Michael Osinski wrote the software that turned mortgages into bonds
March 30, 2009 2:09 PM Subscribe
The article seems to have two excuses: I didn't get a lot of money for it, so my responsibility is limited. And I certainly did not get as much money as these very very bad boys did; look at how bad they are!
To me, it has seemed that each layer of abstraction of the concepts of wealth and income removed from physical, tangible things creates, almost by default, yet another, broader set of error bars around How Much Do You Really Have.
First you have a plot of land and a structure on it, and, barring the government or the bank taking it away from you, you know what you have. Then comes the mortgage layer, where you say "I think, and the market thinks, this is worth this much (* estimation), lend me this money, I will get it back to you and then some." Then a layer where you estimate how much you can earn (* estimation) from that mortgage. But that's too uncertain, so if we get a bunch of them together, the risks cancel out (* assumption). And I'll get this guy who rated that package (* abstraction) AAA last time to rate this package this time (* inadvertent collusion). And let's trade these abstractions of estimations of estimations, creating a market for them.
It's amazing, we have these teeny little houses on the bottom, and more and more concepts of wealth built atop these houses, wobbling from one side to another. When one layer goes, all the layers above go, so that a single foreclosed home triggers this storm of money (* abstraction) flying about.
Not that we could return to barter, but it seems clear that each layer of abstraction should come with greater and greater regulation, not less.
posted by adipocere at 2:40 PM on March 30, 2009 [6 favorites]
To me, it has seemed that each layer of abstraction of the concepts of wealth and income removed from physical, tangible things creates, almost by default, yet another, broader set of error bars around How Much Do You Really Have.
First you have a plot of land and a structure on it, and, barring the government or the bank taking it away from you, you know what you have. Then comes the mortgage layer, where you say "I think, and the market thinks, this is worth this much (* estimation), lend me this money, I will get it back to you and then some." Then a layer where you estimate how much you can earn (* estimation) from that mortgage. But that's too uncertain, so if we get a bunch of them together, the risks cancel out (* assumption). And I'll get this guy who rated that package (* abstraction) AAA last time to rate this package this time (* inadvertent collusion). And let's trade these abstractions of estimations of estimations, creating a market for them.
It's amazing, we have these teeny little houses on the bottom, and more and more concepts of wealth built atop these houses, wobbling from one side to another. When one layer goes, all the layers above go, so that a single foreclosed home triggers this storm of money (* abstraction) flying about.
Not that we could return to barter, but it seems clear that each layer of abstraction should come with greater and greater regulation, not less.
posted by adipocere at 2:40 PM on March 30, 2009 [6 favorites]
so that a single foreclosed home triggers this storm of money (* abstraction) flying about
I would argue that it was a lot more than a single foreclosure. The expansion of lending to sup-prime borrowers and the securitization of mortgages into CDOs are separate things. The popularity of CDOs led in some part of the expansion of the market for bad loans, but it wasn't the sole cause. If we'd only had one of these two things and not the other the situation wouldn't be nearly as bad.
posted by GuyZero at 2:48 PM on March 30, 2009
I would argue that it was a lot more than a single foreclosure. The expansion of lending to sup-prime borrowers and the securitization of mortgages into CDOs are separate things. The popularity of CDOs led in some part of the expansion of the market for bad loans, but it wasn't the sole cause. If we'd only had one of these two things and not the other the situation wouldn't be nearly as bad.
posted by GuyZero at 2:48 PM on March 30, 2009
Nice read.
I can't really blame the guy that wrote the software for the mess. I blame regulators and the banks who didn't do any due diligence on the loans they were writing and foisting off on the market.
Well, that and synthetic CDOs and the backed-by-nothing 'insurance' of CDS.
posted by wierdo at 2:49 PM on March 30, 2009
I can't really blame the guy that wrote the software for the mess. I blame regulators and the banks who didn't do any due diligence on the loans they were writing and foisting off on the market.
Well, that and synthetic CDOs and the backed-by-nothing 'insurance' of CDS.
posted by wierdo at 2:49 PM on March 30, 2009
The article seems to have two excuses: I didn't get a lot of money for it, so my responsibility is limited. And I certainly did not get as much money as these very very bad boys did; look at how bad they are!
"I was just following market orders."
posted by Blazecock Pileon at 3:04 PM on March 30, 2009 [5 favorites]
"I was just following market orders."
posted by Blazecock Pileon at 3:04 PM on March 30, 2009 [5 favorites]
The packaging of heterogeneous home mortgages into uniform securities that can be accurately priced and exchanged has been singled out by many critics as one of the root causes of the mess we’re in.To me, it seems a lot more like the packaging of heterogenous home mortgages into uniform securities that can't be accurately priced.
Or, at the very least, weren't and still aren't.
posted by Flunkie at 3:11 PM on March 30, 2009
"The aim of software is, in a sense, to create an alternative reality. After all, when you use your cell phone, you simply want to push the fewest buttons possible and call, text, purchase, listen, download, e-mail, or browse. The power we all hold in our hands is shocking, yet it’s controlled by a few swipes of a finger. The drive to simplify the user’s contact with the machine has an inherent side effect of disguising the complexity of a given task. Over time, the users of any software are inured to the intricate nature of what they are doing."
Sounds like a coherence-driven approach being applied to a correspondence-driven problem. How do you make a nuclear power plant 'user friendly'?
posted by anthill at 3:12 PM on March 30, 2009 [1 favorite]
Sounds like a coherence-driven approach being applied to a correspondence-driven problem. How do you make a nuclear power plant 'user friendly'?
posted by anthill at 3:12 PM on March 30, 2009 [1 favorite]
adipocere: "The article seems to have two excuses: I didn't get a lot of money for it, so my responsibility is limited. And I certainly did not get as much money as these very very bad boys did; look at how bad they are!"
