00:00:02
this knowledge at wharton podcast and
00:00:05
videocast is brought to you by wharton
00:00:07
executive education for more information
00:00:09
on Wharton's executive programs such as
00:00:12
the executive development program
00:00:14
leading and managing people or the new
00:00:17
high potential leaders course please
00:00:19
visit executive education wharton upenn
00:00:22
dot edu although the subprime crisis
00:00:28
seems to be showing some signs of easing
00:00:30
debate over what caused it whether it
00:00:33
could have been prevented and how long
00:00:35
it might last will continue for some
00:00:37
time to come knowledge at Wharton asked
00:00:39
Wharton finance professor Richard
00:00:41
Herring for his perspective on the
00:00:43
latest economic developments welcome
00:00:46
professor herring glad to be here the
00:00:48
subprime crisis seems to have aggravate
00:00:50
been aggravated by the use of
00:00:52
mortgage-backed securities that many
00:00:54
people didn't understand and now are
00:00:57
hard to value what was different about
00:00:59
these securities from The Plain Jane
00:01:01
mortgage-backed securities that Fannie
00:01:04
Mae and Freddie Mac have been writing
00:01:06
for years you're absolutely right Jeff
00:01:09
the origination of the securitization
00:01:12
market which is probably one of the most
00:01:14
important financial innovations in
00:01:16
post-world War two ERA was based on
00:01:20
securities that were guaranteed by
00:01:23
fannie and freddie which had the
00:01:25
implicit guarantee of the US government
00:01:27
so the private sector didn't have to
00:01:29
worry about the creditworthiness of
00:01:31
these instead of these instruments
00:01:33
although they did have other tricky bits
00:01:36
such as pre payment options and so forth
00:01:39
private securitization evolved as people
00:01:42
found ways to substitute for the credit
00:01:46
guarantees from Freddie and Fannie and
00:01:49
there were essentially three ways that
00:01:50
people did this one was reliance on
00:01:53
credit rating agencies which would
00:01:55
examine each of these securities and
00:01:59
issue them a letter grade with several
00:02:03
notches around it indicating how safe
00:02:07
they felt that would be the second
00:02:09
approach was to my
00:02:12
them very carefully using statistical
00:02:14
models based on past repayment history
00:02:17
to try to indicate that the amount that
00:02:21
you would need to set aside to make
00:02:23
securities perfectly safe and the third
00:02:26
technique was credit risk crashing um
00:02:29
you designated certain claims on a pool
00:02:33
of securities as bearing the brunt of
00:02:36
the risk and if you had a sufficient
00:02:37
amount of those securities then you
00:02:40
could make most of the rest of the
00:02:41
Securities very very safe this was an
00:02:45
extremely popular innovation it led to a
00:02:50
profusion of perhaps six trillion or
00:02:53
more in securitizations many of these
00:02:57
mortgage-backed and it began to become
00:03:00
more and more complex the residential
00:03:04
mortgage-backed securities are fairly
00:03:06
straightforward it's relatively easy to
00:03:09
see what the underlyings are there is
00:03:12
some pretty close relationship between
00:03:14
your claim on the pool and the pool
00:03:17
itself but there was a much bigger
00:03:20
demand for investment grade securities
00:03:23
then there are investment grade issuers
00:03:26
in the world so the financial engineers
00:03:29
began to to innovate to find ways to
00:03:33
synthesize investment grade securities
00:03:35
by making combinations of these
00:03:37
underlying mortgage pools and other
00:03:40
sorts of of securities one famous
00:03:44
innovation were called collateralized
00:03:48
debt obligations and collateralized debt
00:03:52
obligations would take pieces of the
00:03:56
securitized pools often times lower
00:03:59
pieces but sometimes upper pieces they
00:04:02
would add to them pieces from other
00:04:04
pools and they would also tranche the
00:04:06
CDL so that they designated with the
00:04:11
assistance and it's a little bit
00:04:15
ambiguous to say the role of the ratings
00:04:18
agencies it was something less than
00:04:20
being consultants and how you do it but
00:04:23
it had the same
00:04:25
effect because a potential issuer would
00:04:27
essentially show a securitization to an
00:04:30
issue to a ratings agency asked what
00:04:34
kind of reading do I get with this if
00:04:35
they didn't like the answer they would
00:04:37
go back and try another set so you were
00:04:40
able to get pieces that would bear the
00:04:44
first trash and these would be below
00:04:46
investment grade but that mean meant
00:04:48
that the rest of the pool had a very
00:04:50
very high quality and boring cressman
00:04:53
investment-grade ratings this could be
00:04:57
complicated still more by doing cdo's of
00:05:01
CDOs or CBS Squared's or cubes in
00:05:04
principle and sometimes they were fed
00:05:07
