econ-251: Financial Theory
Lecture 25 - The Leverage Cycle and the Subprime Mortgage Crisis [December 8, 2009]
Chapter 1. Assumptions on Loans in the Subprime Mortgage Market [00:00:00]
Professor John Geanakoplos: So I'm going to talk about what led up to the crisis and how the crisis makes us realize that the things about the theory that we've been learning all semester that ought to be slightly different, and I've been working on a new version of the theory for the last ten years, what I call the Leverage Cycle, which I have to say didn't get very much attention ten years ago, but which now seems to be, I don't know, it seems to be getting more attention.
It's almost like life imitated art, really. It couldn't have turned out more like the theory than I would have expected. So I'm going to tell you the facts and then try to fit a theory to it, and probably will spill over until Thursday, and I think Thursday I won't spend that much time finishing. I'll have a review class if anyone's interested for the whole material.
And I think when you see the critique of the theory it might help put the theory in a little bit more perspective. But anyway, nothing of what I say these last two days you should feel is going to be on the exam. It's just more information that you might find interesting.
So we talked last time, remember, at one of those special classes about the sub-prime mortgage market, and the idea was that you would pool together a bunch of loans. These are the--I can do this, I guess. You'd pool together a bunch of individual mortgages like this, and then you'd put them in a giant pool, and then cut the pool up into a triple-A bond, a double-A bond, a single-A bond, a triple-B bond, and this is the over collateralization and the residual.
And I'm not going to go through the whole structure, but suffice it to say that the idea of this is you take individual loans which are very risky, after all these are sub-prime borrowers so everyone knew that they were risky, and you create what's supposed to be an incredibly safe bond. The definition of triple-A bond is that it has a 1 in 100 chance of going bankrupt in 10 years, of losing any principal in 10 years.
That was the definition they claimed. I heard a talk about power failures last night, by the way, the power grid which went out all across from Cleveland to Manhattan in 2003, and they said that was an event that should happen once every 10 years. And then when you hear what happened in the event, someone didn't trim a tree, and the line sagged into the tree, and so then it shut itself off, and then the city next door they had the same problem with someone not trimming a tree, and that line shut off, and once you have two or three lines shut off the system automatically goes down. It seems like it could happen much more often than once every 10 years.
But anyway, it's the same thing here. This is supposed to be once every 10 years, 1 in 100 chance, and it happened in a giant way. So the idea of the triple-A, so they're supposed to very safe because if the first house defaults, the first homeowner defaults, instead of paying back the 100 he owes, let's say he just stops paying, and then you throw the guy out of the house, but then you get to sell the house.
The house is there as collateral for the loan. Now, the loan should have been 100. The house should have been worth much more than the 100. That's why it should have been collateral. So you should have had 120 dollars of the house to sell so even if the house went down by 33 percent to 80 that's still only a 20 dollar loss, or let's say 20 cent loss on the dollar loan, and that 20 cents comes out of the triple-B piece. So you notice how many losses do you have to have if you're losing 20 cents on a house?
To have 5 dollars worth of losses you've got to have 25 of the 100 houses go down, and another 25 of the houses have to go down to lose this one, and then another whole bunch of them, 35 more to go down. So you've got to have like 85 houses going down before you touch the triple-A, because the losses always come from the bottom and work their way up.
So when they made these deals they thought it was inconceivable that you'd have 85 percent of the houses defaulting, or that you'd lose so much more than 20 percent on each house because things like that just hadn't happened before. And so it seemed like they were incredibly safe. So that was the structure.
And the idea, again, was to take risky securities and by securitizing them create a whole bunch of apparently perfectly safe securities. Because, as we said, the goal of most investors is to invest in something that's completely safe, they don't have to worry about getting more in one state or more in another state. And this seemed like an ingenious way of turning 1 trillion dollars of very risky sub-prime loans into, this is 81 percent, into 800 billion of very safe bonds and maybe riskier bonds down below that. So you could never find 1 trillion dollars worth of people willing to lend to sub-prime borrowers, but you didn't have to after the securitization, because 800 billion of the loans to them were by people who thought they were perfectly safe.
You only had the guys buying these bonds down here who thought that they were running a very big risk. So the whole idea of securitization, the whole idea of the American financial system was to--well, let's see what the whole idea was, was to create as many safe loans as possible. So you could put it a little bit more generally. I should have had this picture before.
You could say if the states of the world--if you're here today and you might have many of these different states of the world tomorrow, some people know that they're going to be rich in some states and poor in other states. What they'd like to do is get the same amount in every state. They can do that by directly buying safe bonds. That's what everybody's trying to do. They're trying to hedge themselves to get completely safe bonds. Most people don't know how to hedge so the financial system creates safe bonds for them.
Other people, they do know how to hedge, but their cash flows are different in the different states because, for example, they run a business and if interest rates go up that might be bad for their business. If interest rates go down it might be good for their business, so they've got different amounts of money in the different states. And then to create hedging instruments for these people you'd like to create securities that pay a lot in states where they want the money so that way they can buy hedging instruments.
So the purpose of the financial system is to create cash flows in states that people want them. For the majority of people that's the same thing in every state. For other people it's something like the Arrow securities. So they can buy money in the states when they're running short.
And why is this connected to mortgages, because how do you know that anyone's going to keep their promise to deliver money in those different states, or money in every state, how do you know the people making the promise will keep their promise? Well, you don't. That's why you need collateral, and the best collateral in the country is the houses.
So the purpose of the financial system, as we said all through the course, is to create securities that pay off in different states of nature, of the type people want, and that are guaranteed, so that people keep their promises, guaranteed by collateral. Houses as the best collateral made it inevitable that the most sophisticated part of the market was going to turn into the mortgage market.
So at first glance you would think the most complicated part of the market is the stock market, but no. It turns out that collateral is the key, because without guaranteeing a promise the promise isn't really very good, and so houses are the best collateral. So we'll see now that the change in the theory that I want to describe is what happens when you suppose people might break their promises and you have to take into account collateral.
All right, but let's get back to the facts. So there were 1 trillion dollars of these sub-prime mortgages around. So then what happened? I want to skip through this pretty quickly. People thought that it might be dangerous and they wrote insurance on the sub-prime mortgages.
So an insurance on a sub-prime mortgage, CDS, Credit Default Swap, would say if 1 dollar of the principal of the triple-B bond disappears, and we saw how that could happen, you take one homeowner who loses 20 cents, I mean, that will come out of this first, but after all this stuff is wiped out, 20 cents of the losses because the homeowner instead of paying his dollar doesn't pay it all. The house is sold. You only get 80 cents back, that 20 cents comes out of the triple-B, so the insurance would then say--the buyer of insurance would get to collect the 20 cents. So this seemed like another way for people even in the BBB range to guarantee their payments.
Well, it turned out that the writers of insurance didn't just write 20 cents in case there was a 20 cent loss, they might write 10 dollars of insurance on the 20 cent loss. So these Credit Default Swaps are getting very close to Arrow securities. So the theoretically inclined economists who didn't think very far ahead in the Clinton Administration, namely Larry Summers and people like that, Bob Rubin, they were ecstatic about these Credit Default Swaps because to them it reminded them of Arrow securities.
And they thought, my gosh, the market is coming close to creating Arrow securities and isn't it great that people will now be able to hedge much better because they've got these Arrow securities that they can trade. Now, problem was that who's writing the insurance? The people writing the insurance are big banks and people like AIG. So they're saying if you lose 20 cents we'll pay the insurance. Now, of course, that's a promise just like the promise to repay is and that should have been collateralized.
