Former U.S. Secretary of the Treasury Timothy F. Geithner and Professor Andrew Metrick survey the causes, events, policy responses, and aftermath of the recent global financial crisis.
Lecture: Hypothesis 3: Bubble Thinking
Loading...
From the course by Yale University
The Global Financial Crisis
1135 ratings
Meet the Instructors
Andrew MetrickMichael H. Jordan Professor of Finance and Management
Yale School of Management
Timothy GeithnerLecturer in Management, Yale SOM, Former U.S. Secretary of the Treasury
Yale School of Management
[MUSIC]
Let's turn now to the evidence for Hypothesis 3.
And we will see this explanation,
this hypothesis is consistent with preponderance of the evidence.
We've eluded to some of it already and let me just look at a few more things now.
First the demand side.
Home owners and whether or
not they are thinking will house prices continue to go up.
One thing that we know about home owners and about retail
investors in general is that they are prone to extrapolate recent trends.
And there's a survey of what is likely to
happen to house prices, and house prices here are based on the Case-Shiller index.
This is a well-put-together index that gives us both national and
city-by-city of trends in what's going on in housing prices.
As the bull accelerated in 2003, the expectations of future increases,
in the Case-Shiller index were far above their historical averages.
So here's some survey evidence.
Unfortunately, we don't really have survey evidence that precedes the build up,
because we didn't really have a good index until then.
But right around this time in 2002, 2003, we start to get some survey
evidence, in this case from a firm called Pulsonomics,
of what to expect for housing prices over the next 10 years.
So they are surveying thousands and thousands of people.
And what do they find?
As of 2004, just as we're really seeing the amount of lending explode,
the average expectation for ten year home price increases were 12%.
Now again remember over the previous 100 years or
so, before the great buildup to the housing crisis.
Real home prices are approximately flat.
So thinking that they're gonna go up 12% in the next ten years is quite bold.
Now over the years that follow, this euphoria falls, but only slightly.
Even in 2008, as the financial crisis is beginning,
the expectations of home price rises are approximately 8% over the following years.
If you expect your home price to go up, that's gonna be a pretty good darn reason
to decide to make a bet, to take out a loan, and to buy a house,
even if you think after a few years, you might have to refinance that loan.
And that's what the evidence shows.
Turning next to the mortgage supply side,
what did the seemingly sophisticated investors who were making these loans,
or who were ultimately buying them, what did they think?
In an earlier lesson of this module,
we already saw that actually they lost a lot of money and had a lot of exposure.
But what did they think?
Is there any way to know whether they had the belief or
they really understood what was going on?
And in fact, what we do see is that they did understand that if there was a housing
downturn, it would have a very, very bad effect on mortgage-backed securities.
So that's effectively math.
You design these securities.
And you're designing them, if you recall from our discussion of securitization,
so that there's a nice layer of protection.
So that if you have the triple A tranche of a mortgage-backed security,
prices need to fall a significant amount before it's going to hurt you.
They understood that as we will see.
They just didn't think that it was very likely.
So let's take a look at, this is just two examples of reports.
Extracts from reports.
The first is from Lehman Brothers and the second is from JP Morgan.
And the Lehman Brothers report kind of
breaks out some different scenarios for what could happen to housing prices.
So this is the top half of the exhibit that you're now looking at.
On the left-hand side, you kinda have their cases.
So, the top case is an aggressive case, and
that scenario is an 11% rise in housing prices, over
the life of some pool of mortgage-backed securities that were put together.
Their second case would be an 8% rise over the course of those loans,
and then what they call their base case, which is that housing
price appreciation actually slows to 5% by the end of 2005.
Then a pessimistic case and their pessimistic case is 0%.
So already the language that their using suggests that perhaps they're not well
calibrated to history, just in terms of percentages.
And we'll see that, too, because the pessimistic case
would really be the reality for many of the previous decades.
And then what they call a meltdown case, 5% drop per year for
the next three years, and then an increase in 5% per year after that.
Now, what make this most interesting, in addition to the language,
are the probabilities that they assign to each of these cases.
Their base case in the middle gets a 50% probability.
All the way at the bottom is the melt down case.
And in the meltdown case, they're assigning a 5% probability.
A 5% probability for
dropping by 5% over the next three years, and then 5% thereafter.
Now, if you look all the way on the right hand side column,
it tells you what the losses would be to the mortgage backed securities,
or all pooled mortgage backed securities in each of these cases.
They do understand at the bottom there's a fair amount of losses, 17.1%.
The real problem, though, is that they assign such a low probability for
that happening.
Only 5%.
And it's not just Leeman Brothers.
Leeman Brothers of course we know failed, so
perhaps it's not that surprising that they got the probabilities wrong.
The bank that did the best during the crisis is JP Morgan.
Let's just look at a few extracts down at the bottom from JP Morgan analysts.
These are views on house price appreciation.
And the title of the report, or
the blurb that would come out describing these views, is given over on the right.
And the date of the report is all the way on the left and
it draws from data that is given in the second column.
So for example, in December 2006, with data drawn through October of 2006,
the title of the report is more widespread declines with early stabilization signs.
So it was indeed in 2006,
that observers of the market were seeing a broad based declines in housing.
So we didn't have to wait for the crisis that began in 2007 to see that.
Housing was already weak towards the end of 2006.
But right away we see early stabilization signs.
You look at the next line, one month later, continuing declines,
with stronger stabilization signs.
Okay then.
One month later, tentative stabilization in home price appreciation.
One month after that.
Continued stabilization in Home Price Appreciation, and
then jumping ahead six months, near bottom on Home Price Appreciation,
that's data through July 2007.
The housing crisis really beings in earnest or rather the financial crisis
that came from the housing crisis begins in earnest in August of 2007.
Right before there, we see near bottom on home price appreciation.
Just two months later we get an all caps ugly.
By then it was clear that we were on the way down.
But before that, in the years leading up to the worst parts of the housing crisis,
broadly across all of the evidence that we have seen in this module,
insiders had a lot of exposure.
Insiders lost an enormous amount of money.
And they were telling themselves,and their investors,
that things were likely not going to get that bad.
So we can ask if this is what we see, if indeed there was a bubble, and
that bubble came out really between 2003 and 2007.
Why, why then?
The simplest explanation for why then is that there was a global macroeconomic
condition leading to enormous demand for these underlying securities.
Housing bubbles, indeed asset bubbles in general, are always lurking.
And the longer we go without having one, the fewer people will be immune
to thinking that a bubble will go on and on forever.
The spark that lights this fire is just lots and
lots of investors willing to take these things onto their portfolios.
[MUSIC]
Explore our Catalog
Join for free and get personalized recommendations, updates and offers.
Coursera provides universal access to the world’s best education,
partnering with top universities and organizations to offer courses online.
© 2018 Coursera Inc. All rights reserved.