From the article:
Why I had been able to retire at 45 for simply writing a computer program befuddled me and aggravated others who felt they had worked as hard. Life is not always fair, I told them.
Retiring at 45 counts as "lots of money" to me.
Banned from Wall Street, I discovered that my summer house, on the North Fork of Long Island, included five acres of underwater land. I applied for permits to grow oysters.
Not that oyster fishers tend to be readers of New York magazine. But I wonder how they would respond to this.
posted by Joe Beese at 3:12 PM on March 30, 2009
From the article:
Why I had been able to retire at 45 for simply writing a computer program befuddled me and aggravated others who felt they had worked as hard. Life is not always fair, I told them.
Retiring at 45 counts as "lots of money" to me.
Banned from Wall Street, I discovered that my summer house, on the North Fork of Long Island, included five acres of underwater land. I applied for permits to grow oysters.
Not that oyster fishers tend to be readers of New York magazine. But I wonder how they would respond to this.
posted by Joe Beese at 3:12 PM on March 30, 2009
Retiring at 45 counts as "lots of money" to me, too. Place it, though, within the context of the article:
"I got a $50,000 bonus check, a 50 percent dollop on top of my salary. Peanuts to the traders, but a bloody fortune to me, for the easiest work I’d ever done."
"I made $125,000 in my bonus that year and bought an apartment on Gramercy Park. I had first-tier seats to the ballet, but I still rode my bike to work. The traders pocketed multiple millions. I wasn’t poor, but I wasn’t a plutocrat."
He keeps placing his income against the excesses of the traders. And we get additional informative, though not entirely relevant, bits about peeing for distance. It just has this "yeah, I did it ... but look at these guys over here! feel to it."
posted by adipocere at 3:26 PM on March 30, 2009 [1 favorite]
"I got a $50,000 bonus check, a 50 percent dollop on top of my salary. Peanuts to the traders, but a bloody fortune to me, for the easiest work I’d ever done."
"I made $125,000 in my bonus that year and bought an apartment on Gramercy Park. I had first-tier seats to the ballet, but I still rode my bike to work. The traders pocketed multiple millions. I wasn’t poor, but I wasn’t a plutocrat."
He keeps placing his income against the excesses of the traders. And we get additional informative, though not entirely relevant, bits about peeing for distance. It just has this "yeah, I did it ... but look at these guys over here! feel to it."
posted by adipocere at 3:26 PM on March 30, 2009 [1 favorite]
This is rich: The software proved to be more sophisticated than the people who used it
It's not a bug, you're not sophisticated enough to use it.
posted by digsrus at 3:30 PM on March 30, 2009
It's not a bug, you're not sophisticated enough to use it.
posted by digsrus at 3:30 PM on March 30, 2009
He lost me at "provided banks don’t resume writing mortgages to people who can’t afford them". By lost, I mean, I realized that here is just another apologist for the greed and excesses of the Masters of The Universe. With less than zero credibility. Literally. No point in continuing.
Here's what I mean. Do the math yourself. Don't ask me. Google "mortgage default rate". I dare you. I double dog dare you.
Here's some real rough numbers for you. By rough, I mean that I am just ballparking this. Not rough as in, will make you cry. But, ah, you be the judge.
The mortgage default rates over the last ten years, depending on your sources, ran about 2% to 2.5% each year. In 2008 they jumped to between 2.5% and 3%. Somewhere between an increase between 25% and 50%; let's just say it went from 2% to 3%, a 50% increase in the rate of default, but only 1% more of the total first home mortgages. 50% sounds like a safe bet, maybe a little on the pessimistic side.
So, for every 100 dollars the banks risked, they lost three - in 2008. Except they didn't actually lose it, they lost the loan value; they kept the collateral, the house, which we know is less than the loan value. But nevertheless, it's not a loss of 3 cents on the dollar, it's better than that. But I digress.
Yes, I understand, that's still a lot of money. There were roughly 51 million first mortgages in 2008. Not new ones, overall. So, at 3%, that's 1,531,000 mortgage defaults.
The average home mortgage (new) in 2008 was $176,000. That does not include older mortgages, valued at less - but just for shits and grins, lets say that's the overall average. Sounds low, to me, but I live in California. We'll call it $200,000 just to be safe.
That's: $ 306,200,000,000
A LOT of money. 306 Billion dollars.
But wait. At least 2/3 of that is no different a loss than all the other years. In other words, there is some amount of money expected to be lost on people who get cancer, die, lose their limbs, jobs, etc. That's normal, and it's factored in. You can't win all the time, just 97%-98% of the time. The net increase in the default rate - at 50% - over the previous ten years average - would be...
$ 101,046,000,000
$ 101 Billion dollars.
There was an additional 101 Billion dollars lost in 2008 than they had expected based on historical numbers.
How much is the bailout? I call bullshit.
The REAL problem was that securitized mortgage bundles were insured through CDS's at a rate of 1:20 by companies that did not - and could not have possibly had - the assets to back up the claims. They lied, basically. They made bets they knew they could not possibly pay off if they lost.
That, my friends, is greed. Not by stupid, lazy, ignorant, thieving, lying, swine proletariat homebuyers, either.
posted by Xoebe at 4:31 PM on March 30, 2009 [29 favorites]
Here's what I mean. Do the math yourself. Don't ask me. Google "mortgage default rate". I dare you. I double dog dare you.
Here's some real rough numbers for you. By rough, I mean that I am just ballparking this. Not rough as in, will make you cry. But, ah, you be the judge.
The mortgage default rates over the last ten years, depending on your sources, ran about 2% to 2.5% each year. In 2008 they jumped to between 2.5% and 3%. Somewhere between an increase between 25% and 50%; let's just say it went from 2% to 3%, a 50% increase in the rate of default, but only 1% more of the total first home mortgages. 50% sounds like a safe bet, maybe a little on the pessimistic side.