The trouble is that when AIG as a triple-A rated company wrote its promises many of the people who bought the insurance just figured that's such a great company it's obviously going to have the money, probably isn't going to happen anyway, so I'm not going to worry about the collateral, and they managed to make all of these promises without putting up collateral to guarantee that they are going to keep their promises.
This mistake that people make, "Well, it's probably not going to happen anyway that we're going to even collect the insurance," you shouldn't have bought the insurance if you didn't think you were going to collect it. So obviously, even though it's a low probability event, you have to think what will happen when it comes time to collect the insurance. Are they going to have the money?
Now, many banks in Europe bought these things and they are accused of having realized that they wouldn't be paid off. They only bought it so they could convince their regulators that they actually had a very sound bank and therefore didn't have to set aside capital, and they knew very well that they probably wouldn't be paid off. They just didn't think that it mattered because things wouldn't go that bad. So in any case, to circle back, we've got these sub-prime mortgages where you've written a bunch of apparently safe bonds.
A lot of people, but not many, 20 percent holding dangerous bonds, realizing that they're dangerous, and you've got a bunch of insurance written on the dangerous bonds, but of a much bigger, five times as big as the amount of dangerous bonds. If something goes wrong with that pool these bonds are going to lose and then the writers of insurance are also going to lose a huge amount of money.
So what happened? Well, on top of that there was a CDO market where you took the triple-B pieces and broke those up like you did before into bonds, more triple-A's and so on. So if something happened to this original pool and a lot of homes started defaulting that would wipe out all the triple-B's forcing all these people to pay insurance.
And on top of that the CDO market which is just the same thing done again, but using the triple-B's as collateral, all these things would get wiped out including the triple-A's here. And then that wasn't enough. Wall Street then wrote CDO-squareds on the triple-B and A pieces of the CDOs. All right, so we talked all about this last time. So now I want to get into what happened. Any questions? Yes?
Student: So when these were being securitized like the triple-A is there any way to distinguish on paper the normal triple-A's and the triple-A's derived from like triple-B's, for example, or are all triple-A's just the same and there is no way to distinguish them?
Professor John Geanakoplos: No, a buyer knows very well where his triple-A is coming from. A buyer has access to all--so if you were a buyer--the hedge fund I work with, Ellington, was a big buyer of these kinds of things, and also you'll see what our strategy was, but we knew very well that--so let's go back to this.
We were a big buyer of things here, like these triple-B's we bought a lot of. So we could get for every single home in this pool we knew the history, we know the loan to value, the income of the borrower, what zip code the borrower lives in. We know a tremendous amount of information about each one of these individual houses, how they've paid before, what kind of credit rating they have.
All that information we have, and so we build the model of how reliable we think these borrowers are and what they're going to do. And so on the basis of that information we can predict what we think the value of the BBB is going to be, and whether it makes sense to buy insurance on it or not.
So what was our strategy, by the way, in our hedge fund? Our strategy was, take a pool where we think that the pool is really a good pool, buy the riskiest piece which is down here, the residual that I described last time, for which there is no insurance.
Now, it may be that the whole world goes to hell and we lose money on our residual, but so how do we hedge that? What we do is then we take what we think is a bad pool of loans, they're from a county that we think people aren't as reliable, say, and we don't buy the residual. In fact, we do the opposite. We buy the insurance on the triple-B. There is no insurance on the residual.
So for the bad pool we've bought the insurance. For the good pool we've bought the residual, and so if we're right the residual is going to pay a lot of money because these people aren't going to default. And if the bad pool is as bad as we think they're going to default and we're going to collect the insurance. So we're going to win on both, and if the whole market collapses, well, our residual will go to zero, but so will the BBB in the bad pool and we'll collect insurance and so we'll be protected. So that was our strategy. So, yes, the answer is you have a lot of information about what you're buying.
Student: It's not actually a triple-A, it's a triple-A of a bad pool.
Professor John Geanakoplos: Yeah, this appears the triple-A of a bad pool, but the people buying the triple-A know exactly what's in the pool.
Now, they may be dumb and not bother to think about it and just trust the triple-A rating without thinking about it, but probably a lot of them did think about it. And if we take this pool, the CDO, these triple-A's were held mostly by big banks, and you would have thought they were the most sophisticated of all the buyers because they're the ones creating the pools. So they certainly know what's in the pools. They're the ones who created this original pool here.
They're the ones who underwrote those loans and so they're the ones, they know everything about it, and they're the ones holding all these triple-A's. So we're not talking about people who have no idea what they're doing. We're talking about people who supposedly do know what they're doing.
All right, so that was the basis of the market. Now, what happened? So I can describe our hedging strategy and our models, but I'm not going to do that. So this was our strategy. You go long the residuals short the bad pool. Now, what happened?
Here's what happened. There was an index that was created in the beginning of 2006. You just put together a huge number of triple-B bonds and triple-A bonds of different pools and average their prices. That's the index, and so they're all 100, you see? And then in January of 2007 all of a sudden the market starts to go crazy, and by March and April it's collapsed. These bonds, a lot of them, the triple-B's, have collapsed from 100, where everything was, all the way down to 60, just like that. In a few months it went from 100 to 60, 100 here, three or four months later it's 60. This is March of 2007.
Now, what was the stock market doing then? You might remember the stock market, S&P, so here's the stock market, too far in historical time, but anyway, here's the stock market. It had this NASDAQ collapse, but this point right here is October 31st, 2007. So ten months or nine months later the stock market reaches its peak. So all this time from here up the stock market is climbing and climbing and climbing. So the sub-prime mortgage market has collapsed, but clearly the rest of the population doesn't think that things are that bad.
So just to put everything in perspective, where else are we? What other big--housing, let's look at housing prices here. We're going to come back to this graph if you can see it. So this is 2000 here. This is 2009 here. Green is the housing. This is Shiller's housing index. So if you set it at 100 in 2000, up here at the very top, that's the middle of 2006, it hit its peak and then started to come. So housing hit its peak in the middle of 2006, the third quarter of 2006, and then sort of didn't come down very fast, but started to come down.
Sub-prime mortgages collapsed in January of 2007 and the stock market started its very quick collapse in October-November of 2007. So it was first housing prices turned, then sub-prime mortgages collapsed, then the stock market ten months later collapsed, so we have to explain all of these things and what was going on.
So in the--sorry, where was I?
Chapter 2. Market Weaknesses Revealed in the 2007-2009 Financial Crisis [00:18:27]
So this is the sub-prime market collapsing. Why did this market collapse? What happened to make them collapse?
This is actually an example of the wonderful nature of the American financial system. The fact the market collapsed is a great thing because why did it collapse, what was the information that made people think that things were going to go down? Well, it was this information, cumulative loss by reporting month. Let's look at some.
These are Countrywide, they're one of the bad guys, that is they're one of the ones whose loans went bad and the company went out of business, basically, and got bought by Bank of America. So look at delinquencies historically. So for 2003, these are Countrywide loans given in 2003. It's the brown thing. So here are their delinquencies, right? They start at 0, naturally and then they go up to 2 percent and they stay there as a function of the pool.
