Author Topic: California still leads the way (but not how Rafi and several others thought...)  (Read 2979 times)

Fenring

  • Member
  • All Members
    • View Profile
No, Fenring, you're really not getting it.  Take a look at the trend line https://fred.stlouisfed.org/series/DRSFRMACBS.

It only goes back to 1991. How is that helpful, since the previous significant crisis was in the early 80's? The early 90's were more or less a boom period (not quite a wartime boom, but still a prosperity time), so that's a weak basis for making long-term forecasts. In any case, the graph shows homeowner delinquency rate, which is of course correlated to the mortgage market crisis but isn't directly the thing that failed. The problem wasn't just that homeowners began to spontaneously cease being able to pay their mortgages, which led to the crisis, like for instance if they all lost their jobs or something. The speculative nature of the trading is something not measurable in a simple graph like "homeowners going bust"; the growing instability of the traders themselves to accept these mortgages has little to do with the 15 year trend in people making their mortgage payments. Looking at a graph like this (even if it went further back) would be useless in assessing the mounting instability of the actual reality of what was being counted as assets on the books.

So yes, I think a real expert would have been able to assess the soundness (or lack thereof) of how they were doing their calculations. Real experts don't rely just on data models, but learn real facts on the ground. For instance, looking at the long-term trend-line for a corporation you're thinking of investing in is ok, and reading the quarterly statements is good too, but nothing can substitute for physically being on premises, knowing the operators involved, local conditions (if it's in another country), and other such details. No graph can replace real knowledge, and I believe that there were people in NYC with real knowledge prior to 2008. Greenspan can say whatever he wants, this was not some totally unaccountable blip in history, but worked based on the same physics other system work on for anyone who understands them to an extent. We haven't got that great a grip on this 'physics' but it's not just some unknowable mystery leaving us with only using trend-lines to think forwards.

Quote
What good is a model that takes a rate that has been say stable at 1% but never less than 0.9% and never more than 1.1% and "predicts" that it will go to 4%?  Sure it could give you an idea of what could happen, but to bet on it would be folly.

You're right, it's not good. That's why using convenient models to justify dangerous behavior is bad science, even though it looks good in the board room. Do you seriously think executives green lighting these transactions were thinking of their little deals having a broader effect? Their business was to get their deals done that were right in front of them, not to worry about what happens when everyone else is doing the same.

Quote
It's like betting your life savings on a 16 seed to win the NCAA basketball tournament.

If you're in the process of poisoning the other team's drinking water then maybe it's not such a bad idea. It just helps to know what is or isn't poison.

Quote
I'm beginning to think that basic concepts of finance are not as basic as I believed.

Agreed. Actually, I don't think there are any basic concepts of finance. Most 'beginner' or introductory texts on economics strike me as being either so reductionist that they say nothing, or else so theoretical that they have nothing to do with reality. The fact is that complex systems analysis is the only basis of examining huge systems of numbers moving around, and anything less convoluted doesn't describe the reality. There's no equivalent here to Newtonian physics where the dumbed down version is pretty much accurate for large-scale events. In economics it really isn't.

Quote
At their fundamental core, CDOs are tools to reduce risk exposure.  The idea that they were created by "wild dogs" is so beyond wrong I don't know how to even process it.

The wild dog analogy was in response to your agreement that banking is essentially based on greed first, which I translated as implying predatory behavior. Mixing metaphors by implying that CDO's, specifically, are the work of 'wild dogs' will tend to break down the metaphor, yes.

Quote
Well you are right that banks are de facto public entities in our system.  You are beyond wrong to think they are trusted and not heavily regulated.  I'm not aware of any other industry that provides office space and internal email accounts to regulators that are permanently on site.  Over 10% of the employees of a bank are in compliance and regulatory functions.

It's the system that is bad, the rules are beyond arcane and complex.  That's on the government, who made the rules.

I think you missed my point here. It seems to be your contention that the government and its regulators are what cause these problems in the private sector, but when I say they need to be watched like a hawk you seem to double back and say that of course we keep an eye on them, through these regulators! If your point is that the regulation could be better, then...well, duh. But if it's that direct regulation is always going to lead to these kinds of problems then I wonder how you think the dogs will be kept on a leash since they control so much of the public well-being in their hands? Private/public ceases to mean anything at a certain point if the people in question can directly make or break the economy. And actually this is what the Fed does on occasion when it's threatened, which is to use the WMD of tanking the economy as a retaliatory measure against being prodded.

