A provocative discussion of Goldman Sachs

[NeedOPT]

I would agree for the most part, but it seemed like more of a domino effect. The weekend Lehman went bankrupt, it looked like Morgan Stanley and Merrill would be next, then Citigroup, then B of A, Goldman Sachs, and the rest of the financial sector except Northern Trust, US Bank, and Bank of New York. Maybe JP Morgan had a fighting chance. But the point is that the guys at GS were screwed- especially when they were staring at $10 Billion in counterparty risk with AIG.

How did GS have 10 bill in counterparty risk with AIG?

As far as the assets involved in the Maiden Lane III transaction, they had 1.2 bill in counterparty risk with AIG (according to Goldman's calculations; according to AIG's calculations there was no counterparty risk because they'd provided full collateral, but I'm inclined to trust Goldman's numbers on this more), offset by 1.2 bill in CDS on AIG, according to the SIGTARP report on MLIII.

In the event of an AIG failure, SIGTARP goes on to state that additional counterparty risk with AIG may have appeared on Goldman's books because the market for CDOs might have been driven even further down (if the underlying assets in the MLIII transaction went to exactly 0 then this would have created an additional 4.3 billion in losses). Additionally, SIGTARP notes that some of Goldman's counterparties on the CDS held against AIG might not have been able to pay out fully. The counterparty risk here (which was not with AIG!) would have been something less than 1.2 bill, as the CDS were at least partially collateralized.

In particular, I would note that Warren Buffet noted he wouldn't have made an investment in Goldman Sachs if he couldn't count on a TARP-like bailout.

I haven't seen this. I'd be interested in reading about this statement if you have a source.

Goldman Sachs is filled with some pretty smart people, but they wouldn't have survived without the US taxpayer.

This seems like a fairly bold statement. I'm not sure anybody really knows what would have happened in the counterfactual world of no bailouts. However, I doubt that a few billion in losses (say 5, as the absolute maximum of its exposure with AIG-related business) would have driven Goldman completely under.

---------- Post added at 09:23 AM ---------- Previous post was at 09:19 AM ----------

How exactly? I did not get a cheque from the US government from the money presumably made off the bailouts.

If you want a cheque from the US government every time it makes money then first you're going to have to send them a cheque for your share of the national debt...

 

[GoIllini]

How did GS have 10 bill in counterparty risk with AIG?
This seems like a fairly bold statement. I'm not sure anybody really knows what would have happened in the counterfactual world of no bailouts. However, I doubt that a few billion in losses (say 5, as the absolute maximum of its exposure with AIG-related business) would have driven Goldman completely under.

It's hardly a bold statement. Goldman Sachs's biggest counterparty exposure, like every investment banks', comes from its market-making business. Unless it will have a credit balance with all of its counterparties in the event of a market capitulation that forces more than half the banks into bankruptcy protection, Goldman Sachs would be in trouble too. If you looked at GS's balance sheet in September 2008, it had a 95% liability ratio. A lot of those liabilities were owed to the firm's counterparties but offset by other counterparties who were getting ready to declare bankruptcy.

So you tell me. Could GS have handled 50-60% of its counterparties going bankrupt with its 95% liability ratio? I think anyone with a basic accounting background would be pretty skeptical. You also have to remember that the credit market was in panic mode back then. Even if GS somehow had the solvency to pay off its creditors, it might not have had the short-term liquidity if there was a bank run and the market was as volatile as it was and the margin calls started coming in.

GS was better positioned than most investment banks for a market crisis, but the guys at GS clearly don't have the same wisdom that the conservative trust banks who supposedly "missed the boat on sub-prime mortgages and credit derivatives" did.

I haven't seen this. I'd be interested in reading about this statement if you have a source.

Buffett said that he may regret having bought the stake in Goldman if the rescue package doesn’t pass.

