multi-million-dollar bonuses and reckon that they’re protecting their long-term franchise at the same time. Not everybody was as egregious as this. But the differences between Merrill and other investment banks were only of degree, not of kind.


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

Weird, according these two fuckwits previous “research”, where they showed funds designed to invest in asset-backed securities invested in asset-backed securities, in 2005 H2, ML sponsored CDOs invested in over 75% of ML CDOs and then when this shadowy group was set up that percentage went DOWN to 52%, then 37% and finally settled at 42%. But as we know from their “work” on Magnetar, money lent to banks under the PDCF, and that piece you linked to on Goldmans CLO they are apparently proud to show off their innumeracy and complete and utter cluelessness. I am sure the “former executives” they are speaking to are having a good laugh at these two retards who will clearly believe anything even vaguely anti-bank someone says.

For the record, the ML CDO desk had 45bn of exposure on their desk of which they lost 43bn. Does that sound like people who knew the crash was coming and were offloading the junk?

Posted by Danny_Black | Report as abusive

@ Danny_Black wrote “For the record, the ML CDO desk had 45bn of exposure on their desk of which they lost 43bn. Does that sound like people who knew the crash was coming and were offloading the junk?”

Is that supposed to be an exoneration? Or an indictment?

ML’s CDO desk is just too stupid to be blameworthy, Felix, stop pointing fingers at underlying structural misincentives! They would have blown themselves up without all the bonuses, so what’s the big deal about the excessive and ridiculous comp plans!?!

It sounds like the ML CDO desk was, at best, grossly incompetent and conflicted by the risk-undermining disconnect between bonus-paying but very risky & ultimately money-losing transactions, and more probably knowing perfidy by using bonuses to lure another in-house ML desk to take on (and thus evade CDO-desk risk limits) known crap that they otherwise wouldn’t have agreed to take on without extra bonuses because it was, in fact, known crap.

Why did ML have $45n of exposure left on their desk to begin with? Anything to do with the fees & bonuses earned on the creation side even after it became apparent that the market didn’t have much appetite for them left?

Posted by SteveHamlin | Report as abusive

Someone had to be left with the hot potato…

Posted by hsvkitty | Report as abusive

SteveHamlin, I think it is generally accepted that the CDO desk of ML represented dumb money.

That said this article is clearly false:

1) It claims this desk was set up to take on extra paper in early 2006 because of the lack of investor demand. The fact is that anyone who claims there was a lack of demand in late 2005- early 2006 is a balls out liar. I would suggest that the writers of this article are simple out of their depth but they have posted “research” that even shows this is simply not true.

2) The desk thought its risk was on the order of tens of millions and were nowhere near breaching their risk limits. The fact that the ACTUAL risk was nowhere is something they only found out long after this event.

3) I have no insider knowledge of this but I can make a pretty good guess what was actually going on. Primary issuance desks like league tables. ML wanted to jump to number one and so “sells” a portion of the paper to this investment desk who in return get a cut of the extra bonus the issuance desk makes due to increased business. That turned out to be a bad deal, they blew up and now are blaming someone else – as by the way are the ML CDO desk.

This story is completely misreported and has nothing to do with “misincentives”.

Posted by Danny_Black | Report as abusive

Making me laugh now Danny! You have no insider information but working on a sales desk means you know how the deal went down, of course… as it was the order of the day…excuse the pun!

As someone on the sales desk, were you then structuring the very instruments that helped meltdown your economy in the UK? Everyone knew they were bad trades, but as a hot potato, it blew up in the hands of those who couldn’t unload it.

Here’s hoping some of the people involved in risk management or those who knew what was happening start to leak that there was pay off with ‘bonuses’ to look the other way … Bribery by any other name is still bribery.

Posted by hsvkitty | Report as abusive

hsvkitty, you do of course realise it was commercial banks that went under and needed to get bailed out in the UK right? You are of course aware it had absolutely nothing to do with CDOs in the UK? You are of course completely in control of all the facts whilst pontificating in public? Noooooo? I am shocked.

PS worked on a trading desk.

Posted by Danny_Black | Report as abusive

Danny_Black, “league tables”? “Clean tables”? Or do I not understand the term?