Quote
I think you're confused about what fractional reserve means. It's not possible if you are dealing in an actual commodity.  How exactly would one loan out more gold than one has?  In order for it to be possible, you have to be issuing notes (in that case exchangeable for gold).  A bank with 100 gold bars, issues notes for 200 gold bars, knowing that since the notes are tradeable, its not likely to get a run on the bank for more gold bars than it currently has.

You're right that a side currency is usually in place to do this, but it doesn't have to be a currency controlled by the moneylender. Usually there has been more than one currency in place at a given time historically. Sometimes it's a fiat currency, sometimes bimetalism, whatever. In England they used to use sticks. An example of an alternate currency is a bank credit note, which of course is issued by the bank. But let's leave off alternate currency for a second and do a small thought experiment. If ten people each deposit $100 into a bank that previously had no liquid float, they now have $1,000. If they loan out $900 to someone else (10% reserve rate), their books are 'balanced': $1,000 assets, $1,000 debts. And now that same $900 is deposited back into the bank by nine other people(the person who borrowed it used it to pay his workers, let's say), and it's again loaned out to someone, this time to the sum of $810 (10% reserve rate again). The books are now balanced again ($1,900 of assets and debts), except there's a problem: on a potential $1,900 of debts the bank only has $190 on-hand, and additionally in this model there is actually only $1,000 of actual physical money in the system at all, of which currently $190 is in the bank and $810 in private hands. So we can see that it doesn't actually take the issuing of a secondary currency to create an effective increase in the money supply through a fractional reserve, although as I mentioned that is usually done too and if bank notes are involved it can increase at an even greater rate.

Quote
Isn't this basically the plot of "It's a Wonderful Life"?

I forget :(

Quote
Except, when your defaults are guaranteed by the government (Fannie Mae) and the government will loan you whatever you need (the Fed), there is no possibility of a "run on the bank".  This is related to point that Pyrtolin and I always argued about that with fiat money its not possible to have a default unless the government lets it happen.

It's not that a run on the bank can't happen, but rather because everyone trusts the government backing of the banks the psychological safety net prevents people actually believing that they must ever physically run to the bank to get their money out. Back in the day when there was no such assurance they'd assume their money was toast if they didn't get it back in their hands, whereas now people simply don't believe that will happen. It's that belief which the current system has achieved, but if literally everyone who had money in a bank account simultaneously went to withdraw it, of course the entire economic system would collapse immediately since it's a debt-based system. The money supply would contract to an infinitesimal amount compared to what it is now, and there wouldn't be enough money in the government's wildest dreams to back it up. Not even the Fed could print that much. As it is the FDIC only insures up to $100,000, but even so it could never pay back even that much to everyone if it had to. The actual FDIC reserve is tiny and IMO mostly serves to make people feel good. The real source would be the Fed, but at that point you'd be printing up the entire economic system from nothing, rather than pumping it in piecemeal as a stimulus. So yeah, there can be a run, but it would take a catastrophic fear from the public to generate it.

Quote
Quote
Quote
One of the benefits of this way was the use of private greed and profit motivation to ensure that the money was going to good risks with strong ability to repay loans.  That worked for a really long time.

You mean for like 15 years? ;)

No, more like several hundred.

How do you figure? I meant specifically private greed backed up by the government, btw. Because prior to 1913 there were  bank crises all the time, with banks closing and people losing everything. The system was total chaos. We don't even remember them any more but there was major crises practically every decade back then (maybe more, I haven't looked at the dates in a while).
« Last Edit: February 13, 2018, 04:18:12 PM by Fenring »

Seriati

  • Member
  • All Members
    • View Profile
Some pretty positive news for CA, I really found interesting that 20% of US growth came from the CA economy.  https://www.nytimes.com/2018/02/13/business/economy/california-recession.html?hp&action=click&pgtype=Homepage&clickSource=story-heading&module=first-column-region&region=top-news&WT.nav=top-news

Of course there was also this tidbit:

Quote
For California and the nation, there is a long list of things that could go wrong. A surging budget deficit could stoke higher interest rates. And if the recent upheaval in stocks signals a longer-term decline, it would hurt California in particular because its budget relies heavily on high earners whose incomes rise and fall with the market. President Trump’s moves to upend longstanding trade arrangements could be a setback for the state, home of the country’s biggest port complex. And because the growth of the technology industry has played a huge role in California’s recent boom, a drop in company valuations or in venture capital investments would reverberate swiftly through the state’s economy and tax receipts.