“If congress doesn’t do anything… I may wish I hadn’t gone in….I think anybody buying anything this week will wish they hadn’t if Congress doesn’t act,” Buffett said.

http://www.foxbusiness.com/story/ma...comments-goldman-sachs-stake-government-bail/

Again, I was there in September, and it honestly looked like the end of the world (unless you worked at a conservative trust bank that others had spent the past seven years foolishly attacking for missing the boat on mortgage bonds and the credit derivatives markets). Several of my friends at Goldman Sachs (albeit with student loan debt) were saying they wished they worked at Harriman or BNY. If everyone on the street had gotten a libertarian bail-out, 90% of the banks would have disappeared, including Goldman Sachs. In fact, if you drew the counterparty graph, they were probably next in line after Citigroup and Morgan Stanley.

In any case, a number of bulge-bracket investment banks did not require government bail-outs, but GS did. It was probably the best positioned of the investment banks, but the Treasury department was still concerned about its exposure.

How exactly? I did not get a cheque from the US government from the money presumably made off the bailouts.

No, but it reduced the amount of money that the feds will either have to tax you for in the future, default on (thereby throwing the economy through a loop), or print (thereby debasing your currency).

 

[NeedOPT]

It's hardly a bold statement. Goldman Sachs's biggest counterparty exposure, like every investment banks', comes from its market-making business. Unless it will have a credit balance with all of its counterparties in the event of a market capitulation that forces more than half the banks into bankruptcy protection, Goldman Sachs would be in trouble too. If you looked at GS's balance sheet in September 2008, it had a 95% liability ratio. A lot of those liabilities were owed to the firm's counterparties but offset by other counterparties who were getting ready to declare bankruptcy.

You do realize that if a counterparty goes bankrupt and cannot pay its debts to Goldman then Goldman doesn't have to pay its debts to the counterparty, right? You can't simply look at liability ratio and assume that all of the creditors were different than all of the debtors.

So you tell me. Could GS have handled 50-60% of its counterparties going bankrupt with its 95% liability ratio? I think anyone with a basic accounting background would be pretty skeptical.

I don't know, because I don't have a copy of their balance sheet in front of me, as I'm not a GS employee (and I'm sure that even most GS employees don't get access to that info).

I do know that they take management of counterparty risk seriously, so I would think that a heck of a lot of their transactions ARE balanced out pretty well.

GS was better positioned than most investment banks for a market crisis, but the guys at GS clearly don't have the same wisdom that the conservative trust banks who supposedly "missed the boat on sub-prime mortgages and credit derivatives" did.

I don't really think it's useful to have a discussion about "wisdom". It's one of those qualities which is virtually impossible to measure meaningfully. I like sticking to results, and I'm not sure the position of GS was as precarious as you make it out to be. They certainly weren't in a comfortable position, but no firm which was at all plugged into the markets was at that time. Did GS benefit from the intervention of the US gov't into the financial system? Undoubtedly. I just find it a bit incongruous to single them out as having received some sort of extraordinary assistance, when all the indications are that they'd actually done a better job than most of managing their risks (GS and JPM seem to be the good guys, as far as such a thing existed!)

Also, it's pretty easy to look wise when markets implode if you just didn't take part. That doesn't mean that the real market participants weren't accomplishing something good overall. Some or all of them simply got a bit carried away. We wouldn't have stock market crashes without joint stock companies, yet I doubt many people would argue that this proves the wisdom of not participating in the stock market. Overexuberance is a human failure, not a failure of any particular market instrument.

Again, I was there in September, and it honestly looked like the end of the world (unless you worked at a conservative trust bank that research teams had spent the past seven years foolishly attacking for missing the boat on mortgage bonds and the credit derivatives markets). Several of my friends at Goldman Sachs (albeit with student loan debt) were saying they wished they worked at Harriman or BNY. If everyone on the street had gotten a libertarian bail-out, 90% of the banks would have disappeared, including Goldman Sachs. In fact, if you drew the counterparty graph, they were next in line after Citigroup and Morgan Stanley.