So, for every 100 dollars the banks risked, they lost three - in 2008. Except they didn't actually lose it, they lost the loan value; they kept the collateral, the house, which we know is less than the loan value. But nevertheless, it's not a loss of 3 cents on the dollar, it's better than that. But I digress.
Yes, I understand, that's still a lot of money. There were roughly 51 million first mortgages in 2008. Not new ones, overall. So, at 3%, that's 1,531,000 mortgage defaults.
The average home mortgage (new) in 2008 was $176,000. That does not include older mortgages, valued at less - but just for shits and grins, lets say that's the overall average. Sounds low, to me, but I live in California. We'll call it $200,000 just to be safe.
That's: $ 306,200,000,000
A LOT of money. 306 Billion dollars.
But wait. At least 2/3 of that is no different a loss than all the other years. In other words, there is some amount of money expected to be lost on people who get cancer, die, lose their limbs, jobs, etc. That's normal, and it's factored in. You can't win all the time, just 97%-98% of the time. The net increase in the default rate - at 50% - over the previous ten years average - would be...
$ 101,046,000,000
$ 101 Billion dollars.
There was an additional 101 Billion dollars lost in 2008 than they had expected based on historical numbers.
How much is the bailout? I call bullshit.
The REAL problem was that securitized mortgage bundles were insured through CDS's at a rate of 1:20 by companies that did not - and could not have possibly had - the assets to back up the claims. They lied, basically. They made bets they knew they could not possibly pay off if they lost.
That, my friends, is greed. Not by stupid, lazy, ignorant, thieving, lying, swine proletariat homebuyers, either.
posted by Xoebe at 4:31 PM on March 30, 2009 [29 favorites]
adipocere, you got it in one. It's precisely the abstraction of wealth that's messing us up so bad -- it allows those making the abstractions to leverage themselves up and profit enormously with relatively little actual capital. It's an enormous structure built on top of the real economy, and the weird pressures it exerts have eroded the base badly.
Leverage creates risk; that's kind of the whole point of leverage. You use limited capital to take on more risk, in exchange for more reward. They thought they were offsetting it with other derivatives, but even when they moved the risk off their balance sheet, they just took on more risk. Being able to sell risk to some other guy meant that any trader would take any opportunity, no matter how insane, to boost this year's bonus. And, with the Fed's full backing, these dollar-denominated derivatives functioned more or less like money. With that much 'money' floating around the system, we had a pair of giant bubbles, debt and real estate, both of which are now trying very hard to crash, while the Fed and Congress desperately try to keep the fiction running.
I don't really blame this guy. He was a programmer. He didn't force anyone to take on more risk than they could handle. It's pretty unlikely he had any idea that it could be so toxic. Certainly Alan Greenspan, the head of the most-central central bank, was a big cheerleader of the concept. How was one dude with his head buried in code supposed to figure out what all those economic superstars couldn't?
posted by Malor at 4:56 PM on March 30, 2009
Leverage creates risk; that's kind of the whole point of leverage. You use limited capital to take on more risk, in exchange for more reward. They thought they were offsetting it with other derivatives, but even when they moved the risk off their balance sheet, they just took on more risk. Being able to sell risk to some other guy meant that any trader would take any opportunity, no matter how insane, to boost this year's bonus. And, with the Fed's full backing, these dollar-denominated derivatives functioned more or less like money. With that much 'money' floating around the system, we had a pair of giant bubbles, debt and real estate, both of which are now trying very hard to crash, while the Fed and Congress desperately try to keep the fiction running.
I don't really blame this guy. He was a programmer. He didn't force anyone to take on more risk than they could handle. It's pretty unlikely he had any idea that it could be so toxic. Certainly Alan Greenspan, the head of the most-central central bank, was a big cheerleader of the concept. How was one dude with his head buried in code supposed to figure out what all those economic superstars couldn't?
posted by Malor at 4:56 PM on March 30, 2009
Holy crap, it's The Metamorphosis of Subprime Intellect.
posted by localroger at 5:37 PM on March 30, 2009 [4 favorites]
posted by localroger at 5:37 PM on March 30, 2009 [4 favorites]
This guy seems to have an Asperger's like detachment from being able to see how his behavior affects the lives of others. He doesn't seem like he sees other people as real. They're just abstractions and it's too bad if they get messed up.
posted by Maias at 6:13 PM on March 30, 2009 [2 favorites]
posted by Maias at 6:13 PM on March 30, 2009 [2 favorites]
larger issues?
what (if any) software is it immoral to create?
are the writers of software responsible for how it is used?
if this author had refused to write his software, would the market demand go unfulfilled (would someone else write it, and the consequences be the same)?
could different software produced different outcomes (different back-end logic, different front end experiences)?
posted by el io at 7:26 PM on March 30, 2009
what (if any) software is it immoral to create?
are the writers of software responsible for how it is used?
if this author had refused to write his software, would the market demand go unfulfilled (would someone else write it, and the consequences be the same)?
could different software produced different outcomes (different back-end logic, different front end experiences)?
posted by el io at 7:26 PM on March 30, 2009
Wall Street has not been blown up, and that may be part of the problem:
"We have a financial system that is run by private shareholders, managed by private institutions, and we’d like to do our best to preserve that system.”
posted by ornate insect at 8:22 PM on March 30, 2009
"We have a financial system that is run by private shareholders, managed by private institutions, and we’d like to do our best to preserve that system.”
posted by ornate insect at 8:22 PM on March 30, 2009
The "abstraction" of wealth that contributed to this disaster reminds me of the "abstraction" that allowed people to rationalize genocide (e.g., the Holocaust). Or the abstraction that the meat in the nice plastic wrap didn't come from a sentient organism.
posted by cogneuro at 8:30 PM on March 30, 2009 [1 favorite]
posted by cogneuro at 8:30 PM on March 30, 2009 [1 favorite]
Banned from Wall Street, I discovered that my summer house, on the North Fork of Long Island, included five acres of underwater land. I applied for permits to grow oysters.