It's interesting that CA is so exposed to the stock market, which Greg hadn't mentioned and really has nothing to do with whether local policies have merit.

Also of interest, was something I had underappreciated, CA really benefits from controlling a large part of the country's ports and port fees.  That's another of those things that is undersold when you think of CA contributions to the country versus the country's contribution to CA (effectively CA grabs a rent on every made in China product sold in the US).


Seriati

  • Member
  • All Members
    • View Profile
No, Fenring, you're really not getting it.  Take a look at the trend line https://fred.stlouisfed.org/series/DRSFRMACBS.

It only goes back to 1991. How is that helpful, since the previous significant crisis was in the early 80's? The early 90's were more or less a boom period (not quite a wartime boom, but still a prosperity time), so that's a weak basis for making long-term forecasts.

Which is why I said in my next paragraph that I couldn't find the longer chart, but that if you looked at it was long and stable.  Don't argue from ignorance. 

Quote
In any case, the graph shows homeowner delinquency rate, which is of course correlated to the mortgage market crisis but isn't directly the thing that failed.

Sigh.  Really?  You think the primary thing that actually caused the crisis "isn't directly the thing that failed"?

Quote
The problem wasn't just that homeowners began to spontaneously cease being able to pay their mortgages, which led to the crisis, like for instance if they all lost their jobs or something. The speculative nature of the trading is something not measurable in a simple graph like "homeowners going bust"; the growing instability of the traders themselves to accept these mortgages has little to do with the 15 year trend in people making their mortgage payments. Looking at a graph like this (even if it went further back) would be useless in assessing the mounting instability of the actual reality of what was being counted as assets on the books.

What that graph showed is a fundamental change in a metric that was fundamental to any plausible calculation of risk - you literally could not have validly calculated risk without that metric.  Given the leverage that was in play, and the order of magnitude of that change in that chart, makes it completely obvious what happened.

Seriously you can literally read all about it.

Quote
So yes, I think a real expert would have been able to assess the soundness (or lack thereof) of how they were doing their calculations.

I see.  So there were no "real experts" in the entire credit industry, the ratings industry, with the regulators, with the investor community (which included almost every single sophisticated investor on the planet).  Yep, you're probably right, a "real expert" would have seen the "obvious."

Quote
Real experts don't rely just on data models, but learn real facts on the ground. For instance, looking at the long-term trend-line for a corporation you're thinking of investing in is ok, and reading the quarterly statements is good too, but nothing can substitute for physically being on premises, knowing the operators involved, local conditions (if it's in another country), and other such details.

Except the "facts on the ground" were exactly what caused them to predict the stability of the mortgage default rate.  Literally, decades upon decades of data from thousands of banks across the country.

But I'm sure you're right, that you could have walked into someone's - what?  house - and seen the problem, if only someone had done their "job."

Quote
No graph can replace real knowledge, and I believe that there were people in NYC with real knowledge prior to 2008.

I see.  So "no graph" of millions of data points spanning decades can replace "real knowledge".

Honestly, don't ever cite a study to me again.

Quote
Greenspan can say whatever he wants, this was not some totally unaccountable blip in history, but worked based on the same physics other system work on for anyone who understands them to an extent. We haven't got that great a grip on this 'physics' but it's not just some unknowable mystery leaving us with only using trend-lines to think forwards.

Again, it literally was such a blip. 

Quote
You're right, it's not good. That's why using convenient models to justify dangerous behavior is bad science, even though it looks good in the board room.

Yeah, but that didn't occur so moot.

Quote
Do you seriously think executives green lighting these transactions were thinking of their little deals having a broader effect?

Yes, literally they were.  In fact the entire purpose of this industry was to spread credit risk of an industry - collectively - across a bigger and more diverse group of lenders than was other wise possible.

These transactions were literally designed to exploit the seeming national immunity to a mortgage crisis to smooth out the local risks of one.

Quote
Their business was to get their deals done that were right in front of them, not to worry about what happens when everyone else is doing the same.

That's sort of true.

Quote
If you're in the process of poisoning the other team's drinking water then maybe it's not such a bad idea. It just helps to know what is or isn't poison.

I agree.  Congress poisoned the water.

Quote
I think you missed my point here. It seems to be your contention that the government and its regulators are what cause these problems in the private sector, but when I say they need to be watched like a hawk you seem to double back and say that of course we keep an eye on them, through these regulators!