into structured investment vehicles or
00:05:11
sometimes an asset-backed commercial
00:05:13
paper conduits and all three of these
00:05:16
and indeed the whole CDO market has
00:05:19
completely collapsed and essentially the
00:05:22
reason is that the three props that made
00:05:24
people feel confident in making
00:05:26
investments in these things have turned
00:05:28
out to be not very reliable the ratings
00:05:32
have shown themselves to be completely
00:05:36
unreliable for this whole category of
00:05:39
structured finance we've seen multi
00:05:42
notch downgrades of many of these
00:05:44
securities that are huge multiples of
00:05:47
what one would expect even in the worst
00:05:49
times for corporate debt so the implicit
00:05:53
assumption and I would stress it it as
00:05:55
an assumption because there were lots of
00:05:57
reasons that people should have believed
00:05:59
otherwise but the implicit assumption
00:06:01
that a double a corporate bond was the
00:06:04
same as a double a trash of a
00:06:05
securitization was simply false people
00:06:09
also lost faith in the underlying
00:06:11
statistical models because the most
00:06:14
sophisticated players in the business
00:06:16
the people who are really making markets
00:06:18
in this bear stearns city bank ubs had
00:06:22
major losses Merrill Lynch had major
00:06:24
losses and indeed had trouble figuring
00:06:27
out even what their exposures were so
00:06:29
people completely lost faith in the
00:06:31
valuation models
00:06:32
and then finally monoline insurance that
00:06:35
that last safe guide became unreliable
00:06:39
as people started to worry about the
00:06:41
mounting losses relative that they had
00:06:43
guaranteed relative to their equity base
00:06:46
and their ratings came under pressure
00:06:49
once they've lost the ratings they're of
00:06:52
no value and that actually led to
00:06:54
contagion to other kinds of products
00:06:57
that are guaranteed by monoline ensures
00:07:00
such as muni bonds one of the problems I
00:07:03
gather is that if you you go back to the
00:07:05
kinds of securities that Fannie and
00:07:07
Freddie been issuing for a long time
00:07:09
they have track records they know how
00:07:11
many homeowners default or fall behind
00:07:13
on their payments or and so they can
00:07:16
they can project pretty accurately based
00:07:19
on history how new batch of these
00:07:21
securities will behave but some of these
00:07:23
newer securities had no such track
00:07:25
records and some of the models that were
00:07:28
used to guess what the track records
00:07:30
were or just faulty is that right that's
00:07:32
one of the problems behind the
00:07:33
statistical models there simply wasn't
00:07:35
enough data over enough cycles to be
00:07:38
able to predict another problem was that
00:07:41
underwriting standards had deteriorated
00:07:43
in an amazing way which undermined the
00:07:47
assumptions about diversification
00:07:49
because one of the ways you can perform
00:07:52
the financial al coming of turning less
00:07:55
than investment grade securities into
00:07:57
investment grade securities is by
00:07:59
diversifying out all of the
00:08:01
idiosyncratic risk well it turned out
00:08:03
that underwriting standards had had
00:08:05
deteriorated across the board almost all
00:08:08
over the United States now just to focus
00:08:11
on these the alchemy that you talked
00:08:14
about and to try to put it into layman's
00:08:16
terms as much as possible as I
00:08:18
understand it you you you have say a
00:08:20
large pool of mortgages and all together
00:08:22
it has a certain level level of risk a
00:08:24
kind of average risk for the whole thing
00:08:27
but then you could slice it up and say
00:08:29
well we're going to give certain
00:08:30
investors first dibs on the payments
00:08:33
made by homeowners and they will be very
00:08:37
low risk and they'll get relatively low
00:08:39
interest rates and then other investors
00:08:41
will have sort of last dips they'll be
00:08:43
the first ones to suffer if people fall
00:08:45
behind it
00:08:46
but in return they'll be rewarded with
00:08:48
higher interest rates assuming they get
00:08:51
what they're expecting is in layman's
00:08:52
terms is that roughly it that's pretty
00:08:54
much it but there was some other bells
00:08:56
and whistles attached to make it safer
00:08:58
still if you were not utterly persuaded
00:09:01
that simply having a buffer of creditors
00:09:05
who would suffer loss before you were
00:09:07
you might insist that you have a trigger
00:09:10
clause in your contract such that if
00:09:13
there were cash flow problems you could
00:09:15
demand that the entire pool be
00:09:17
liquidated and you be paid off or if you
00:09:20
were very very very conservative you
00:09:23
might say well gee this is all good but
00:09:25
I'd feel even better if I had monoline
00:09:28
insurance guaranteeing my part of it and
00:09:31
most of the monoline insurance went to
00:09:33
the the Triple A which we came in a