Just as overexuberance is a normal human failure, so is becoming overly cautious/risk-averse when things turn south. I don't trust reactions in the midst of the crisis any more than I trust statements about the strength of people's positions which were put out in 06/07. I trust numbers, but I also know how difficult it is to predict alternate histories. My major point is that it seems a bit ridiculous to go back and ask "what if, what if". What if the US gov't hadn't stepped up and provided what (at least now) appears to simply have been a ridiculous amount of liquidity? What if companies X,Y and Z had gone under due to this inaction? I don't really know, and I don't know if anybody does. At that point you're talking about the knock-on effects of counterparties of counterparties of counterparties. GS was insured against direct counterparty risk with AIG. It looks like it bought some protection against counterparties of counterparties too. Beyond that, I have no idea.

Thanks for the reference to the Buffett statement. However, it looks like he's saying that the market as a whole (not just financial companies) was going to suffer without some sort of government help. Doesn't seem particularly damning a statement about Goldman.

 

[bigbadwolf]

No, but it reduced the amount of money that the feds will either have to tax you for in the future, default on (thereby throwing the economy through a loop), or print (thereby debasing your currency).

I don't believe it (although that is the theoretical argument). The premise of the argument is a rational financial and economic system. That doesn't exist. The deficit spending is being piled on relentlessly and with no forethought. I don't know how many "billions" the taxpayer theoretically made but that is offset a hundred times -- maybe more -- by profligate spending on several different fronts. People like myself will either not see a social security cheque, or it will be in debased dollars in a wildly inflationary regime. People like myself are going to see our medicare drastically cut back when we retire. The argument that the taxpayer "made money" off the bailouts is camouflage for something that shouldn't have happened. In principle, the government should not be assisting GS and other large firms. These are the same people who have been singing the praises of unfettered capitalism -- but promptly take recourse in what they would previously have called "socialist measures" when it suits them. I don't want it done in my name nor do I want to be told how much money I theoretically made (but which I will never see) in a corrupt political deal.

 

[GoIllini]

You do realize that if a counterparty goes bankrupt and cannot pay its debts to Goldman then Goldman doesn't have to pay its debts to the counterparty, right? You can't simply look at liability ratio and assume that all of the creditors were different than all of the debtors.

That's true- you typically get right of offset, but at the same time, your individual exposure to each counterparty is going to look in the aggregate like the inverse of their market view. When the markets become very correlated as they did in September, all of your counterparties that were betting on a milder recession owed you money- and probably were in weaker financial shape. And naturally, if the same counterparty enters into an offsetting trade, you get to take that off your balance sheet.

Also, the right-of-offset can get tied up under bankruptcy protection. I would note that JP Morgan ran into trouble when it tried to bring Lehman's collateral onto its books. Creditors can wait if they think there might be a solvency issue, but if there's a liquidity crisis, you don't have time to get your collateral out of bankruptcy court.

So again, we could have been looking at a potential 50% default rate with counterparties. Do you mean to claim that there is an immediate 90% recovery rate on those defaults AND that on top of that, Goldman could get the necessary liquidity on 95% of its entire balance sheet to meet its margin calls during a financial crisis? Smart people can disagree on stuff, but I'm inclined to be pretty skeptical about their ability to stay liquid if the crisis had played out for another two weeks.

I don't know, because I don't have a copy of their balance sheet in front of me, as I'm not a GS employee (and I'm sure that even most GS employees don't get access to that info).

Their balance sheet is public record- you can certainly discover their liability ratio pretty easily via Yahoo. Just take liabilities divided by assets (actually do that for basically all of the banks back in September) and you'll understand why we were in a lot of trouble if there was a system-wide problem with counterparty risk.

I do know that they take management of counterparty risk seriously, so I would think that a heck of a lot of their transactions ARE balanced out pretty well.

Yeah, everyone takes counterparty risk pretty seriously, now. And if GS had eliminated all counterparty risk and all other market exposures back in September, they would not have needed to report a 95% liability ratio.

I don't really think it's useful to have a discussion about "wisdom". It's one of those qualities which is virtually impossible to measure meaningfully. I like sticking to results, and I'm not sure the position of GS was as precarious as you make it out to be. They certainly weren't in a comfortable position, but no firm which was at all plugged into the markets was at that time. Did GS benefit from the intervention of the US gov't into the financial system? Undoubtedly. I just find it a bit incongruous to single them out as having received some sort of extraordinary assistance, when all the indications are that they'd actually done a better job than most of managing their risks (GS and JPM seem to be the good guys, as far as such a thing existed!)