No my complaint is with people like Greg that constantly want government to solve the problems it created.  This crisis was directly the result of deliberate government policy.  The fact that industry responds to incentives is just a fact.  Do you recall my example of an industry forming to carry ocean water from the beach to the end of the pier. 

These CDOs don't ever get this crazy, maybe they never form at all, without government pressure to force lending into areas where the borrowers couldn't repay it.

Quote
If your point is that the regulation could be better, then...well, duh. But if it's that direct regulation is always going to lead to these kinds of problems then I wonder how you think the dogs will be kept on a leash since they control so much of the public well-being in their hands?

Personally, I take the government funding out of their hands.  I can not comprehend - at all - why our government lends money to banks to exploit us.

Quote
You're right that a side currency is usually in place to do this, but it doesn't have to be a currency controlled by the moneylender.

It actually does.  The lender has to have the power to create/issue more of the "money" than they have assets.  Whether it's because they can create bank notes, or because the government gives them massive loans for nearly free is immaterial when the government policies are mandatory.

Quote
If ten people each deposit $100 into a bank that previously had no liquid float, they now have $1,000. If they loan out $900 to someone else (10% reserve rate), their books are 'balanced': $1,000 assets, $1,000 debts.

Lol that's not how it works.  In your example, the bank would have a $1,000 in deposits, and be able to loan out $10,000 for a leverage ratio of 10:1.   Like I said though, it's not that straight forward, and the actual leverage ratios were several hundred to 1.

If banks could only lend out the amount of their deposits we'd never have had this crisis, and of course our economy would be less than a hundredth of its current size.

Quote
It's not that a run on the bank can't happen, but rather because everyone trusts the government backing of the banks the psychological safety net prevents people actually believing that they must ever physically run to the bank to get their money out.

Actually, you're wrong.  It literally  can't happen in a fiat currency system unless the government lets it.  The government can always print more bills to cover the shortfall. 

A classic run on the bank only works against a system that has a specific backing (like say a gold standard), or one where the government lets a bank collapse.

Did you ever hear of SDIC insurance?  99% of the population is literally immune from a "run on the bank" problem - as guaranteed by the federal government.

Quote
Quote
One of the benefits of this way was the use of private greed and profit motivation to ensure that the money was going to good risks with strong ability to repay loans.  That worked for a really long time.

***

How do you figure? I meant specifically private greed backed up by the government, btw. Because prior to 1913 there were  bank crises all the time, with banks closing and people losing everything. The system was total chaos. We don't even remember them any more but there was major crises practically every decade back then (maybe more, I haven't looked at the dates in a while).

Sure, individual banks.  And because of that problem individual bankers were risk adverse.  They'd do "silly" things like insist on proof of ability to repay, deny loans to people they thought were bad risks, and insist on 20% down on a mortgage.

None of which served Congressional desires to have easy credit flood the market and put people in homes.  So  they meddled.  They insured the bank accounts.  They removed the regulations separating wall street banks from main street banks, and allowed both to operate at higher leverage ratios.  They specifically forced banks to lend into neighborhoods that had bad credit.  Politicians actually made claims that it was discriminatory to ask for proof of ability to repay.  Then insured mortgages, and even "helpfully" reduced the amount down on insured mortgages to less than 3%.  They changed tax policy to make mortgage interest deductible (even though rent is not and it's very likely that renters need more help).  Then aggressively investigated banks who didn't make politically desirable loans, and complained loudly whenever credit markets "contracted" (and not about the risk they were avoiding, but rather the "greed" that was causing bankers not to make loans).

TheDrake

  • All Members
    • View Profile
Quote
Politicians actually made claims that it was discriminatory to ask for proof of ability to repay.

That sounds pretty much wrong to me. I think what they were saying is that you have to ask everybody for the same proof, if you're going to ask. NINJA loans weren't a government creation. They were about subprime lenders filling an unquenchable thirst for more MBS. Banks took these risks thinking "oh well, we can always foreclose" - except that because of grossly inflated real estate market, the burst bubble dropped all those loans underwater.

