Week 4 Flashcards
(42 cards)
How many main explanations for the crisis are there?
3: Bad behaviour, government thinking, Bubble thinking
Evidence for which of the two theories is the strongest?
For the third hypothesis.
What movie als makes us think that moral hazard was the reason for the crisis?
Inside job.
So according to the moral hazard hypothesis, where does the demand for mortgages come from?
from confused or mislead buyers, who had been intentionally mislead. We would have in this hypotheses the mortgage lenders pushing borrowers in to mortgages that were wither not appropriate for them or not appropriate for anybody. And doing that with sales techniques.
What is the second component of the moral hazard hypothesis?
The second component talks about mortgage supply. So who’s providing all of the money for these loans? And now what’s emphasized is that the financial intermediaries, the banks, the mortgage brokers, are conflict-ridden.
What is true typical story of the moral hazard ?
So the typical story would be, well a mortgage broker just arranges the loan and gets paid a commission. Never has to worry if the loan actually fails. And a bank that provides the funds for the loan is just going to resell the loan out immediately to somebody else, and they’re never going to get harmed if the loan doesn’t perform well. So that’s where the mortgage supply comes from.
What supposedly facilitates the moral hazard theory?
The whole process is then facilitated under hypotheses one by a revolving door. People going from government to industry to the academic world and around and around into circles. And effectively what happens is when you’re in government, you pass the rules that are good for industry. When you’re in industry you take advantages of those rules. And when you’re in academic you write lots of articles saying that these rules are very, very good while you’re being paid by industry. Now that’s kind of broadly how hypotheses one fits together.
What does hypothesis 2 emphasise?
Hypotheses two emphasizes government failure. Not specific actors, not people who are in a revolving door, but rather government itself and policies that may have been designed to promote goals that we really like, but in the end, end up leading to crisis.
What is the classic example of how the government caused the crisis?
In this case the demand, the homeowner demand, for housing that caused housing to go up. Comes from explicit and implicit government subsidies trying to expand home ownership. The clearest example of that is a tax deduction that you are allowed for mortgage interest. So if you had a mortgage in the United States and you are itemizing your deductions on your tax return, then any interest you pay on your mortgage, you can deduct from your income before you pay taxes.This is actually a very, very large subsidy to taking out a mortgage relative to, for example, renting. So if you’re considering the buy versus rent decision, your rental payments just go to the landlord. Your purchase payments, you are allowed to deduct. So these types of incentives are in fact quite strong and have been in existence for a long time.
For th mortgage supply, from which angle do we come from ?
The mortgage supply, where we’re getting all the money from, here the government angle on it comes from the government sponsored enterprises. And one thing that is often pointed to, is a law passed in 1992 that forced the government sponsored enterprises to follow and to have fair housing goals. So they would have goals to try to bring more lower income people into becoming home owners. And the idea there is well that pushed them to take on more risks and to encourage more of what later became a huge surge in subprime lending.
Where is the demand side coming from in the bubble thesis?
And here, again, we have a demand and a supply side and then a overall facilitation that brings those two things together. What does bubble thinking mean from the demand side? Demand is coming from consumers, from homeowners who view houses as a great investment that never goes down in value. And in fact, this is quite, this comes from survey evidence and it certainly comes from
Where does a lot of confusion concerning housing prices come from?
In part, that comes from a little bit of a confusion that many of us have the difference between real and nominal prices. And nominal prices will just go up when inflation goes up. And for a lot of our recent history, there was inflation just making house prices go up. Whereas real prices require an appreciation relative to say the basket of goods that you actually buy. So if prices of goods and services go up at the same time as house price go up, it’s not an increase in real house prices. Nevertheless, it does make some sense for people to think that when there’s inflation, it’s kind of good for you as a homeowner, since typically you have a nominal mortgage. And that nominal mortgage does not rise with inflation, whereas your home value can. Nevertheless, the overall picture is consumers thinking my house is going to go up in value over time.
Which of the two sides is more interesting to analyse - supply or demand?
Now the supply side is the more interesting one. Because typically we think that the suppliers who ultimately are banks or these large cash pools that we talked about in an earlier module. Of institutional investors who are ultimately buying packages of mortgages. That they somehow should get this right more often than would consumers. But in fact what we will see in the evidence is that, no, that’s not really true. That even these types of intermediaries view mortgage-backed securities as safe assets. Because the assumption is and the assumption backed up by a lot of models that they will look at. Is that housing prices will never fall by very much. And thus if you have this demand to have to hold safe assets either for collateral reasons. Or just for your own investment portfolio, then this housing based assets are a wonderful opportunity.
When is bubble thinking more likely to happen?
After a long period of relative stability. And in the United States we had a long period of stability. Really since the Great Depression, there was not a major, major crisis. We had recessions, but we had nothing that would make people think that the entire financial system or housing system would collapse. And the very long time that it had been, made this really ripe, made it really possible. Just like if there’s a very long time without any kind of epidemic or any kind of disease. People simply haven’t built up immunity for it. In this case we did not have immunity built up to prevent bubble thinking.
Why is the thesis that moral hazard was the reason wrong?
Because it had happened before. Non prime loans performed very well for many years before the crisis. Evidence would not have thought that they were designed to fail.