00:09:36
super triple-a tranches now some people
00:09:39
have described the collapse of all this
00:09:41
and the subprime mortgage market as a
00:09:43
kind of perfect storm because a lot of
00:09:47
these securities were predicated on the
00:09:50
assumption that home prices would
00:09:52
continue rising and that interest rates
00:09:54
would stay low and that people's incomes
00:09:57
would rise enough to cover future resets
00:10:00
as their as their mortgage payments went
00:10:02
up and on the other hand it looks like
00:10:05
those three things although they don't
00:10:08
always happen at the same time a
00:10:09
reversal in those three things is not
00:10:12
all that unpredictable if you say well
00:10:14
interest rates were to extraordinarily
00:10:15
low levels assume well they've got to go
00:10:18
up and the same with home prices that
00:10:19
were extraordinarily high was this an
00:10:21
unpredictable perfect storm or were
00:10:23
people just ignoring the risks that were
00:10:26
apparent there it's the sort of thing
00:10:30
that happens time and again bubble after
00:10:32
bubble there was in no sense a perfect
00:10:36
storm in this case it was simply a
00:10:37
return to no more normal sorts of
00:10:40
economic relationships if you look at
00:10:43
the growth and house prices in the
00:10:45
United States it's been relatively
00:10:48
modest until about nineteen ninety-eight
00:10:50
you were making about as much on your
00:10:53
house as on your treasury belt it's as
00:10:55
if you'd invested in Treasury bills
00:10:58
the appreciation started just about as
00:11:01
interest rates went down they were very
00:11:03
which was very fortuitous because it
00:11:06
helped soften the collapse of aggregate
00:11:09
spending in the recession but it led to
00:11:13
a circumstance where most of the people
00:11:17
doing the business had not in their
00:11:20
lifetime experienced a downturn in house
00:11:23
prices and there are a number of
00:11:26
psychological mechanisms that suggest
00:11:28
that when people don't actually know the
00:11:30
probabilities of things they will tend
00:11:33
to extrapolate recent experience they'll
00:11:36
use something called the availability
00:11:37
doctrine to see how easy it is to
00:11:40
imagine something happening well even if
00:11:43
you do start with the notion that these
00:11:44
could decline if over time you've had
00:11:48
more and more favorable experience that
00:11:50
tends to fall to a very very low number
00:11:52
and if you believe that house prices are
00:11:55
going to continue to go up then you
00:11:57
don't care so much about the ability of
00:11:59
the borrower to repay because you figure
00:12:02
you can always get the value out of the
00:12:04
house now obviously these are all
00:12:08
especially in from the perspective of
00:12:12
hindsight utterly foolish assumptions
00:12:14
yet they are characteristic of virtually
00:12:17
all the bubbles that we've experienced
00:12:19
and there seemed to be bubbles at least
00:12:21
once every 10 years or so well in this
00:12:24
bubble as in the.com bubble before and
00:12:27
in others that many of the participants
00:12:29
have been punished pretty badly
00:12:30
shareholders have lost a lot of money
00:12:33
top executives that Wall Street firms
00:12:36
have lost their jobs although they don't
00:12:38
seem to wandered away impoverished is
00:12:41
the marketplace punishing people enough
00:12:44
to deter this kind of behavior in the
00:12:46
future or do you think something needs
00:12:48
to be done on the regulatory area well
00:12:51
there are thats a very complicated
00:12:53
question the market is to a very large
00:12:55
extent self-regulating the ratings
00:12:58
agencies are already all three of them
00:13:00
very busy trying to figure out ways to
00:13:03
improve their methodology and to make it
00:13:05
more transparent
00:13:07
as you mentioned several executives have
00:13:09
lost their jobs but not necessarily
00:13:11
their bonuses and there are some real
00:13:14
questions about compensation question
00:13:16
compensation structures that I think
00:13:18
probably will need to be reviewed I
00:13:22
think more fundamentally we need to
00:13:24
rethink the way in which we have
00:13:27
redesigned bank regulation bank
00:13:30
regulation under Basel to the new view
00:13:33
that's been worked on for over a decade
00:13:35
relies very heavily on credit ratings in
00:13:39
the one hand or for the more advanced
00:13:40
banks the internal models and both of
00:13:43
them have turned out to be not very
00:13:45
reliable beyond that liquidity issues
00:13:50
have turned out to be enormously
00:13:52
difficult those are to be handled under
00:13:54
pillar 2 but in a as yet to be
00:13:59
determined why we're promised that the
00:14:01
Basel Committee is going to tell us
00:14:02
something about that later on another
00:14:05
issue that is surfaced that we thought
00:14:08
we'd put to bed with Enron is the use of
00:14:11
special purpose vehicles or