Some of those guys are defaulting and being thrown out of the pool. That's how delinquencies can go down. Also people can start paying after being delinquent. Delinquent just means you're missing some payments. Look at 2004. They're still hardly doing anything. And then look at 2005. Well, all of a sudden delinquencies start going up. In 2005--they're 5 percent by the end of 2007, but delinquencies, 5 percent delinquencies.
We're just talking about people who missed a few payments, 5 percent delinquencies. They haven't actually been thrown out of their house. It takes 18 months. Nobody's lost anything. They've just been delinquent for a little while. So what was the catastrophe that made the market collapse?
Well, this is it. So WALA means Weighted Average Age. So look at the months. So you take things that were issued in--this is the securitization. They're bundled together in the beginning of 2006. This is the second half of 2006, but the loans actually started in 2005 they just didn't get--the homeowners got their mortgages in 2005. They got securitized in 2006.
So you look at the delinquencies and you see that nothing happens the first 12 months, these are losses, and then eventually it goes it 1 percent. The losses go to 1 percent. We're not talking about giant numbers. Well, everybody saw that. Then they looked at the losses for things issued second half of 2006, and the losses, that's this blue line, they are going to, we're talking about 20 months later, they're going to 1 percent or 1.2 percent.
You look at the things issued in 2007, which means that the loans you're really measuring from somewhere in the middle of 2006. At 1 year you're at half a percent. And in the second half of 2007 you're at .2 percent losses, so no losses at all.
So nothing horrible has happened except for the fact there's an obvious pattern here. It seems like every newer vintage is getting worse, and worse, and worse, and worse. And so it was on that basis, just seeing things getting worse before there were any losses. Remember, these are very sophisticated people betting hundreds of millions of dollars. They're thinking hard about this. Even though nothing had happened yet the market collapsed.
So this collapse that happens here is before anybody has lost any money. Hardly anyone's been thrown out of their house. Nothing's happened in the economy, but the traders have already figured out things are on a bad course, things are going to look really bad and the market collapsed. So we had warning. This is the beginning of 2007. We're almost at the end of 2009, so it was almost three years ago.
Nearly three years ago everybody should have known that the market had figured out that there was going to be a catastrophe in the sub-prime world, a real catastrophe because the losses were going to be 40 percent. That's what they were expecting. You can't have losses like that unless you throw out, you know, it's a huge number of people that have to get thrown out of their houses, and you have to lose a huge amount of money on each house in order to get losses of 40 percent.
Like I said, if you lose 50 percent on every loan you have to throw 80 percent of the people out of their houses to get 40 percent losses. So this meant that people were expecting gigantic catastrophe. By the way, that was an exaggeration because this is the triple-B piece. I meant the triple-B piece was going to lose 40 percent. So the triple-B piece is only protected by 6 or 8 percent so it means the total losses would be 10 percent, 10 or 15 percent, which means if you're losing 50 percent on a house 30 percent of the people are going to be thrown out of their houses.
So it's like 30 percent losses and 30 percent of the people thrown out of their houses. That was the expectation then, but that's still terrible. There were five million sub-prime people, so the market at this point is expecting 30 percent of them to be thrown out of their houses with 30 percent losses even though the losses are 1 percent or less. So you can see how far sighted the market is. So our government, everybody should have been warned right at this point that something was going to happen.
Now, did the stock market realize this was such a catastrophe, absolutely not because the stock market waited until the end of 2007, which we're talking about here, and now look at where the losses are in sub-prime bonds. Here the stock market started to go down. So it had all this warning that something horrible was happening in the mortgage market, but the stock market owners thought nothing of it. They thought, "Well, how could this possibly affect us?"
And only here do they start to catch on. So to give you an example of the losses, this is only up to December 2007, people kept announcing losses. Merrill announced 9 billion. Citi Corp announced 10 billion and then the next week another 10 billion. UBS announced 3.7 billion and then another 10 billion. Goldman announced 5 billion, Morgan Stanley 4 billion. Every other day someone was coming out saying that they had lost all these billions. And that was only in December 2007.
The losses accelerated like crazy through 2008, so there were hundreds of billions of dollars lost. Now, the whole sub-prime market is only 1 trillion, but remember there was all that insurance written on it. And who was writing the insurance? These very same people were the ones writing the insurance. So you not only lost the hundreds of billions in the sub-prime market, but also multiply that by 5, the hundreds of billions all these guys wrote in insurance and in CDO triple-A's that they held.
So, all right, now let's say one other fact that's quite interesting. These are prepayments. So I told you that in sub-prime loans for the first two years, the sub-prime loan, you'd be locked into the loan. A huge penalty if you prepaid, but between year two and year three you could prepay, get a new loan, refinance it and then in year three the interest rate would jump. So what would people do?
Well, people who made all their payments the first two years would be regarded as reliable payers and so of course they would refinance. They'd no longer be such sub-prime people. They'd refinance into loans with lower interest because they'd be regarded as better credit risks. So if you go back to 2003 you see that after the second year prepayments go up to 70 percent. So 70 percent of the people would refinance their loan between year two and three.
In 2004, that's the pink one, the same thing happened, 60 or 70 percent of all those people over the course of a year refinanced their loans. So that means that the original lenders got 70 percent of the money back. It's like a prepayment like we have in mortgages. Well, in sub-prime these guys had a huge incentive to do it once they proved that they were good people.
And as I told you, most of the sub-prime borrowers are young people who had screwed up with their credit cards and things like that, and after they paid for a while people would realize that they'd settled down and give them a new loan, but look what's happened to these numbers, so if you jump to 2007 all these things just collapsed, no more prepayments, and once there are no more prepayments that means the lenders don't get their 70 percent back.
So if you're a hedge fund figuring, "How much can I lose if I buy one of these sub-prime things," you know that between year two and year three you're going to get 70 percent of your money back so you can't lose more than 30 percent. All of a sudden you're getting 10 percent of your money back which means you could lose 90 percent.
So, why would that have happened? Well, we're going to come to that. So those are the basic facts. I don't have time to say more than that, so any questions about those basic facts? Now we need something to explain how they happened. Yes?
Student: So ultimately the slope of the line is getting steeper, is that what you're saying, so the projected loss percentage, right?
Professor John Geanakoplos: Exactly. So people were realizing--yeah, that was all that it was. People just said to themselves every--that's exactly what I'm saying that people said, "Look, we know that losses are going to increase, but they're probably not going to get very far. Historically they get to 3 percent or 4 percent or something like that. That's it. That's all we have to worry about. And it's early on, so it's not like we've seen any losses, but they just seem to be happening so much faster than they did before. Let's extrapolate that curve."
Now, how do you extrapolate a curve like that? There are so many ways you could extrapolate it. So the market extrapolated and thought, "Gosh, things are getting just much worse," and the market started to collapse. So you see, in real trading markets people respond very quickly to very little information. I mean, their whole livelihood depends on it. It's all they're thinking about and so they're coming up with quick conclusions, which in this case like in many others, they're not so crazy.
They're realizing something bad is happening and they're acting very quickly, and so huge--billions of dollars are being lost just on the basis of this amount of evidence. I think this brings home how fast the market acts and how expectations about the future are driving prices, and therefore how much information--you remember, we said that if you just pay attention to the market you can learn a lot.
Chapter 3. Collateral and Introduction to the Leverage Cycle [00:29:00]
Well, we didn't learn that much, the whole stock market. They weren't paying attention to what was happen. The mortgage market was collapsing. That should have meant something, but they weren't paying attention to it. Nothing in the world seemed to change because there were no people on the streets yet.