Many Americans have property which is underwater at the moment.
posted by atrazine at 8:53 PM on March 30, 2009 [1 favorite]
Many Americans have property which is underwater at the moment.
posted by atrazine at 8:53 PM on March 30, 2009 [1 favorite]
I blame the lawmakers and ineffectual regulating bodies. Fer crissakes, these guys were worse than FEMA!
Somewhere right now, some real-world Mr. Burns is reading these words, rubbing his hands together and saying "Excellent, excellent..."
Yeah, you know, one thing that occurred to me recently is that so much of the more recent "Conservative" arguments rests on the claim that government is inherently too incompetent or corrupt to regulate markets.
Well, it's pretty obvious by now that markets aren't capable of regulating themselves, and in any event, if they were, that argument wouldn't hold any weight. After all, the government, according to this line of thought, is only too incompetent to regulate the markets because markets are too good at bucking the intended regulatory outcomes and seeking profit maximization anyway--the argument isn't that people won't still lie, cheat, and steal in the absence of regulation (at least, I've never heard anyone make such a stupid claim), it's more that since people are still able to lie, cheat and steal despite whatever regulations we might impose, why should we even bother trying to stop them? Especially if it costs me money?
When you realize that that's essentially all this particular "Conservative" argument comes down to--basically, "We can't stop people from ripping us off, and I've managed not to get ripped off yet, so what gives you the right to use my tax dollar to stop scammers?"--it's not hard to imagine where this kind of thinking ultimately leads.
posted by saulgoodman at 8:53 PM on March 30, 2009 [1 favorite]
Somewhere right now, some real-world Mr. Burns is reading these words, rubbing his hands together and saying "Excellent, excellent..."
Yeah, you know, one thing that occurred to me recently is that so much of the more recent "Conservative" arguments rests on the claim that government is inherently too incompetent or corrupt to regulate markets.
Well, it's pretty obvious by now that markets aren't capable of regulating themselves, and in any event, if they were, that argument wouldn't hold any weight. After all, the government, according to this line of thought, is only too incompetent to regulate the markets because markets are too good at bucking the intended regulatory outcomes and seeking profit maximization anyway--the argument isn't that people won't still lie, cheat, and steal in the absence of regulation (at least, I've never heard anyone make such a stupid claim), it's more that since people are still able to lie, cheat and steal despite whatever regulations we might impose, why should we even bother trying to stop them? Especially if it costs me money?
When you realize that that's essentially all this particular "Conservative" argument comes down to--basically, "We can't stop people from ripping us off, and I've managed not to get ripped off yet, so what gives you the right to use my tax dollar to stop scammers?"--it's not hard to imagine where this kind of thinking ultimately leads.
posted by saulgoodman at 8:53 PM on March 30, 2009 [1 favorite]
Metafilter: not sophisticated enough to use it.
posted by Michael Roberts at 9:02 PM on March 30, 2009
posted by Michael Roberts at 9:02 PM on March 30, 2009
This is quite a blast from the past. My little company was Mike Osinski's Internet consultant and his ISP back in 1995-97, helping him put his Gramercy Park apartment/office on-line and hosting his e-mail, while he was starting up the business he describes in the article, selling his mortgage-securitization software to investment banks at (yes, GuyZero) a half-million bucks a pop. I remember him as a smart, genial, enthusiastic "real programmer" type who had written a lot of good modular code and was making terrific money during the Silicon V/Alley bubble without being part of it.
He's a problem solver, and the problem was "how do we manage and track the securitization of home loans" — he wasn't one of the evil quants who "solved" the "how do we model millions of loans so that we can predict, sort, and sell their risk" problem, or the managers who "applied" those models as if there was no such thing as a black swan.
He's a lot more conscious of his minor enabling responsibility for the mortgage-backed mess than the people we're hearing from who were directly involved; and his awareness that only his non-compete, not any moral instinct, prevented him from becoming complicit in the CDO debacle gives the lie to those — me included, sometimes — whose anger at the bad guys includes a lot of people that didn't do anything they themselves wouldn't.
posted by nicwolff at 9:11 PM on March 30, 2009 [5 favorites]
He's a problem solver, and the problem was "how do we manage and track the securitization of home loans" — he wasn't one of the evil quants who "solved" the "how do we model millions of loans so that we can predict, sort, and sell their risk" problem, or the managers who "applied" those models as if there was no such thing as a black swan.
He's a lot more conscious of his minor enabling responsibility for the mortgage-backed mess than the people we're hearing from who were directly involved; and his awareness that only his non-compete, not any moral instinct, prevented him from becoming complicit in the CDO debacle gives the lie to those — me included, sometimes — whose anger at the bad guys includes a lot of people that didn't do anything they themselves wouldn't.
posted by nicwolff at 9:11 PM on March 30, 2009 [5 favorites]
As a programmer myself (although not one working at the same level as the author of the piece), I actually don't hold the guy as personally accountable as he holds himself.
It's not like he was writing vote-rigging software or something. He didn't write code to rip people off. He wrote code to meet the end users business requirements. That's what programmers do. I'm sure a lot of dirty dealing on Wall Street gets done using software tools no more specialized or exotic than MS Word or Outlook. Now, maybe MS is evil for a host of other reasons, but it's not because they make products that people on Wall Steet can use to rip other people off. Nor is MS evil merely because they charge a lot for their products (well, that one's moot).
Still, his passing remark about the role played by "mortgages people couldn't afford" rankled me, too. But I think it's less apologia than simple confusion. For so long now, the press has beaten it into our heads that debt instruments referencing sub-prime mortgages were primarily what got us into this mess, without nuance and despite all evidence to the contrary.
It's become clearer and clearer in the intervening time that the much bigger problem is with the multitrillion dollar market for the even more questionable derivative devices (CDOs, synthetic CDOs, third-party swaps, etc.) that gained such currency after this guy left Wall Street.