What does this chart tell us?
We go back to 2000 on the X-axis, and you can see that from 2000 to 2004, the failure rate for subprime mortgages is below 20% and is actually falling. Now whether or not you think this is a big number, it’s a number that had been, this particular failure rate had been around for a while. And if you frame it as almost 20% are failing, that sounds like a lot of failure. If you frame it as more than 80% of the people who took out these loans, who would not otherwise have been able to get home loans were successful in being able to pay off their mortgage, that sounds a lot better. And so, the way that you frame, of course, success or failure here is important. But let’s emphasize just the statistics which is that this was a stable or declining percentage until 2005, 2006 where it starts to go up. If you were standing in 2005 thinking about taking out such a loan, or thinking about making such a loan, and you looked at the evidence, the evidence looks like, wow, this is actually a product that is doing relatively well and certainly even doing a little bit better over time as we increase the amount of loans that we make.
These are loans that have, they’re called low doc or no doc, which means very little documentation. So, somebody is going and they say their income is something, but they have no documentation for their income. They have no W2s, for example. The number of originations of low doc or no doc loans also explodes during the global savings glut period beginning in 2002, 2003. We see that on the origination line. But if you look at failure rates, the failure rates once again are relatively low. So they are below 10% in 2000, and they fall during the period of time when the number of originations is exploding.
Finally, the type of loan that is really quite rare but it’s worthwhile to look at, nevertheless. These are loans that you would think seem completely crazy at the outset. These are negative amortization loans that is loans where people will continue to borrow more rather than in these loans paying off some of your equity in each period, you’re actually increasing the amount of the loan. And in these loans, the LTV starts at 90% or more. So you’re putting at most 10% down. And then over the first few years of the loan, you’re actually borrowing more. Plus, these will be low doc or no doc loans. So we’re putting very little money down, and we are having no documentation of our income. These are the types of loans that you would sometimes hear people call liar loans. So if anything, we would expect the failure rate on these loans to be very high. We don’t see a significant amount of these loans at all until 2003, when the Global Saving Club begins to greatly increase the demand to own the other side of this, to own mortgages. In the early years of these loans, the failure rates were below ten percent. Once again, it is not until we see housing prices falling that we start to notice that these loans are not performing well at all.
What is the second piece of evicence why moral hazard could not be the reason for failure of these morgages?
Another piece of evidence is that mortgage defaults seem actually to be unrelated to rate resets.
Why do people think that people were fooled?
In an adjustable rate mortgage, you start off with a relatively low rate. Because it’s only going to be fixed for a short period of time, typically two years. So we have a two year mortgage, where you will have a fixed interest rate and then after two years is over, that interest rate will adjust. And it will adjust to relative to whatever interest rates look like at that time. So, one thing that is often discussed as sub-prime loans being designed to fail is the idea that you’re starting people with a very low rate, but then after two years that rate is going to jump up much higher. And unless they can pay that much higher rate, they’re either going to end up in foreclosure or they will have to refinance their mortgage. And if they refinance their mortgage then they may have to pay a penalty rate. So it seems some combination of you’re trying to fool people because this all won’t come down the road for a couple of years.
What does this picture prove?
These pictures also come from the Foote, Gerardi, and Willen study. And now, what we’re gonna do is we’re gonna break up these sub-prime adjustable rate mortgage loans into three different groups based on their vintage. The vintage is when the loans were taken out. In the top-left corner is the January 2005 vintage. So these are adjustable-rate mortgages taken out in January 2005 and set to reset with their new interest rate two years later in January 2007. On the top right is the vintage from January 2006, which is set to reset in January 2008. And on the bottom is the January 2007 vintage, which will have its reset in January of 2009. So let’s now look at the evidence.On the right hand axis is the default rate. And along the bottom, on the x-axis, is the year. So we start off with the blue line that’s in between 7 and 8%, and then what we see happen in January of 2007 is a large reset of four percentage points. The interest rate goes way up after two years, exactly the way that we would expect would cause a great increase in defaults. Now look down at the bottom at the cumulative defaults. If you look at this line and you ask is anything actually happening in January 2007 that’s special, the answer is no. And you can do that statistically. But you don’t need to do it statistically. You can do it just by looking at it. There is very little that is going on in January of 2007. Now let’s turn to the January 2006 vintage.
First for January 2005. On the left hand axis is the interest rate, and we see that these loans on average were taken out with an interest rate between 7 and 8%.
First for January 2005. On the left hand axis is the interest rate, and we see that these loans on average were taken out with an interest rate between 7 and 8%.
What does this graph tell us?
Here the interest rate begins a little bit higher, between 8 and 9% when the loans are first taken out. And now the adjustment, which occurs in January 2008, is smaller, because by January 2008, interest rates were lower than they had been for the previous years adjustment, January 2007. So there’s only a slight uptick in the interest rate in January of 2008. Nevertheless, the cumulative defaults just appear to be rising steadily throughout the entire period. In fact, the January 2006 vintage eventually reaches a default rate of 45%, significantly higher than the default rate for the January 2005 vintage, which here never goes above 20%. Despite the fact that it was the January 2005 vintage that had that 4% spike when the interest rates were reset.