Now there are people unemployed, there are people on the streets, there are going to be millions of guys on the streets pretty soon, homeowners on the streets. And of course so now the stock market has caught on that things are really bad. All right, so any more questions about this? So now I want to talk about a way of understanding what happened, and then I'm going to end.
So I'm going to go from more general to more specific. So I'm going to talk about a way of understanding what happened and then we're going to talk concretely about how you would change the mathematics we've been doing all semester. So my view of all this is that it's the leverage cycle, which is--the leverage cycle, as I'll define it in a second, has to do with putting up collateral to make sure you're going to pay, and so far in the course we've paid no attention to that, and in all of economic theory they never paid attention to that.
You can read anyone of those textbooks that were on the reading list by all of the great Nobel Prize winners, and all the great financial economists, and you'll never see the word, hardly ever see the word collateral, and you'll never see, what's the equilibrium amount of collateral? And so that's been entirely missing.
So, in fact, if you think about macroeconomics and the Fed they always tell you that if things go bad in the economy reduce the interest rate. That's the way we can stimulate investment. We can raise the price of assets by lowering the interest rate. If we raise the interest rate we'll lower the price of assets. We can always get the economy to boom or slow down by adjusting the interest rate.
The most important thing ever done in macroeconomics in theory and practice is to establish the Federal Reserve and give it the responsibility of managing interest rates. Now, in my view it should be managing collateral rates and not interest rates. And in fact it has the power to manage, it's been given the authority to manage collateral rates when it was created, but it just never exercises that authority.
Now, as I said in class, Shakespeare already had this idea. In The Merchant of Venice, as we said, they were haggling over the interest rate and not just over collateral and Shakespeare anticipated the impatience theory of interest and Shylock, who you know, the modern reading of The Merchant of Venice is that it's about anti-Semitism and Shylock is supposed to be a bad guy. I don't think Shakespeare intended that at all.
In fact, Shylock is much more sophisticated, by far, and much better understanding of what's going on in the world than either Antonio--well, certainly than Antonio who is the merchant of Venice and Bassanio's actually, I think, quite clever, although Harold Bloom makes Bassanio out to be an idiot and a dupe, and can't imagine how Portia could possibly marry such a guy. But anyway, I think that Bassanio is a very attractive guy and I think that Shylock, he's the only one who really understands what's going on. So he explains the entire modern theory of impatience to Bassanio and Antonio and says that's why the interest rate should be high.
And then they haggle over the collateral. And of course, Shakespeare, in my view, thought the collateral was much more important than the interest because nobody can remember the rate of interest they agreed on, but everybody remembers the pound of flesh collateral. And remember that the play ends with the ships all sinking, apparently, and so should Shylock take his collateral or not, and the court rules not that you should reduce the collateral or you shouldn't take the collateral but you should have a different collateral. It should have been a pound of flesh, but not a drop of blood.
So it's the responsibility of the judicial system and the regulatory body to monitor not the interest rate, not the amount of the borrowing, but the collateral that's put up for it. And that's exactly what my message is going to be. We have to manage the collateral. And so I wrote about this starting in 1997, and 2003, and then again in 2008, and I've written with a bunch of people, and I have students who wrote on it. And Araujo and Pascoa were two Brazilians who wrote about it. I gave a talk there in 1997. I got them going on it.
And then Bernanke himself wrote about collateral in the '90s at the same time, even earlier than I was, but he didn't have the idea of leverage and collateral rate. He knew that you needed collateral. He talked about collateral, but he didn't talk about the ratio and the rate. So what do I mean by that?
If you have a house that's worth 100 dollars, say, it's collateral for a loan because if you don't pay the loan they can take your house. So that's called no recourse collateral if they take your house, but they can't go after you for anything more. Now that, in fact, is not by law what happens, but in practice what happens. If you default on your mortgage they'll take your house, but they aren't going to come after you to try and get your income or something like that.
So it's as if all they can do is take your house. In some states explicitly by law they can only take your house. In other states you still owe the money, but they always just say, "Forget about it. We're just taking your house." So I'm going to assume that they take your house and no more. If the house is worth 100 dollars and you borrow 80 then the collateral rate is 125 percent, because a 100 dollar house is protecting an 80 dollar loan, so it's 125 percent.
The loan to value, it's an 80 dollar loan on a 100 dollar house, is 80 percent. The margin is the margin of safety, it's a 100 dollar house so the lender thinks that he's protected because he's holding a 100 dollar house and he only gave you 20 dollars. So it's 20 percent and the leverage is 5 because with 20 dollars of cash you've managed to buy a house worth 5 times as much, 100. So these numbers are all the same thing. It's just different ways of saying the same thing. So I prefer leverage, but collateral rate, they're all the same thing.
Student: Could you explain what the 5 is?
Professor John Geanakoplos: 5 is a 100 dollar house and you only paid 20 dollars. You borrowed the other 80. So with 20 dollars of cash you can manage to own a 100 dollar asset, so the leverage is 5. So people who borrow a lot can own huge quantities of things even though they had hardly any money.
So, as I say, when you're negotiating the loan you have to negotiate the interest rate, but you also have to negotiate the leverage. How much cash do you have to put down in order to get the house? How much can you borrow using the house as collateral? Now, in the standard theory that Fisher taught that we've described in this class, you've got supply and demand determining the interest rate. Leverage never appeared. We just always assumed everyone was going to repay their loan.
So in my theory supply and demand are going to determine both the interest rate and the leverage. Now that seems a little shocking because how can one equation determine two variables? So I think it's for this reason that economists, basically all these years, ignored leverage because they just didn't see any way to fit two variables with one equation and solve for two variables. So they just said, "Well, for simplicity we'll assume everyone always repays." Now, what is it that's going to determine leverage?
Well, this is common sense as Fisher and Shakespeare before him said that interest rates are determined by impatience. In my view leverage, there are many determinants, but the most important determinant is volatility, the volatility of the asset price.
Why is that? Because you've got a house of 100 protecting the 80 dollar loan, if you think the housing price is going up and down and it might fall below 80 then you're not going to feel very safe. If you think the housing price is rock solid at 100 you're going to feel very safe and you'll even loan more than 80. So the volatility of the housing price, obviously, is what controls how safe the lender feels, and so volatility has got to be the most important determinant of leverage, but it's hard to see how to fit that into a supply and demand equation.
So the theory, my theory, is going to be that the higher the leverage is, which supply and demand is going to determine, the higher the asset prices are, and the lower leverage is the lower the asset prices are. Now, why should this be the case? Well, in my view there are many different potential buyers, let's say a continuum of potential buyers and each of them has a different view of what the asset is worth, and so you have this continuum.
Now, let's, for simplicity even, suppose that everyone has a number in his head about what the asset's worth. That's what we did on the very first day. Everyone had a reservation price of the asset. And let's say, though, that they're willing to buy more than one football ticket. They're willing to buy as many football tickets as they can get as long as the price is less than the reservation price.
So what happens, the people who have the highest valuation naturally are going to buy the assets, and the people with low valuations are going to sell them, and somewhere there will be a marginal guy. So people above here are going to be the buyers, and people below here are going to be the sellers. And what's the price going to be? It's going to be the valuation of this marginal guy.