I think like most of us, he's still a little confused about what the hell's actually going on here and that's understandable.
posted by saulgoodman at 9:40 PM on March 30, 2009 [1 favorite]
It's not like he was writing vote-rigging software or something. He didn't write code to rip people off. He wrote code to meet the end users business requirements. That's what programmers do. I'm sure a lot of dirty dealing on Wall Street gets done using software tools no more specialized or exotic than MS Word or Outlook. Now, maybe MS is evil for a host of other reasons, but it's not because they make products that people on Wall Steet can use to rip other people off. Nor is MS evil merely because they charge a lot for their products (well, that one's moot).
Still, his passing remark about the role played by "mortgages people couldn't afford" rankled me, too. But I think it's less apologia than simple confusion. For so long now, the press has beaten it into our heads that debt instruments referencing sub-prime mortgages were primarily what got us into this mess, without nuance and despite all evidence to the contrary.
It's become clearer and clearer in the intervening time that the much bigger problem is with the multitrillion dollar market for the even more questionable derivative devices (CDOs, synthetic CDOs, third-party swaps, etc.) that gained such currency after this guy left Wall Street.
I think like most of us, he's still a little confused about what the hell's actually going on here and that's understandable.
posted by saulgoodman at 9:40 PM on March 30, 2009 [1 favorite]
This guy seems to have an Asperger's like detachment from being able to see how his behavior affects the lives of others. He doesn't seem like he sees other people as real. They're just abstractions and it's too bad if they get messed up.
That's not "Asperger's like" that's how must people are when they are thinking about people they've never met. It's much easier to push a button to drop a bomb then it is to stab them.
posted by delmoi at 10:19 PM on March 30, 2009 [1 favorite]
That's not "Asperger's like" that's how must people are when they are thinking about people they've never met. It's much easier to push a button to drop a bomb then it is to stab them.
posted by delmoi at 10:19 PM on March 30, 2009 [1 favorite]
I honestly don't see how this guy is any more responsible for what happened when he wrote software to specifications provided to him by his employers than say, Mikhail Kalashnikov is responsible for the Taliban. Both made tools, what others did with those tools is on the others, not the creators.
Kalashnikov wasn't thinking of insurgents and ultrareligious tribal warfare when he designed his famous AK, he was thinking about how the Germans used their machineguns to turn Russians into hamburger during the latter days of WWII and that if the Russians had a comparable weapon maybe that would have worked out differently. And Osinski wasn't thinking about how his software would be misused to bring Wall Street's giants to their knees and shake the whole economy to its foundations, he was designing a software tool to automate processes the traders were already doing by hand. He made it faster, but it was the traders' idea, and they were looking for a software tool to help. If it wasn't him doing the work, someone else would have done it. And as I say, at the time he wrote it, nobody had any idea what stupidity the traders would be able to inflict with it.
posted by barc0001 at 1:43 AM on March 31, 2009 [1 favorite]
Kalashnikov wasn't thinking of insurgents and ultrareligious tribal warfare when he designed his famous AK, he was thinking about how the Germans used their machineguns to turn Russians into hamburger during the latter days of WWII and that if the Russians had a comparable weapon maybe that would have worked out differently. And Osinski wasn't thinking about how his software would be misused to bring Wall Street's giants to their knees and shake the whole economy to its foundations, he was designing a software tool to automate processes the traders were already doing by hand. He made it faster, but it was the traders' idea, and they were looking for a software tool to help. If it wasn't him doing the work, someone else would have done it. And as I say, at the time he wrote it, nobody had any idea what stupidity the traders would be able to inflict with it.
posted by barc0001 at 1:43 AM on March 31, 2009 [1 favorite]
At the heart of this issue of responsibility is a split between ends and means. In his account, the ultimate end of this kind of securitization was someone else's responsibility. He was a means guy, and evidently rather good at it, although I do find his narrative a little technical (in its somewhat fussy literariness, rather than in any software engineering sense).
The ends guys - senior management, regulators, whoever - mistook the software's technical coherence for its fitness for purpose in handling the messiness of the real world. I hold both of them responsible for this catastrophe. But I consider the ends guys more culpable, because they did not do their jobs, whereas this guy (and I've come across a lot like him) is merely rather short-sighted or selfish in imagining that just doing his job and then going for a bike ride or growing oysters was an acceptable way to live your life in a world full of others. Kind of pathetic, but understandable in a culture that places a fleeting premium on certain technical skill when those skills are rare.
The problem as I see it is in this disastrous bifurcation of technical competence and moral responsibility. Shouldn't we - don't we - have both the ability to do certain things, well or not, and the ability to reflect on them?
posted by GeorgeBickham at 6:18 AM on March 31, 2009
The ends guys - senior management, regulators, whoever - mistook the software's technical coherence for its fitness for purpose in handling the messiness of the real world. I hold both of them responsible for this catastrophe. But I consider the ends guys more culpable, because they did not do their jobs, whereas this guy (and I've come across a lot like him) is merely rather short-sighted or selfish in imagining that just doing his job and then going for a bike ride or growing oysters was an acceptable way to live your life in a world full of others. Kind of pathetic, but understandable in a culture that places a fleeting premium on certain technical skill when those skills are rare.
The problem as I see it is in this disastrous bifurcation of technical competence and moral responsibility. Shouldn't we - don't we - have both the ability to do certain things, well or not, and the ability to reflect on them?
posted by GeorgeBickham at 6:18 AM on March 31, 2009
I'll just grab two quotes, and spare you my vitriol.
“Mike,” he told me when denying my request, “can you really look for people dumber than you and then take advantage of them? That’s what trading is all about.”
Yes, I assured him, yes, yes.