Greg Davidson

  • Member
  • All Members
    • View Profile
Quote
You think the primary thing that actually caused the crisis "isn't directly the thing that failed"?
Okay, maybe we are closer to understanding the difference of opinion. I don't disagree that government policies to encourage home ownership contributed to the default rate. However, without the actions of Investment Banks and the ratings agencies, we would have had a problem scaled at $7T based on the value of housing and not a crisis concerning $54T (see Greenspan testimony 10/23/2008). So there is some degree to which "government" policies contributed to the default rate (along with private sector mortgage lenders and actual borrowers who also count as private sector actors and not the government). But you are not losing 800,000 jobs per month, with the collapse of the auto industry imminent, all just from the collapse in housing prices. The much larger sums of money beyond real estate played a much more prominent role in causing damage from the collapse.

But let's also be crystal clear that "government" is an overly broad term, because we have two political parties in the US that implement government policy in very different ways. And anti-government rhetoric is a common weapon in the Republican quiver, but we need to hold the parties accountable for when they specifically take the lead for actions which turn out disastrous.

Quote
From his earliest days in office, Bush paired his belief that Americans do best when they own their own homes with his conviction that markets do best when left alone. Bush pushed hard to expand home ownership, especially among minority groups, an initiative that dovetailed with both his ambition to expand Republican appeal and the business interests of some of his biggest donors. But his housing policies and hands-off approach to regulation encouraged lax lending standards.

Bush did foresee the danger posed by Fannie Mae and Freddie Mac, the government-sponsored mortgage finance giants. The president spent years pushing a recalcitrant Congress to toughen regulation of the companies, but was unwilling to compromise when his former Treasury secretary wanted to cut a deal. And the regulator Bush chose to oversee them - an old school buddy - pronounced the companies sound even as they headed toward insolvency.

As early as 2006, top advisers to Bush dismissed warnings from people inside and outside the White House that housing prices were inflated and that a foreclosure crisis was looming. And when the economy deteriorated, Bush and his team misdiagnosed the reasons and scope of the downturn. As recently as February, for example, Bush was still calling it a "rough patch."

The result was a series of piecemeal policy prescriptions that lagged behind the escalating crisis.

"There is no question we did not recognize the severity of the problems," said Al Hubbard, Bush's former chief economic adviser, who left the White House in December 2007. "Had we, we would have attacked them."

Looking back, Keith Hennessey, Bush's current chief economic adviser, said he and his colleagues had done the best they could "with the information we had at the time." But Hennessey did say he regretted that the administration had not paid more heed to the dangers of easy lending practices.

And both Paulson and his predecessor, John Snow, say the housing push went too far.

"The Bush administration took a lot of pride that home ownership had reached historic highs," Snow said during an interview. "But what we forgot in the process was that it has to be done in the context of people being able to afford their house. We now realize there was a high cost."

For much of the Bush presidency, the White House was preoccupied by terrorism and war; on the economic front, its pressing concerns were cutting taxes and privatizing Social Security, a government retirement and disability benefits program. The housing market was a bright spot: Ever-rising home values kept the economy humming, as owners drew down on their equity to buy consumer goods and pack their children off to college.

Lawrence Lindsay, Bush's first chief economic adviser, said there was little impetus to raise alarms about the proliferation of easy credit that was helping Bush meet housing goals.

"No one wanted to stop that bubble," Lindsay said. "It would have conflicted with the president's own policies."
  http://www.nytimes.com/2008/12/21/business/worldbusiness/21iht-admin.4.18853088.html

TheDeamon

  • All Members
    • View Profile
Quote
As long as we're going to be content to categorize them as wild dogs looking for their meat then you shouldn't have any qualms about treating them like a kennel and putting a leash on them.

I'm beginning to think that basic concepts of finance are not as basic as I believed.

At their fundamental core, CDOs are tools to reduce risk exposure.  The idea that they were created by "wild dogs" is so beyond wrong I don't know how to even process it.

I understood it. Maybe it would help Greg if you compared it to insurance "risk pools" as his rooting around in health care issues should have introduced him to it.

It is only a small step away from the origin of insurance back in the Age of Sail where any particular voyage (particularly long ones)  could be a ship's last even though ship's seeing decades of service wasn't uncommon, even then. But as the cost of trans-oceanic expeditions were enough to ruin investors should "their ship" get lost, they "spread the risk around" (via Lloyds of London) and made the British Empire a reality in the process.

yossarian22c

  • Member
  • All Members
    • View Profile
I’m pretty sure Greg gets it. The problem wasn’t only that the cdos that helped hide the risk due to negligent ratings agencies. The result of a housing decline with just those components would have been bad but probably wouldn’t have sunk any major firm or bank. It was all of the derivative bets on the cdos that made a 1 dollar decline in the cdo a 5-10 dollar loss for the firm. The derivative bets coupled with high leverage levels turned a economic downturn into an economic crisis. You can reasonably argue what role government policy had in the 1 dollar decline. But the additional 4-9 dollar loss plus high leverage is almost exclusively private sector.