He's marginal precisely because he's indifferent between being a buyer or a seller which means the price for him is just right. These guys up here think the price is too low. That's why they're buyers. These guys down here think the price is too high. That's why they're sellers. So this is all consistent with what we saw in the football ticket example.
Chapter 4. Contrasts between the Leverage Cycle and CAPM [00:38:53]
Now, one little twist is where does this have to be? Well, it depends on how much money these guys up here have. So Fisher actually said this. If you make people who really like to consume now richer you're going to get a different interest rate, or if you make the patient people richer you'll have a lower interest rate. Well, the same logic here.
If you make the natural buyers richer you're not going to need as many of them to buy the assets and the marginal guy will go up here and the price will rise, not because the payoffs of the asset have changed, but because the marginal buyer has gone up. So to put it another way, a more important way, if it's easier to borrow because leverage is higher, so you can buy that 100 dollar house with only 10 dollars down instead of 20 dollars down, then each of these natural buyers instead of being able to afford to buy one house can buy two houses or two mortgage securities, or buy a house that's twice as big.
So the natural buyers, with a lot of leverage, a few of them will be able to buy all the assets, and the marginal buyer will be very high and the price will be very high. If there's no leverage and no borrowing it's going to take a lot of them to buy all the assets, and the marginal buyer will be down here and the price will be very low. So leverage is clearly going to affect the price.
And notice how shocking this is. What would Fisher say about this? So this sounds like such common sense, how could this be wrong? It sounds like it's so obvious. But now, why didn't anybody say this before? Why didn't Fisher say this?
Because remember what Fisher said. He said the price of every asset is the fundamental value of the cash flows. You take the future cash flows and you discount them by the interest rate, say, and that's what the price is. I'm saying, even if the interest rate doesn't change, you just change the amount of leverage so these natural buyers at the top, fewer of them are going to be able to buy all the assets, then that's going to make the price higher. So the price is not going to be equal to the fundamental value.
In fact, there isn't a fundamental value. It all depends on the different opinions of the different people. So what are the reasons they differ? Why should there be people who disagree about what the value is?
I think that there are basically four reasons, or five reasons. Four of them, anyway, are one, these people, let's say, are more risk tolerant than the people down here. So the people down here are so scared to hold these assets because they might pay low numbers, low payoffs in some states and high payoffs in other states.
So that assumes that they can't insure themselves against everything. So that's another crucial ingredient to my theory which I'm not going, you know, in the background here is the idea there's not insurance for everything. So the asset here is going to be regarded as risky if it pays less in some states than other. So because some people are less risk averse, they're more risk tolerant, they're going to be willing to pay more for the asset. That's one reason.
Another reason is these people might just be more optimistic than these people. So far we've assumed that everybody believed in the same probabilities. What if these guys' probability of the up state is always higher than these guys' probability of the up state? Then they're going to be willing to pay more than these guys are.
Another reason might be these people simply like the asset. They like living in the house. These people are the New York bankers. They're just lending money. They couldn't care less about living in the house. The house to them is just a source of income. For these people, they really want to live in the house.
So another thing could be some people just have more expertise. They know how to read all these statements of loans. They can analyze the loan level data, and these people don't have the time, or it gives them a headache, or they don't know how to find the loan level data to go through, house by house, who the homeowners are and how reliable they are, so naturally they're going to be scared of buying because they're going to figure these guys know more than they do.
So these are all different reasons why some people would be willing to pay more than other people. And so therefore, and as I say, in the standard theory the asset price is supposed to be the fundamental value. We don't talk very much about the heterogeneity of the people and how that makes the definition of fundamental value hard to maintain. So in my theory, in short, in equilibrium in normal times or good times there's going to be too much leverage and therefore too high asset prices, and in bad times there's going to be too little leverage and therefore too low asset prices, and this recurs over and over again, and it's what I call the Leverage Cycle, what I called the Leverage Cycle.
Chapter 5. Leverage Cycle Theory in Recent Financial History [00:43:36]
Now, let me give you an example of what people really had to pay for these things. So for a bank buying a triple-A security there's regulation which forces them to put a certain amount of cash down. So they have their depositors' money, so they can buy things with depositors' money, but they have to put some of their own capital down. So how much do they have to put down? 1.6 percent, so they could leverage 60 to 1, a 100 dollar asset, they only paid 1.6 dollars in cash.
If you look at all the so called toxic mortgage securities of which there were 2 and 1 half trillion, these are the things that crushed all the banks that they held, I went through security by security and made a guess as to how much cash was put down by the buyer of each of those securities and on average I got--so this is not absolutely accurate, there are a lot of guesses, but I got 16 to 1 as the leverage. So that meant that, if that number's right, it means 150 billion dollars of cash was paid, and 2.35 trillion was borrowed.
That's 16 to 1 leverage. So that means that, you know, think about it, Bill Gates and Warren Buffett, two people by themselves, in 2006 had almost 150 billion dollars. So that means two guys in the whole economy could have bought every single toxic mortgage security because they could have borrowed the rest of the money they needed to buy.
So when you think of where this line is what I'm saying is that for all the mortgage securities, all the toxic mortgage securities, sub-prime, all those supposedly dangerous securities you could have had the marginal buyer way up here. That's all that was required. Those are the only number of people, this top little echelon of people bought all those securities, could have bought all those securities because they had enough money to do it, and they hardly had to put anything down.
All right, the same was true with housing. Now, let's look at this theory a little bit. So I should have--sorry, I could have put this in a better thing. So I give all these talks with Shiller for alumni during the crisis. We haven't given one in a while, but there was, you know, lots of them. Like every month or two months we'd be giving one of these talks and they'd bring in all these donors to say, "Yale needs your money now," and they'd say, "We don't have any money. So anyhow, I'm just joking.
So we'd have a lot of these talks and Shiller would always put up the green graph. So on the right hand side, I showed you this before, it's his famous diagram. He did a brilliant thing. He decided to collect data on every transaction. Every time someone purchased a house he'd keep track of it, because it's all public information, and then he'd compare it to the same house that was sold before, and do a few other tricks that I don't have time to explain, to make an index of the housing prices in the whole country.
And so he'd done this going back many years, but starting in 2000 the thing that he called attention to before anybody, he said, "Look at this. Look at this line. The housing price index from 2000 to 2006 it's gone up by almost 100 percent." The housing index was 100 in 2000 when it started, that's 110, sorry. This is 100. It was 100 when it started here in 2000 and it goes up to 190 in 2006. That's 90 percent in 6 years.
To almost double in six years that means it's gone up 12 percent a year or something. So a staggering increase in the price of housing, and then it started to go down. So he called attention to this already as it was coming up. So in 2004 and 2005 he was saying, "My gosh, there's a bubble. It's going crazy," and he'd say, "Everybody is nuts. Americans, there's irrational exuberance and people are so crazy that they just think things are always going to go up and the price is going up higher and higher and higher."
And then when it got so high the narrative changed. People started getting worried and they started telling everyone the world was coming to an end and the prices went down, and down, and down because we have irrational exuberance. And things are picking up a little bit, but we don't know whether people are irrationally exuberant, or pessimistic, or they're blipping, but anyway, they're irrational.
So that's a story that has a certain compelling nature to it and he deserves a lot of credit for calling attention to it. But in my story I say these prices are related to leverage. So I went through every loan, loan by loan, that wasn't a government loan. So these are private loans. You can't get the data on government loans. So on private loans you know every time someone took out a mortgage they were selling the loan, so therefore they published the data. You know what the appraised value of the house was or the sale price if they were buying it for the first time, and you know what all the loans are.