The supply of raw material, new mortgages, disappeared. We had to lay off half of research. After a day of bloodletting, one of the bosses cornered me in the hallway. Did I get a sexual thrill out of firing people, he wanted to know, because it had always worked for him, big time.
posted by bitmage at 7:30 AM on March 31, 2009 [1 favorite]
“Mike,” he told me when denying my request, “can you really look for people dumber than you and then take advantage of them? That’s what trading is all about.”
Yes, I assured him, yes, yes.
The supply of raw material, new mortgages, disappeared. We had to lay off half of research. After a day of bloodletting, one of the bosses cornered me in the hallway. Did I get a sexual thrill out of firing people, he wanted to know, because it had always worked for him, big time.
posted by bitmage at 7:30 AM on March 31, 2009 [1 favorite]
I really think that some of you are combining hindsight with hostility.
The only people who I condemn are the bankers who made bad loans, the people who took loans they couldn't afford, and people who mixed the bad loans into securities.
It seems to me like this guy's software was only bad for the part where he facilitated subprime securities trading. The rest of it seems fine.
Having securities of good loans (some of which might occasionally go bad) still doesn't seem like a problem to me.
posted by HappyEngineer at 1:22 PM on March 31, 2009
The only people who I condemn are the bankers who made bad loans, the people who took loans they couldn't afford, and people who mixed the bad loans into securities.
It seems to me like this guy's software was only bad for the part where he facilitated subprime securities trading. The rest of it seems fine.
Having securities of good loans (some of which might occasionally go bad) still doesn't seem like a problem to me.
posted by HappyEngineer at 1:22 PM on March 31, 2009
The only people who I condemn are the bankers who made bad loans, the people who took loans they couldn't afford
HappyEngineer: Where have you been?
The really serious problem now--the thing that makes this not merely a downturn in the housing market with a quick recovery immediately visible on the horizon, as it would likely be if we were just dealing with an increase of a few points in prime or subprime default rates--are the multi-trillion dollar unregulated derivatives markets.
These products, instead of hedging risk exposure as advertised, in effect maximized everyone's risk.
Since buyers could take out third-party swap contracts on financial instruments they didn't even have a stake in, all of the institutions that issued swaps just turned around and bought swaps of their own on those same instruments they guaranteed from other institutions to hedge their own risks, and so that their own exposure appeared minimal on the balance sheets.
This process repeated with speculators and other investors joining in the fun until, eventually, everyone in the financial sector ended up exposed to more risk than ever existed in the first place. Instead of hedging risk, or even spreading risk around to minimize exposure, because of this particular loophole in the way the unregulated derivatives markets worked everyone ended up being more exposed.
Now in the over-leveraged, over exposed financial sector these kinds of practices led to, the failure of just one major financial institution can have a cascading effect that wipes out hundreds of others and renders trillions of dollars of assets completely worthless. (Here's a simple-stupid link on the subject.)
If it weren't for the bigger problems created by the unregulated derivatives markets, the downturn in the housing market due to the subprime collapse would be just another downturn. It wouldn't be a small issue, certainly, but it wouldn't be a deadly serious one either.
The bigger problems in the financial sector caused by the unregulated derivatives markets, on the other hand: this is a deadly serious issue.
posted by saulgoodman at 2:27 PM on March 31, 2009
HappyEngineer: Where have you been?
The really serious problem now--the thing that makes this not merely a downturn in the housing market with a quick recovery immediately visible on the horizon, as it would likely be if we were just dealing with an increase of a few points in prime or subprime default rates--are the multi-trillion dollar unregulated derivatives markets.
These products, instead of hedging risk exposure as advertised, in effect maximized everyone's risk.
Since buyers could take out third-party swap contracts on financial instruments they didn't even have a stake in, all of the institutions that issued swaps just turned around and bought swaps of their own on those same instruments they guaranteed from other institutions to hedge their own risks, and so that their own exposure appeared minimal on the balance sheets.
This process repeated with speculators and other investors joining in the fun until, eventually, everyone in the financial sector ended up exposed to more risk than ever existed in the first place. Instead of hedging risk, or even spreading risk around to minimize exposure, because of this particular loophole in the way the unregulated derivatives markets worked everyone ended up being more exposed.
Now in the over-leveraged, over exposed financial sector these kinds of practices led to, the failure of just one major financial institution can have a cascading effect that wipes out hundreds of others and renders trillions of dollars of assets completely worthless. (Here's a simple-stupid link on the subject.)
If it weren't for the bigger problems created by the unregulated derivatives markets, the downturn in the housing market due to the subprime collapse would be just another downturn. It wouldn't be a small issue, certainly, but it wouldn't be a deadly serious one either.
The bigger problems in the financial sector caused by the unregulated derivatives markets, on the other hand: this is a deadly serious issue.
posted by saulgoodman at 2:27 PM on March 31, 2009
Related: The Formula That Killed Wall Street:
...Li's formula, known as a Gaussian copula function, looked like an unambiguously positive breakthrough, a piece of financial technology that allowed hugely complex risks to be modeled with more ease and accuracy than ever before. With his brilliant spark of mathematical legerdemain, Li made it possible for traders to sell vast quantities of new securities, expanding financial markets to unimaginable levels.
His method was adopted by everybody from bond investors and Wall Street banks to ratings agencies and regulators. And it became so deeply entrenched—and was making people so much money—that warnings about its limitations were largely ignored.
Then the model fell apart. Cracks started appearing early on, when financial markets began behaving in ways that users of Li's formula hadn't expected. The cracks became full-fledged canyons in 2008—when ruptures in the financial system's foundation swallowed up trillions of dollars and put the survival of the global banking system in serious peril.
posted by ornate insect at 6:04 PM on March 31, 2009
...Li's formula, known as a Gaussian copula function, looked like an unambiguously positive breakthrough, a piece of financial technology that allowed hugely complex risks to be modeled with more ease and accuracy than ever before. With his brilliant spark of mathematical legerdemain, Li made it possible for traders to sell vast quantities of new securities, expanding financial markets to unimaginable levels.