Not true at all. Northern Trust had a huge business- actually bigger than GS- as an asset manager and was also heavily involved in the mortgage business. They were probably a little nervous back in September, but I don't think any of their creditors lost sleep. Compare that with GS.

Also, it's pretty easy to look wise when markets implode if you just didn't take part. That doesn't mean that the real market participants weren't accomplishing something good overall.

Subprime mortgages accomplished empty and dilapidated McMansions in the deserts of Arizona and California. Smart investors who focus on the fundamentals try to avoid things that will wind up that way, whether or not the price could go up in the near-term. It helps us sleep better.

Some or all of them simply got a bit carried away. We wouldn't have stock market crashes without joint stock companies, yet I doubt many people would argue that this proves the wisdom of not participating in the stock market. Overexuberance is a human failure, not a failure of any particular market instrument.

The question is whether you should participate in a bubble. Buffet basically took Berkshire Hathaway out of the market from 1970 to 1974 because he knew it was fundamentally overvalued. He also cut back from 2007 to 2009.

My view is that if you know the market is overpriced, you shouldn't buy into it on the hopes that a bigger sucker will buy your stock later. It's a dangerous and possibly immoral business model.

Thanks for the reference to the Buffett statement. However, it looks like he's saying that the market as a whole (not just financial companies) was going to suffer without some sort of government help. Doesn't seem particularly damning a statement about Goldman.

I think if you search through the news back then, he clarifies that he would never have bought into Goldman Sachs if he didn't think there would be a bailout.

 

[NeedOPT]

Their balance sheet is public record- you can certainly discover their liability ratio pretty easily via Yahoo. Just take liabilities divided by assets (actually do that for basically all of the banks back in September) and you'll understand why we were in a lot of trouble if there was a system-wide problem with counterparty risk.

I don't mean the one which tells you assets, liabilities and broken up by class. I mean the master list of every single position held with every single counterparty.

I don't know that Goldman would have remained solvent, but I also don't think it's obvious they would have been driven into the ground. Again, I'm not sure it's even a meaningful question.

Not true at all. Northern Trust had a huge business- actually bigger than GS- as an asset manager and was also heavily involved in the mortgage business. They were probably a little nervous back in September, but I don't think any of their creditors lost sleep. Compare that with GS.

I think you misunderstand me. I'm not saying that Northern Trust wasn't doing business; what I am saying is that there IS net positive value in the financial instruments they chose not to (I'm assuming this is what you're claiming) trade in. Just because there wasn't as much value in those instruments as some people hoped doesn't mean that it's necessarily "wise" to simply steer clear of all innovations. By the way, I know virtually nothing about NT's business. Good for them to not get hit by the crisis, but there is room (and the need!) for both innovators and people who hold out from innovation in the marketplace. Personally, I didn't feel the need to mock the Northern Trusts of the world when they didn't take part in '06 and I don't feel the need to mock the GSes of the world in '09 because they did. If you are going to take part in creating/building a new market then I think you should do your homework, and I respect GS for managing its risks better than most of the other participants.

The question is whether you should participate in a bubble. Buffet basically took Berkshire Hathaway out of the market from 1970 to 1974 because he knew it was fundamentally overvalued. He also cut back from 2007 to 2009.

Warren Buffett is one of the few examples of men who CAN reliably value assets based on fundamentals (though he's been very wrong a few times). Good for him to call things right, but not everybody's going to be as smart as him every time.

By the way, "participating in a bubble" is the sort of meaningless drivel I absolutely hate reading in newspapers because it tells me that the reporter writing it has almost no clue what he's talking about. Who the hell "participates in a bubble" by buying assets they know are overvalued? If "participating in the bubble" means selling assets short then great! We need all the smart money possible to participate in and deflate bubbles. When somebody does "participate in a bubble" by buying overvalued assets then they've simply made a mistake valuing the asset. Or are you making some sort of Taibbian claim about a giant conspiracy to inflate asset bubbles and then profit from their popping?