Seriati

  • Member
  • All Members
    • View Profile
Greg, you'll love this article.  It's a pretty resounding condemnation of the finance industry.  https://www.theatlantic.com/business/archive/2016/09/fairfield-county/501215/

You'd probably be surprised to know that I agree with a good chunk of it, though I think it misses the boat on the resolutions (higher taxes on the rich is political claptrap that doesn't address any of the underlying concerns and ignores that they will literally move out of state).  It glosses over a lot of facts (like that the CT rich pay an enormous amount of taxes in CT, literally dwarfing the taxes of everyone else).  It also misses that the best cure would be to incentivize better investment practices, practices that generate better jobs, rather than taking and redistribution.  Put the smart people to work on the problems you want them to be solving.  But its a good critique about why market success is no longer translating as directly as it once did into more industry and better jobs.

I actually, came at this because of a continuing fascination I have with the "1%" concept.  CT as a whole already has a 1% line that's over $600k, and given the concentration of wealth in Fairfield county, I bet the line there is closer to $1m.

Edited to add: Just found this directly on the 1%er question.  Fairfield county's line is about $1.4m.  https://www.nytimes.com/2016/09/25/business/your-local-1-percenters-may-not-be-as-rich-as-you-think.html
« Last Edit: February 16, 2018, 09:31:22 AM by Seriati »

Greg Davidson

  • Member
  • All Members
    • View Profile
Yossarian, some of the cdos were leveraged between 10 and 20 times.

Seriati/TheDaemon,
I fully understand risk pools, and the whole notion that in an orderly market with many independent actors, even if some behave irrationally there should be others who recognize the irrationality and self-optimize by betting the other way. Michael Lewis's The Big Short was about a few individuals who just did that, but the problem was that so many many others all bet the wrong way. This was specifically what Alan Greenspan confessed as a "fundamental flaw" in free market economic theory - in a functioning market, enough people are supposed to recognize irrationality and in their own self-interest pursue an arbitrage that richly rewards them by providing a market correction. It is a beautiful and elegant system, and it often works - but liberal economics (Post Keynsian and others) recognizes that it does not always work, and that vulnerability to catastrophic failure legitimizes some level of government regulation

Seriati

  • Member
  • All Members
    • View Profile
Yossarian, some of the cdos were leveraged between 10 and 20 times.

Actually, they were all leveraged more than 20 times (that's before you add on synthetics) and before you consider that the banks and mortgage cos were themselves leveraged (with the banks being several hundred times) and that the investors often also used leverage to increase their stake.

The CDOs were so inherently leveraged that the lowest tranches of debt were expressly considered equity rather than debt because of the risk.

Quote
Seriati/TheDaemon,
I fully understand risk pools, and the whole notion that in an orderly market with many independent actors, even if some behave irrationally there should be others who recognize the irrationality and self-optimize by betting the other way.

Greg, you've already lost me here.  You can't claim to understand these risk pools, then assert that the actors were behaving irrationally.  They literally were not being irrational.  Betting against them would have been irrational.

In fact, to put into terms anyone here can understand.  The concept on a CDO getting a triple A rating is like loaning someone $800 on a collateral of a $1000 gold bar.  There's no "irrationality" on that. 

To see the whole picture, you'd also see someone else loaning $150 at B, someone loaning 100 at junk and someone putting in $25 in equity.  That's $1075 on a $1000 gold bar - what gives?  You don't sell the bar for 10 years, it's a bet that the increase in the gold bar is greater than the cumulative interest.  Given the AAA stuff has a very low interest rate because it's over collateralized, you can easily get there.  In fact, the certainty on getting there was 95% plus, and the risk of default on the triple A?  less than 0.4% (and there were specific measures to require selling the collateral to de-risk if it even looked likely - unfortunately this doesn't work when there is an industry wide event and everyone is selling).

When you look at the interest rates that actually apply in a CDO (ie the mortgage interest rates, versus the note rates) mathmatically even with a default rate that's off 2 standard deviations the bond holders - all - get paid, it takes an even bigger swing to put the AAA at risk.  Unfortunately that's what we got.