His method was adopted by everybody from bond investors and Wall Street banks to ratings agencies and regulators. And it became so deeply entrenched—and was making people so much money—that warnings about its limitations were largely ignored.
Then the model fell apart. Cracks started appearing early on, when financial markets began behaving in ways that users of Li's formula hadn't expected. The cracks became full-fledged canyons in 2008—when ruptures in the financial system's foundation swallowed up trillions of dollars and put the survival of the global banking system in serious peril.
posted by ornate insect at 6:04 PM on March 31, 2009
Afroblanco, AIG has massive amounts of exposure due to its sales of unregulated swap contracts. Its obligations alone might be enough to ruin us all, from what I understand.
But unfortunately, AIG is also the biggest bond guarantor going, so untold numbers of municipal bonds (which cities and towns issue to fund infrastructure projects, fill short-term budget gaps, etc.) and other types of bonds would basically become worthless if AIG went bankrupt, leaving many states, towns and businesses who rely on issuing bonds dead in the water as their bonds become worthless and no one wants to invest in future issues anymore.
Not to mention all the banks that go bust because suddenly they've got all these worthless bond assets on their books. And then of course, a lot of those institutions probably took out default swaps on their bond holdings, and the bankruptcy laws under Bush were rewritten to require that swap contracts can't be dismissed through the receivership process and effectively jump to the line ahead of other obligations.
And on and on. Until finally, everybody, in one way or another, is busted, because the total value of all these derivatives is something like five times the total notional value of the underlying reference assets on account of all the speculative buys and side-bets people made on swaps and other exotic derivatives derived from swaps.
ironmouth does a good job of explaining these aspects of the problem here.
posted by saulgoodman at 7:11 PM on March 31, 2009
But unfortunately, AIG is also the biggest bond guarantor going, so untold numbers of municipal bonds (which cities and towns issue to fund infrastructure projects, fill short-term budget gaps, etc.) and other types of bonds would basically become worthless if AIG went bankrupt, leaving many states, towns and businesses who rely on issuing bonds dead in the water as their bonds become worthless and no one wants to invest in future issues anymore.
Not to mention all the banks that go bust because suddenly they've got all these worthless bond assets on their books. And then of course, a lot of those institutions probably took out default swaps on their bond holdings, and the bankruptcy laws under Bush were rewritten to require that swap contracts can't be dismissed through the receivership process and effectively jump to the line ahead of other obligations.
And on and on. Until finally, everybody, in one way or another, is busted, because the total value of all these derivatives is something like five times the total notional value of the underlying reference assets on account of all the speculative buys and side-bets people made on swaps and other exotic derivatives derived from swaps.
ironmouth does a good job of explaining these aspects of the problem here.
posted by saulgoodman at 7:11 PM on March 31, 2009
At the heart of this issue of responsibility is a split between ends and means. In his account, the ultimate end of this kind of securitization was someone else's responsibility. He was a means guy, and evidently rather good at it, although I do find his narrative a little technical (in its somewhat fussy literariness, rather than in any software engineering sense).
The ends guys - senior management, regulators, whoever - mistook the software's technical coherence for its fitness for purpose in handling the messiness of the real world. I hold both of them responsible for this catastrophe. But I consider the ends guys more culpable, because they did not do their jobs, whereas this guy (and I've come across a lot like him) is merely rather short-sighted or selfish in imagining that just doing his job and then going for a bike ride or growing oysters was an acceptable way to live your life in a world full of others. Kind of pathetic, but understandable in a culture that places a fleeting premium on certain technical skill when those skills are rare.
The problem as I see it is in this disastrous bifurcation of technical competence and moral responsibility. Shouldn't we - don't we - have both the ability to do certain things, well or not, and the ability to reflect on them?
posted by GeorgeBickham at 2:18 PM on March 31 [+] [!]
Well, the point is that there's nothing wrong with the ultimate end of securitization. It's a perfectly good idea that worked for some time, and he was providing the means for something that was - in his eyes at the time - a good thing.
And the 'ends guys' were correct in assessing his software as being fit for purpose. His software only handled "prime" mortgages (though only a non-compete contract kept him from extending it to handle sub-prime), and for that purpose it was perfectly suited.
The problem with securitising mortgages is that it because it makes the financing process more efficient it can allow bubbles to inflate much more rapidly. Previously, even if there was a sustained run of very low real interest rates, the rise in house prices was slower. Even if you did have a billion dollars you wanted to loan to house buyers RIGHT. NOW. You still couldn't easily do it, you'd need the infrastructure of local banks first. With securitisation, it was super easy for money to flow into mortgages.
But as saulgoodman points out, the real problem was the unregulated derivatives market.
It isn't so much that he didn't take the time to foresee the consequences of his actions, as that they were not reasonably forseeable.
So it's not so much the case that this guy is like a gunsmith. A gunsmith is obviously not responsible for all the uses of his weapons, but he does know that he is making tools that can kill.
The writer is more like a guy who sells a hammer which is used responsibly by most users, but which some irresponsible users use to build shaky houses.
Unknown to the original hammer vendor (now retired to farm oysters) these shaky houses are connected to all the well built houses with guy wires, but no-one has regulated the number of poorly built houses guyed to the well built ones! So now, when all the houses look they might come down together, is it really hammer guy's fault?
posted by atrazine at 1:29 AM on April 1, 2009
The ends guys - senior management, regulators, whoever - mistook the software's technical coherence for its fitness for purpose in handling the messiness of the real world. I hold both of them responsible for this catastrophe. But I consider the ends guys more culpable, because they did not do their jobs, whereas this guy (and I've come across a lot like him) is merely rather short-sighted or selfish in imagining that just doing his job and then going for a bike ride or growing oysters was an acceptable way to live your life in a world full of others. Kind of pathetic, but understandable in a culture that places a fleeting premium on certain technical skill when those skills are rare.