Yeah, everyone takes counterparty risk pretty seriously, now. And if GS had eliminated all counterparty risk and all other market exposures back in September, they would not have needed to report a 95% liability ratio.

Errr....yes they would have. We just got through discussing this. Liability ratio is not a meaningful measure of counterparty risk without detailed knowledge of the parties involved.

My view is that if you know the market is overpriced, you shouldn't buy into it on the hopes that a bigger sucker will buy your stock later. It's a dangerous and possibly immoral business model.

So you are making some sort of ridiculous conspiracy theory claim. How the hell is Goldman smart enough to ride bubbles it knows are bubbles in order to make money off of the suckers yet at the same time so stupid as not to realize that the suckers it eventually sellls to represent counterparty risk?

Come off it, man. Nobody sits around saying "yeah, I know this market is dangerously overvalued, but I'm going to hold on longer and make money by selling right before the peak". At least no company that's been around for 150 years does that.

 

[GoIllini]

I think you misunderstand me. I'm not saying that Northern Trust wasn't doing business; what I am saying is that there IS net positive value in the financial instruments they chose not to (I'm assuming this is what you're claiming) trade in.

My view is that there's generally no net positive value in helping to build houses that nobody will live in if the materials and labor carry a significant cost. Whenever you get into any market, you need to bring a careful eye to the fundamentals and understand exactly how value is getting created. A good firm with an eye towards the long-term will also be on the lookout for its clients and make sure that when they buy something, they're getting a good value.

Houses occupied by people who can't afford them generally don't have the value of the resources that went into building them; why would you want to sell mortgage securities on those houses to your clients?

By the way, "participating in a bubble" is the sort of meaningless drivel I absolutely hate reading in newspapers because it tells me that the reporter writing it has almost no clue what he's talking about. Who the hell "participates in a bubble" by buying assets they know are overvalued? If "participating in the bubble" means selling assets short then great! We need all the smart money possible to participate in and deflate bubbles. When somebody does "participate in a bubble" by buying overvalued assets then they've simply made a mistake valuing the asset. Or are you making some sort of Taibbian claim about a giant conspiracy to inflate asset bubbles and then profit from their popping?

Actually, it's just as stupid, because your counterparty is probably going to wind up bankrupt unless they can pay cash when you short. Mortgage securities and credit derivatives are traded over the counter and with a lot of credit products, unless the counterparty is putting up the entire protection as collateral, you're looking at a pretty asymmetrical outcome on the trade if you believe there is a bubble that will pop.

Errr....yes they would have. We just got through discussing this. Liability ratio is not a meaningful measure of counterparty risk without detailed knowledge of the parties involved.

And we just got through discussing why your point is invalid during a market crisis. When there is a system-wide financial crisis, who your counterparty is matters a whole lot less than your leverage. I think you were in school back in 2008, so maybe you've already forgotten, but by mid-September, people stopped caring about the creditworthiness of their counterparties- they just wanted out of positions they were ahead on because they knew a crisis would take down pretty much everyone. Sometimes the market stops making sense and when that happens, you want to have lots of liquidity and minimal leverage.

So you are making some sort of ridiculous conspiracy theory claim. How the hell is Goldman smart enough to ride bubbles it knows are bubbles in order to make money off of the suckers yet at the same time so stupid as not to realize that the suckers it eventually sellls to represent counterparty risk?

Hardly. I'm actually saying they got lucky. In fact, they posted huge losses for the first time in the firm's history in the fourth quarter. Northern Trust actually posted some pretty hefty profits during the same time. (No, I don't work at NTRS).

Come off it, man. Nobody sits around saying "yeah, I know this market is dangerously overvalued, but I'm going to hold on longer and make money by selling right before the peak". At least no company that's been around for 150 years does that.

Then explain why they were making a market and therefore keeping an inventory in all of this stuff. At the very least, why would you want to put your clients through that if you knew there was a bubble?