Posted by TFF | Report as abusive

Maybe didn’t understand. League tables like “We are the number one CDO issuer”. For example, a govie market making desk might artificially compress spreads to buy business which the govie primary issuance desk then uses to get business and in return that desk shares some of its P&L with the market making desk.

Posted by Danny_Black | Report as abusive

Ah, I see. That might be a Brit term? Though I can’t think of an American equivalent (standings? rankings?), so I wouldn’t be surprised if it were borrowed for some purposes.

Thanks, DB.

Posted by TFF | Report as abusive

rankings I think is the closest. Walk round a trading floor of any bank and they will have these sort of posters – “number one BBB Vietnamese Rice-backed Bonds research analyst 1994″.

Posted by Danny_Black | Report as abusive

I don’t know for sure, but I strongly suspect that the manner in which the scale of the holdings were concealed, and the manner by which the internal buyers were compensated, were both similar to the way in which these were perpetrated at UBS. In fact, to link this story without any reference to the EBK/UBS Shareholder report on those losses is rather remiss. s?newsId=140339

Posted by drewiepe | Report as abusive

drewiepe, thanks for the link.

Not sure what you mean by the holdings being concealed. ML boasted about it’s warehouse and CDO positions in 2005 and 2006. What was not apparent was the risk they were taking. The desk limits were apparently VaR based without operational limits on the amount of loans taken on. I have to say I find this extraordinary given literally everyone knows the flaws in this methodology, especially for AAA rated FI instruments where the risk is all downside and discontinuous.

I would say that UBS stands to a certain extent as a counterexample to what is claimed here. UBS treasury was compensated on a stand-alone P&L basis and so was incentivised to lodge “excess cash” in higher yielding ABS. Had they had this soft P&L arrangement where they were rewarded for the liquidity they provided to the other bank units they presumably would have focus more on the liquidity than the face yield.

Posted by Danny_Black | Report as abusive

As a side note, I think that UBS report highlights some of the true issues that helped caused the crisis:

1) As I said before I find it extraordinary that VaR was used as anything more than a coarse directional heuristic. I can understand the focus from a reg arb point of view. Ditto ratings.

2) The lack of focus on the management of liquidity as a goal in itself. Capital and “profits” are opinion, cash in cash out are facts. It is something that is relatively easy to measure and harder to game on a rolling basis and a liquid bank can sit out short-term blips that a bank with lots of “capital” but no cash cannot. I also see no reason why banks should not be able to show more or less real time info on liquidity to regulators if not publicly. Much more useful CVA than say share price and tends to be more accurate a predictor of stress with less frequent and smaller jumps.

3) Feedback loops. They hedged some of the SS tranches with monolines. Prices decline, monolines have to put up more cash, they get downgraded, hedges are not properly hedged, bank sells causing prices to decline. Rinse and repeat. Same with the wholely artificial banding for credit risk, the static focus on credit risk and regulatory limits on different grades causing unnecessary discontinuous jumps.

Seems to me the last two are really key and barely covered.

Posted by Danny_Black | Report as abusive

Apologies for not returning sooner Danny. Sorry for not being more clear. I was not really talking about the PNL arrangements but rather about the risk mitigation techniques/theories utilised. Namely allowing positions to “collapse” to zero balance sheet usage and zero risk if they were covered by the purchasal of small amounts of CDS protection.

Posted by drewiepe | Report as abusive

Moreover Danny, if you don’t mind me saying, I think this “real-time liquidity assessment” is something that is a focus for new FSA regulations, as well as Basle III.

Posted by drewiepe | Report as abusive

thats ok was sort of waiting for you…

Only saw a little bit about LCR in coverage of Basel 3, unlike the coverage of capital and exotic debt instruments. Haven’t read a thing about FSA – mostly because I don’t care about the UK terribly much any more and it doesn’t pop up that often in what i read.

Also the little i did read still have a built-in trigger namely the hard barrier on the credit rating of the “risk free” bonds counting towards liquidity.

Cheers anyway for the response, always worth reading!

Posted by Danny_Black | Report as abusive