So I added up the value of all the loans and I divided it by the price of the house, and I subtracted it off to get what the down payment was, the cash down payment, so this is 2 percent down, 4 percent down, 6 percent down, or if you take 100 percent minus this, this is the loan to value, so this is 100 percent. You've borrowed everything. Here you borrowed 98 percent of the value, 96 percent of the house, 94 percent of the house, etcetera.
So in 2000 these are slightly riskier people. Amazingly they're putting down 14 percent, not amazingly. They're putting down 14 percent so it's 7 to 1 leverage. But then look what happened. The leverage went up, and up, and up, and up, and up, and up, and then it was, in 2006, less than 3 percent down. These people were 30 to 1 leveraged. I said the securities market was 16 to 1 leveraged, the triple-A part 60 to 1 leveraged. Here they're 30 to 1 leveraged and so the prices went up.
But yes the prices went up because the leverage went up. You hardly had to put any money down. That's why people were willing to buy such expensive houses because they weren't actually putting that much money down. And then leverage collapsed like this and you had to put more and more money down, and then the whole market, you know, you can't get a loan basically anymore unless you're getting a government loan.
And so leverage collapsed and the housing prices went down. Leverage was collapsing faster and in front of housing prices. So I think it's the leverage which is a crucial determinant of the price, and not irrational exuberance, although I think there is irrational exuberance, but I think leverage is what's underneath it and more basic.
Now, let's do the same thing with prices. So we did the sub-prime index before. This red line is the prime index. So we're talking about not sub-prime homeowners, but prime, people with brilliant credit ratings and stuff. So even now years, three years after the crisis has started, the losses on these pools are still in the single digits, 5 percent, 6 percent, things like that and most of these, 8 percent maybe, and most of these, these are the trip--the top piece of a pool of prime borrowers. So most of these loans are, they're protected by 8 percent of the bottom piece.
So you have to go through 8 percent losses before you touch them at all. So it's just inconceivable that you'd see 40 percent losses, but so what happened? These bonds were at 100. This is 100 over here. The price is 100. They're at 100.
Now, the interest rate is changing and the prices are going up and down because they're supposed to be floaters which always pay more or less depending on the interest rate. So they should be locked in at 100, but they're not floating exactly on the same time, you know, they don't change their float so fast. So the price is going up and down because of that, but it's clear that nobody thinks there are going to be any losses. And then all of a sudden in 2007, but later, this is much later than the sub-prime market collapse, these prices collapse, and they collapse all the way to 60, shocking.
It's just inconceivable, I think, that people could have thought there are going to be 40 percent losses, or that everybody could have thought there are going to be 40 percent losses, and then they've shot back up to 80. So there's been a huge gain. So if you look at a typical hedge fund like Ellington you'll see that we lost a lot of money, and now we've made a huge amount of money. So that was not such brilliant hedging, but anyway.
So, but now why did these prices collapse and go back up? Is it that people just got totally irrationally pessimistic and then they said, "Oh, things can't really be that bad," and things got better? Well, let's look at the leverage, what is the down payment, the same thing as before on triple-A securities. Now amazingly the Fed has never kept these numbers, and they're still not keeping these numbers. So I testified in front of Ben Bernanke and the Board of Governors on October 2008. There were several of us, but it was for four and a half hours. And I showed them this graph.
So this graph is at my hedge fund, Ellington--it's just a bunch of securities that we're used to trading. It's how much they were offering to lend us, what the down payment was they were insisting if we wanted to leverage as much as we could. So we didn't leverage anything like that, but this is what they were offering us on the kinds of securities that we're interested in.
So this is not the greatest data because we're only looking at what we're interested in, and we're only looking at the offers made to us. So, but looked what happened. In '98 when there was a leverage crisis, that was the last one, I must have skipped a slide where I said the last leverage cycle ended in '98. It was 10 percent down. That means 10 to 1 leverage, and then suddenly it went to 40 percent down. So leverage drastically dropped. The down payment, the margin, radically increased for a few months during the crisis then it went back to 10 percent.
Then we had 10 percent down for a long time. That's 10 to 1 leverage. And then for a couple of years it became 5 to 1 leverage, so 20 to 1, 5 to 1, sorry, 5 percent down means 20 to 1 leverage. So the leverage had drastically increased, and then leverage suddenly started collapsing and lenders asked for more and more down payments, more and more down payments. And leverage was just collapsing, and it happened ahead of the prices, and then leverage actually picked up here and went down again.
Then--this graph is quite old now, May. I should have updated it. Leverage has gone way up again and the prices have gone way up. So you see, leverage goes down the prices go down. Leverage goes up the prices go up, and leverage is still heading up and the prices are much higher.
So that's my view. It's leverage that's determining a lot of these things. All right, and so as I said, there's recurring leverage cycles. In 1998 there was a leverage cycle. That's when Long-Term Capital--so there was a book on the collapse of Long-Term Capital that Lowenstein wrote that's on the reading list.
Before that was Orange County. You remember my firm Kidder Peabody went out of business in 1994, the end of '94. The beginning of '94, Orange County got bankrupted partly thanks to buying securities from Kidder Peabody. Some of you may live in Orange County. In any case that was the derivatives crisis, and in 1987 there was a stock market crash, so anyway, it's hard to document all the historical leverage cycles because the Fed has never kept this data, and most other firms don't keep their historical data.
We at Ellington realized after '98 that this was a huge problem, leverage, and so we kept all the historical data. That's when I wrote my first paper on leverage and collateral and that being a crucial ingredient of the--so we have this data and practically nobody else has. So for me the most important thing, the first step the Fed should do is keep track of the leverage that is being allowed for, the margins that are being demanded on every security, housing and securities.
And I heard this talk, I told you, last night on the collapse of the electric grid in 2003, and so I said, "What have you done?" Obviously the obvious question everyone wanted to know. So, "What have you done to make it safer?" And they said the one thing they did is now they've spent tens and tens of millions of dollars getting more data. So they now monitor every single power line in real time. They had farmed that out to other people before and they weren't very careful about it. Now they're incredibly careful about monitoring every single line. So as soon as one line goes down they switch the electricity and they move it evenly across a bunch of other lines.
In the old days when a power line went down the grid automatically--sort of, electricity went, the power went to the next closest place and that would overload the next closest one and maybe make that one go down. Now if they know it's gone down they choose how to re-divide it. But the point is, you can't do all that without information, and they're spending tens of millions of dollars monitoring everything and we still are not monitoring what leverage is.
I think it's a shocking thing. We should be monitoring leverage and then we should be regulating it, and instead we haven't even begun to monitor it, much less to figure out how to regulate it. Now, why was the leverage cycle--I said, this recurs over and over again. So why was the leverage cycle so bad this time than it was before? So this is my short version. I have a long version of this.
The short version is because leverage got higher than it ever did before, and there are lots of reasons why leverage got higher than it ever did before. We had this period of low volatility for a long time, so it made it higher. We then had securitization which we didn't have before. So a sub-prime borrower would have had to put down a huge down payment in the old days, because people would have regarded it as such a risky thing, but once we packaged all the sub-prime loans together in a big security and the triple-A guy got the best of the houses, not each house, not a share of every house, but only a share of the very best houses, it all of a sudden became a much safer loan and so you could borrow a lot more money against it, against that triple-A loan.