00:14:13
off-balance-sheet entities these special
00:14:17
purpose vehicles did in fact comply with
00:14:19
the post Enron rules they just became
00:14:22
more complicated off-balance-sheet
00:14:24
vehicles but at the end of the day a
00:14:27
reputation risk caused most of the
00:14:30
sponsors particularly the large bank
00:14:31
sponsors to bail them out yet neither
00:14:36
the regulators nor the shareholders in
00:14:37
general even knew they existed so we've
00:14:40
in fact created an off-balance banking
00:14:43
system that is huge very nearly as large
00:14:46
as the on balance sheet banking system
00:14:49
one of the proposals for dealing with
00:14:52
that is still forming I think but the
00:14:56
general idea is to require that the
00:14:58
people who create the institutions that
00:15:00
create the off-balance sheet entities
00:15:02
keep some skin in the game as people
00:15:05
have said that they continue to own the
00:15:07
kinds of securities that they're trying
00:15:09
to sell to other people and so that they
00:15:11
share some of the risk do you think
00:15:13
that's a good idea I think that it would
00:15:17
improve the incentive structure for
00:15:19
sure although we have to remember that a
00:15:22
number of the institutions that have
00:15:24
suffered the biggest losses maybe
00:15:26
unknowingly had a lot of skin in the
00:15:28
game as well Merrill Lynch Citibank and
00:15:30
UBS we're all big producers and Bear
00:15:33
Stearns were all big producers of these
00:15:35
securities yet themselves got caught up
00:15:37
in it what is your general assessment of
00:15:40
the way the Federal Reserve has
00:15:41
responded to this they've been very
00:15:45
innovative they have taken the view that
00:15:49
the first thing to do is ensure that
00:15:51
that there's sufficient liquidity to
00:15:55
enable markets to function well they've
00:15:58
been real problems in doing that because
00:16:00
it is more than a liquidity problem it
00:16:03
is potentially a very serious capital
00:16:05
problem and there's a capital problem
00:16:08
not only because of the losses the key
00:16:11
institutions have suffered because of
00:16:13
the new business they will not do but
00:16:16
have an inventory all of the loans they
00:16:17
claimed it and securitize but also
00:16:21
because of the potential that they may
00:16:25
have to bring a lot of this off balance
00:16:26
sheet business back onto their balance
00:16:28
sheets and to do that will require
00:16:30
raising enormous new amounts of capital
00:16:32
so they've been very reluctant I think
00:16:36
to lend to each other a long term
00:16:38
because they think they may well have
00:16:40
very much better uses for that capital
00:16:43
in their own book of business one thing
00:16:46
relative to earlier crisis that I
00:16:49
applaud and think has worked very well
00:16:51
is the prompt corrective action
00:16:56
solutions that we've put in place in the
00:16:58
United States have encouraged banks to
00:17:02
recapitalize very very quickly the
00:17:05
sovereign wealth funds to some extent
00:17:07
private equity to some extent
00:17:08
shareholders have been recapitalizing
00:17:11
these banks and massive amounts which
00:17:14
will enable them to continue and to give
00:17:18
them some time to restructure and and
00:17:20
find new and more profitable business
00:17:23
models but contrast that with what
00:17:25
happened in the 1980s when basically the
00:17:29
hypothesis the Fed was using is that the
00:17:31
best way to recapitalize the system is
00:17:33
to ignore the losses and let banks
00:17:37
aren't enough on other lines of business
00:17:38
so that they can retain earnings to I
00:17:41
replace the capital that it that we all
00:17:43
know is already lost that gets this is
00:17:45
the final question which is do you think
00:17:47
the recovery will be faster than it was
00:17:49
in the 80s and are we closer to the end
00:17:51
now than the beginning I think that it
00:17:55
much depends what's different about this
00:17:57
of course is that it is in the housing
00:17:59
sector and much depends on how long it
00:18:04
takes to work down the huge inventory of
00:18:08
houses that is is really very
00:18:11
substantial and there isn't a lot that
00:18:14
the Fed can do about that there is a
00:18:18
genuine excess supply of houses out
00:18:21
there that will have to somehow be work
00:18:23
through now this is an entirely bad news
00:18:26
if you're a young household tried to
00:18:31
find a home there may well be good
00:18:34
bargains ahead and buy lots of measures
00:18:37
the house prices we were seeing
00:18:39
generally over the last three or four
00:18:40
years were simply unaffordable to Middle
00:18:43
American so there is there's a sense in
00:18:47
which this is a transfer costs problem
00:18:49
but there will be winners as well as the
00:18:52
losers we all read about in the
00:18:53
newspaper I'm sure we'll be unfolding
00:18:55
for quite some time thank you very much
00:18:57
my pleasure