My view is that GS either completely ignored the fact that credit derivatives were way out of whack in 2006 and 2007 that Northern Trust hit, or they foolishly stumbled into a bubble. Arrogance tends to cancel out experience when it comes to wisdom.

 

[NeedOPT]

And we just got through discussing why your point is invalid during a market crisis. When there is a system-wide financial crisis, who your counterparty is matters a whole lot less than your leverage.

No, you just got through talking about collateral. I'm not talking about collateral. I'm talking about about offsetting positions. Counterparty risk is limited to the positive value is the SUM TOTAL OF CONTRACTS WITH THAT COUNTERPARTY. It's not just the sum of positive valued contracts. The negative value contracts offset. Smart companies try to manage their counterparty risk so that they hold some negative positions and some positive positions with any given counterparty. Collateral is a second-best solution precisely because seizure of collateral is sometimes a bit slower than you'd hope, and CDS is for when you can't extract enough collateral.

Actually, it's just as stupid, because your counterparty is probably going to wind up bankrupt unless they can pay cash when you short.

Err....no. That's why you insist on settlement daily/weekly/monthly. You don't just let giant positive positions build up.

In fact, they posted huge losses for the first time in the firm's history in the fourth quarter.

I wouldn't exactly call them "huge losses". Even with the orphan month in Dec. their losses represented only a small fraction of the firm's net worth.

Hardly. I'm actually saying they got lucky.

If they "got lucky" then how the hell is it their "business model" and "immoral"? Goldman's not prescient. They just do a bit better of a job in squeezing a buck than a lot of people.

 

[GoIllini]

Smart companies try to manage their counterparty risk so that they hold some negative positions and some positive positions with any given counterparty

Again, I wish your risk manager the best of luck at that when the equities markets are moving 7% in a day and you're seeing 200 basis point swings in credit spreads. If it wasn't for TARP, Goldman would have run out liquidity in the second or third day of the October panic if it wouldn't have occurred earlier.

Some of your counterparties are accruing net positive positions that they now want out of or want collateral on. Others are going bankrupt and owe you money. Maybe you had everything balanced a day or two ago, but now things have changed dramatically before you barely even had the chance to get all of your new marks-to-market from the pricing system. This is where your liability ratio (along with your counterparties' leverage) starts to become a huge factor in your survival. If you have lots of equity and unencumbered cash or marginable assets, you stand a better chance of surviving when a financial earthquake hits and your counterparties' margin balances are reshuffled before you even have a chance to react.

Err....no. That's why you insist on settlement daily/weekly/monthly. You don't just let giant positive positions build up.

Exactly, but what assumptions are you making here? Maybe the following questions can help clear things up:

What happens when there is a market dislocation like we saw in early October and the market moves 7% before you even get a chance to make margin calls to half your counterparties?

What happens when the bankruptcy judge freezes the collateral as an asset of your counterparty when it files for bankruptcy protection and in the meantime, all of your counterparties are clamoring for collateral from you?

The assumptions you're making about counterparty risk were ones everyone was making back in September, and they have now learned from. The only option is to manage your counterparty risk by keeping the net potential liabilities under control relative to your balance sheet, or to make sure no counterparty's net mark-to-market moves much during a severe market dislocation. The second option means your counterparty has no reason for trading with you and you go out of business, so you're really only left with the first option.

If they "got lucky" then how the hell is it their "business model" and "immoral"? Goldman's not prescient. They just do a bit better of a job in squeezing a buck than most investment banks.

*Fixed*

I wouldn't exactly call them "huge losses". Even with the orphan month in Dec. their losses represented only a small fraction of the firm's net worth.

Typical investors would consider $2.2 Billion a huge loss.

 

[alain]

...
In addition, according to two people with knowledge of the positions, a portion of the $11 billion in taxpayer money that went to Socit Gnrale, a French bank that traded with A.I.G., was subsequently transferred to Goldman under a deal the two banks had struck.
...

OH!!!!!

http://www.nytimes.com/2010/02/07/business/07goldman.html?hp=&pagewanted=all