So the triple-A piece, using that as collateral, you could borrow a lot of money, a big percentage of the price of the triple-A thing you could borrow, whereas if you had a single sub-prime loan people would never lend more than 50 percent of the value of the house or something.
The second reason why things were so bad was because we had a double leverage cycle. We had leverage shooting through the roof not just in securities as it had before. As I showed you in '98 where it was at 10 percent down, and then there was a collapse to 40 percent down, and then back to 10 percent down, that was a big change, in the crisis stage, leverage collapse, but here we had it in housing as well where there was a feedback between the two.
So why is it that all these people stopped being able to refinance their loans in 2007, which contributed so much to the losses? It's not because the interest rates were suddenly too high for them. Those didn't change at all. It's that the lenders suddenly demanded a much bigger down payment.
Instead of 3 percent down they had to put 25 percent down and they couldn't afford to do it, so they couldn't refinance their loan. They didn't have the cash, so they were stuck with the old loan, and so the original lender was in much more jeopardy.
And then the last thing is the CDS market. So I have to explain this. That's why it's going to spill over into next time. So these insurance things, there are many ways of interpreting what they did and why they were so dangerous, but one of them is that they allowed people to write insurance to lose huge amounts of money.
So that's an obvious thing. That's why the losses were so big, but more interesting than that, they allowed the pessimists to leverage themselves, because in our old story from way back here, in this old story--where was the old story with that? In this story, remember, I said that the natural buyers, you'd have fewer of them if they could leverage a lot. So what were these guys down here doing? Well, they couldn't do anything except sell the assets and maybe make the loans.
As the price got higher and higher these guys down here, presumably, were getting more and more disgusted and thinking, "Well, this market's ridiculous, and the price is way too high. We don't want to buy here." And I've talked to a lot of insurance companies like Northwestern Mutual Life, the guy who runs that. He said, "We used to be a big buyer of mortgage securities and sub-prime mortgages and stuff in 2004, and the price got so high in 2006 to us it became ridiculous so we just dropped out of the market."
But dropping out of the market is one thing. What if they could bet against the market? Well, they had no opportunity to do that before. They couldn't express their negative view, but starting in 2005 where this CDS market really took off, they could suddenly express their negative view. So in my view and in my opinion what happened was that at end of 2005 the pessimists for the first time were able to bet on the market going down.
And so that's what started to make prices go down, because of the CDS market. When prices started to go down the mortgage securities that had been sold with all that leverage for high prices started to go down in price. That meant that new securitizations weren't worth it.
You couldn't make the profit that we were talking about before, putting all of these things together and selling off the bonds. The bonds are being sold for a lower price. You couldn't lend all that money to these people. You were getting the money to lend to the people by selling the bonds. If you sell the bonds at a lower price you don't have the money to lend to the people.
So in order to make the bonds sell for a higher price they demanded that the homeowners put up more collateral, and then that meant the homeowners couldn't refinance, and then that started to make the world more dangerous, and that's why, then, on securities, the lenders on securities reduced the leverage and then the whole thing started to spiral. That's my story of how the things started. So let's say that.
In the bottom of the leverage cycle the same bad three things always happen. Number one there's scary bad news that creates more uncertainty and more disagreement. So it wasn't just that people thought that losses were going to go from 40 percent to 20 percent on sub-prime mortgages. It's that they went from 4 percent expected, with a range of 1 to 5 percent, to 30 percent, but maybe they'd be 80 percent, the losses.
There was a huge volatility--nobody really knew what the hell was going to happen. They had to extrapolate those curves. How do you know how to extrapolate? For a decade the curves had been flat. Suddenly they're starting to accelerate. How do you know how to extrapolate them? There's huge uncertainty, huge volatility, and so naturally the lenders are going to be much more nervous and lend only with much more collateral.
Once the lenders lend with much more collateral those people at the top can't afford to buy things anymore. They're going to have to sell. So there's de-leveraging. Sorry, and the people at the top are going to have to sell, but the price is going to go down, and because they borrowed money and using an asset as collateral, their leverage--we talked about that--their losses are going to be gigantic because they're betting, essentially, on things going up, and things are going down so they lost a huge amount of money.
So this is always the same thing. There is scary bad news. Margins get tougher, and the most optimistic leveraged buyers are the ones who got crushed because they were betting on things going up and they go bankrupt.
Chapter 6. Negative Implications of the Leverage Cycle [01:03:55]
So now, what's so bad about this? I'm running out of time here, and I haven't gotten to how the math is going to change. So what's so bad?
So I'm going to talk for four more minutes and hand back the exams, or should I just hand back the exams right now? What do you--are you? I'll talk four more minutes. Some of you aren't going to come next time, maybe most of you, so I'll do how the math changes next time. So what's so bad about the leverage cycle? So let's just think about all the bad things that happened.
Number one, you can have a tiny group of people at the top controlling the price, because they can borrow so much money. Now we've always assumed that everybody was rational. What if these buyers are kind of crazy? A few guys can make the price skyrocket because they can borrow so much money.
Now, once they do that and they're making the price of assets skyrocket, people are going to build more of those assets. So they're going to use real resources to build projects that are probably crazy and are going to be very costly to reverse.
But let's leave aside now people being crazy. What else happens in the leverage cycle? Well, in the leverage cycle when you're leveraged a lot, as we said, that's going on the Tobin diagram to the right. It's like borrowing money to bet in the stocks. That means if the stocks go up you multiply your winnings. If the stocks go down you multiply your losses. So in the leverage cycle people who are lucky, if things keep going up, they get incredibly rich.
It's a tremendous mystery in economics. Why is it that inequality increased so much? Well, one of the reasons, I think, is the leverage cycle, that people were betting and making more and more money and so that's why inequality was--is helping to spread inequality.
Now, another thing is the worst, probably the biggest problem is that once the asset prices fall everything gets hard to do. So if you want to get a new credit card, what does it mean to get a credit card? It means that you sell your promise, right, to a buyer. Now, why should a buyer buy your promise for 100, you're promising to pay back in the future, so he's paying 100 for you when the very same promise that someone made six months ago is selling for 65 because the prices have all collapsed. They're not going to buy the new promises when they can buy the old promises at such a low price.
Who's going to buy a new mortgage when the old mortgages are selling at 65? Who's going to buy a new auto loan when the old auto loans are selling so low? So nobody can get a new loan. Nobody can sell a new promise, and no business can sell its promise. There are old businesses that everybody understands whose bonds are out there being traded at very low prices. The investors are going to all buy the old assets not the new assets. That's why activity collapses.
Then another terrible thing is, what if the optimists are indispensable to the economy like the big banks? They go out of business and then everything topples. Well, why didn't the big banks take that into account? Shouldn't they have realized that the losses would be so big and try to protect themselves?
Well, they're not taking into account that when they go under and the managers lose a lot of money that there are other people they don't care about like all their workers and the rest of the economy that's losing even more that they're not paying attention to. So they're not internalizing all the losses when they go out of business.
Another incredibly bad thing is, what we're suffering from now is, debt overhang. There are a tremendous number of people, homeowners to name one, who are under water. That means they owe more money than their house is worth, or banks that have more debts than the value of the assets they hold, or firms that have more debt than the value of the assets they hold.