The problem as I see it is in this disastrous bifurcation of technical competence and moral responsibility. Shouldn't we - don't we - have both the ability to do certain things, well or not, and the ability to reflect on them?
posted by GeorgeBickham at 2:18 PM on March 31 [+] [!]
Well, the point is that there's nothing wrong with the ultimate end of securitization. It's a perfectly good idea that worked for some time, and he was providing the means for something that was - in his eyes at the time - a good thing.
And the 'ends guys' were correct in assessing his software as being fit for purpose. His software only handled "prime" mortgages (though only a non-compete contract kept him from extending it to handle sub-prime), and for that purpose it was perfectly suited.
The problem with securitising mortgages is that it because it makes the financing process more efficient it can allow bubbles to inflate much more rapidly. Previously, even if there was a sustained run of very low real interest rates, the rise in house prices was slower. Even if you did have a billion dollars you wanted to loan to house buyers RIGHT. NOW. You still couldn't easily do it, you'd need the infrastructure of local banks first. With securitisation, it was super easy for money to flow into mortgages.
But as saulgoodman points out, the real problem was the unregulated derivatives market.
It isn't so much that he didn't take the time to foresee the consequences of his actions, as that they were not reasonably forseeable.
So it's not so much the case that this guy is like a gunsmith. A gunsmith is obviously not responsible for all the uses of his weapons, but he does know that he is making tools that can kill.
The writer is more like a guy who sells a hammer which is used responsibly by most users, but which some irresponsible users use to build shaky houses.
Unknown to the original hammer vendor (now retired to farm oysters) these shaky houses are connected to all the well built houses with guy wires, but no-one has regulated the number of poorly built houses guyed to the well built ones! So now, when all the houses look they might come down together, is it really hammer guy's fault?
posted by atrazine at 1:29 AM on April 1, 2009
It isn't so much that he didn't take the time to foresee the consequences of his actions, as that they were not reasonably forseeable.
So it's not so much the case that this guy is like a gunsmith. A gunsmith is obviously not responsible for all the uses of his weapons, but he does know that he is making tools that can kill.
The writer is more like a guy who sells a hammer which is used responsibly by most users, but which some irresponsible users use to build shaky houses.
Unknown to the original hammer vendor (now retired to farm oysters) these shaky houses are connected to all the well built houses with guy wires, but no-one has regulated the number of poorly built houses guyed to the well built ones! So now, when all the houses look they might come down together, is it really hammer guy's fault?
As I said, I don't really consider him culpable: self-marginalising, more like, in the general sense of how he seemed to have seen his place in the scheme of things, as a kind of honest prole. But I do consider him responsible, in a chain-of-events kind of way. My larger point is about the evaporation of responsibility within a system where roles are sliced up into little boxes. The mortgage brokers, including those selling sub-prime, deferred responsibility to the underwriters, who deferred to those securitizing the bundles of mortgages ... and so on.
Ultimately, few who were in a position to contribute to a big picture of what was going on were doing so or even invited to. This is partly a management failure, but it's really a cultural failure within the financial industry, among other structures, which actively prevent reasonable foresight from even being attempted. It designates certain functions as technical, boxes them up and passes them down the line.
I know a guy, a fairly senior quant at an institution you would have heard of, who was ignored or called out for overstepping his mark, when he warned about systemic risks building up. That was supposed to be a management issue, but management (executives, regulators and legislators) had all left the building. The system is supposed to generate its own rules as emergent properties of the sum of its transactions... and all that (there's where you might find someone to blame, by the way). Maybe what we need is an equivalent to the situation on modern car assembly lines, where the lowliest bolt-fitter can yank the cord and bring the line to a halt if they see a problem? That may imply re-thinking quite a few of our arrangements, admittedly, including those of technicity and responsibility.
posted by GeorgeBickham at 6:36 AM on April 1, 2009
So it's not so much the case that this guy is like a gunsmith. A gunsmith is obviously not responsible for all the uses of his weapons, but he does know that he is making tools that can kill.
The writer is more like a guy who sells a hammer which is used responsibly by most users, but which some irresponsible users use to build shaky houses.
Unknown to the original hammer vendor (now retired to farm oysters) these shaky houses are connected to all the well built houses with guy wires, but no-one has regulated the number of poorly built houses guyed to the well built ones! So now, when all the houses look they might come down together, is it really hammer guy's fault?
As I said, I don't really consider him culpable: self-marginalising, more like, in the general sense of how he seemed to have seen his place in the scheme of things, as a kind of honest prole. But I do consider him responsible, in a chain-of-events kind of way. My larger point is about the evaporation of responsibility within a system where roles are sliced up into little boxes. The mortgage brokers, including those selling sub-prime, deferred responsibility to the underwriters, who deferred to those securitizing the bundles of mortgages ... and so on.
Ultimately, few who were in a position to contribute to a big picture of what was going on were doing so or even invited to. This is partly a management failure, but it's really a cultural failure within the financial industry, among other structures, which actively prevent reasonable foresight from even being attempted. It designates certain functions as technical, boxes them up and passes them down the line.
I know a guy, a fairly senior quant at an institution you would have heard of, who was ignored or called out for overstepping his mark, when he warned about systemic risks building up. That was supposed to be a management issue, but management (executives, regulators and legislators) had all left the building. The system is supposed to generate its own rules as emergent properties of the sum of its transactions... and all that (there's where you might find someone to blame, by the way). Maybe what we need is an equivalent to the situation on modern car assembly lines, where the lowliest bolt-fitter can yank the cord and bring the line to a halt if they see a problem? That may imply re-thinking quite a few of our arrangements, admittedly, including those of technicity and responsibility.
posted by GeorgeBickham at 6:36 AM on April 1, 2009
« Older A Photographic Mystery. | On jetlag, and how to beat it. Newer »
This thread has been archived and is closed to new comments
posted by GuyZero at 2:19 PM on March 30, 2009