Now, many of these people don't just go bankrupt. They keep going because maybe they'll get lucky. Maybe some miracle will happen and the stuff they own is going to go up in value. So they don't stop. They still exist, but they're what people have called zombies. They're not doing anything productive because what's the point in fixing up your house if more than likely you're going to end up losing it anyway.
So all these homeowners are not doing anything to take care of their houses, the banks aren't making loans, these firms aren't making productive investments because they know they're so far under water.
So another horrible thing happens is that if you try to collect the collateral when people default, it's a costly operation. So in sub-prime loans it takes 18 months to throw somebody out of their house. So during those 18 months the guy's not paying his mortgage. He's not paying his taxes, all of which you have to make up for. He's trashing the house, or if he doesn't trash it he leaves the house and some neighbor trashes it and rips out everything in the house.
The average sub-prime loan, when you force someone out of their house and sell it, you get back 1 quarter of the value of the house now. At the beginning people assumed it could never be less than 80 percent. Now it's 1 quarter.
Then there are other more complicated things that can go wrong which I'm going to skip over. So let's just see what happens now with housing. So what is it that's happening with housing? If you look at people underwater, why are people defaulting on their houses? It's because they're underwater. So here is for every different kind of mortgage, starting at sub-prime and going down to prime. These are except government loans.
This is the fraction of people who default every month, not every year, and here's their loan to value. So how did I figure out the loan to value? I had to look at all the original loans. I did this house by house, or Ellington, my hedge fund, actually the guys there who were former students here, they actually figured all this out.
They went house by house. They got what the loan sizes were on all the houses, and then using the Shiller index they know what the house was bought for, and assuming the zip code index dropped by 20 percent they figured every house in the index went down by 20 percent, so they could get the current value of the house. So this is the ratio of the loan to the current value of the house. Current combined loan to value ratio. So 140 means the loans all together, first and second loans are 140 percent of what they estimate the value of the house is.
So look at what happens. If you're sub-prime and you've got that--I can't see my--this is--140's over here somewhere, right? You're defaulting more than 6 percent a month. If you're 160 you're defaulting at 8 percent. Every month 8 percent of those people are becoming serious delinquent and probably will never repay; 8 percent a month. In one year they're all going to be gone. So you can see this is shockingly sensitive to loan to value.
It's not that people are defaulting because they lost their jobs. They're defaulting because it's not worth it for them to pay. They're not stupid. Why should they pay 160,000 when the house is only worth 100,000? It's just how can they, you know, times are tough. What are they going to do, tell their children, well, we can't feed you? You can't go to school, because daddy and mommy have to pay this thing where we don't really have to pay it?
They're not going to do it. So somehow, I couldn't convince Larry Summers of this, I tried hard. I wrote an Op-Ed in The New York Times in October of 2008. I wrote another one in 7, October. No, a year ago. What are we talking about, October of 2008 then March of 2009, saying that the only way to save the housing market was to write down principal.
Now, writing down principal means saying, if the loan is 160,000, maybe the house used to be 200,000, the house is only 100,000 now. You just say, "We're writing off 80,000 of the loan. It's now only 80,000 that the homeowner owes." Now, who's going to pay for the lost 80,000? I say, nobody has to pay for it.
The government doesn't have to pay a penny. The bond holder, the lender would be better off writing down the loan to 80,000 than throwing the guy out of the house and getting 40,000 in the end, because once the loan is written down to 80,000 the homeowner now has 20,000 of equity in the house. He'll either work hard to fix it up and spruce it up and sell it for 100,000, or he'll actually pay the 80,000.
Either way the lender is going to get 80,000, not 40,000. So this is a win-win as I called it in the editorial with Susan Koniak, a coauthor, a law professor at BU. It's a win-win situation and it's not happening. And it's not happening because the loans are all in the hands of the servicers.
So remember I said there was this big pool with all the loans in them and then there were the bonds. And so the bond holders, these are the people who'd like to see the principal written down, but the bond holders don't know who the homeowners are and aren't legally allowed to talk to them. There's a servicer in the middle who's a big bank who's writing the letter saying pay up, and if they pay up they distribute the money to the bond holders. If they don't pay up the servicer throws them out of the house.
The servicer, the big bank doesn't own the loans. They're getting paid a percentage, like .5 percent of the principal, so they have no incentive to write down these loans. If they write down the loan in half they lose half their fee. If they write down the loan to 80,000 and then the guy sells the house they don't service the house anymore. They lose all their fees.
Instead, Obama, thanks to my classmate Larry Summers and all his other advisors, they've given these servicers license not to write down any loans, but to pretend that they're writing down the interest on the loans. Now, that would be great, and that's actually what's happening. The servicers are basically not even kicking the people out of their houses. So these people are not paying anything now. There are millions of them not paying anything still living in their houses.
In a year or two they're going to get thrown out, but of course they're not going to pay. This is the ideal situation for the servicer because he's still collecting his fee, the homeowners can't sell, they don't want to leave because they're not paying anything, and the bondholders are getting screwed.
So it's a catastrophe. And finally, after a year, if you read the Op-Ed's in The New York Times and the The Wall Street Journal, in the last week or two they have been deluged with people saying, "Why aren't they writing down principal?" So this is something that a year or two ago I already advocated. Anyone with any common sense would have seen that was what was inevitably going to happen and we didn't do anything about it.
Chapter 7. Conclusion [01:14:14]
So I've got one more minute to finish. So what should we do about the leverage cycle? Well, the thing to do about the leverage cycle is to reverse the three bad things that happen in a leverage cycle.
The first one was that there was too little--the most important one is leverage collapse. So the Fed has to go and increase leverage. So it used to be 30 to 1 or 20 to 1 now it's close to 1 to 1 on many assets, although it's gone up a little bit. So the Fed has actually done something about this. What does it have to do? Not lend at lower interest. The Fed cut the interest rate to 0 and said, "Look how great we are." Well, that did nothing practically.
The reason the Fed has cut the interest rate to zero is to give money to the big banks because the banks have to pay depositors like you and me. Now, on your checking account or savings account you're getting zero practically because the Fed lowered the interest rate to 0. That means the banks make money because they don't have to waste any money giving it to their depositors. That's the reason why the interest rate went down to 0, and they try to dupe the public into thinking it's because it's a Keynesian stimulus.
It's not stimulating anything basically. What you have to do is lend with less collateral and the banks aren't willing to do it. The Fed has go around the banks and lend with less collateral, which I'm running out of time so I can't explain how it's actually done that and that's one of the big reasons leverage is starting to go up.
Secondly, you have to replace some of the natural buyers who have gone out of business, the ones at the top who are doing all the buying. Without them the price is going to be worth much less. The government is going to have to do some buying. So through the PPIP program and other things it is doing some of that.
And then finally the first thing, the most important thing, the crisis began with housing. Housing is the first thing you have to do to lessen the instability and the uncertainty in the economy. It's now started with housing and it spread to other people who are under water. That's where all the uncertainty comes from.
So we have to straighten out who's going to go out of business and who's not going to go out of business. We have to write down principal of the homes. That's the way to reduce that uncertainty. And in the long run we have to prevent the next leverage cycle by never letting leverage get so high. So we have to keep the data and we have to prevent people from asking for so little margins.
So I'm finally going to hand back the exam. So in case anyone wants to come I'm going to continue this story and see how the math changes next time. I'll talk for 20 minutes or a half an hour about that and then we'll review, if anyone's curious, the class, you can ask me about the exam, but now we're going to hand back...
[end of transcript]