|
One
month ago, as the credit squeeze persisted, ratings
agencies were forced to downgrade thousands of
securities after failing to foresee the recent wave of
defaults, particularly in subprime loans.
On
January 30 alone, Standard & Poor’s downgraded more than
8,000 residential mortgage-related securities worth $534
billion. These downgrades have triggered bitter
recriminations, amid a wave of losses in
asset-management companies and banks. Much of these
losses were in AAA securities, which caused a tremendous
loss of confidence.
The
massive credit downgrades have also left policymakers
and analysts scrambling to determine what has gone so
badly wrong. As this search intensifies, some economists
are starting to suspect that the answer lies in a
striking recent change in American household choices, a
shift that could have important implications for
policymakers and investors alike.
It seems
that mathematical models used to predict future default
rates based on past patterns of losses have gone wrong
because they did not adjust to reflect shifts in
household behavior. Or, in another words: The past is
not always a guide to the future. The mortgage borrowers
did not behave as expected. The issue at stake revolves
around the delinquency rates, the proportion of people
who fall behind on their debt repayments. When American
households faced hard times in previous decades, they
tended to default on unsecured loans such as credit
cards and car loans first, and stopped paying their
mortgage only as a last resort. However, in the last
couple of years, households have become delinquent on
their mortgage much faster than trends in the wider
economy might suggest. And this is particularly true for
the less creditworthy subprime borrowers. As a result,
mortgage lenders have started to face losses at a much
earlier stage than in the past.
One
possible explanation is that it has become culturally
more acceptable this decade for people to abandon houses
or stop paying in the hope of renegotiating their home
loans. The shame that used to be associated with losing
a house has ebbed away in recent years, as underwriting
standards were loosened. But consumers may also be
rationally reevaluating the costs that come with
defaulting on different part forms of debt, in the light
of recent bankruptcy- law reforms in the
United States.
This
analysis will lead to a question on what to do when you
don’t know everything. In the economic profession, the
fact has been to pursue the perfect economic forecast
despite abundant evidence that it does not and cannot
exist. They are proponents of rational-expectancy
theory, which assumes that the economy and the
individuals within it act with perfect foresight. On the
other hand, there is a more fashionable school of
behavioral economics, of which practitioners claim that
although people are irrational, their irrationality can
be modeled so precisely that the future can be forecast
with great precision.
In a new
book, “Imperfect Knowledge Economics” (Princeton
University Press, 2007), written by Roman Frydman from
New York University and Michael Goldberg of the
University of Hampshire, Mr. Frydman sets out an
alternative approach to prediction, in which the
forecaster recognizes that his model will inevitably be
less than perfect. Their work has received glowing
praises from Nobel prize-winning economists such as
Kenneth Arrow and Edmund Phelps, who wrote the
introduction to the book, although it is unlikely to
have gone down so well with Robert Lucas, who won the
Nobel for his work on rational expectation.
There is
nothing new in economics about the idea that people must
make decisions based on imperfect knowledge. John
Maynard Keynes observed that “human decision affecting
the future, whether personal or political or economic,
cannot depend on strict mathematical expectation, since
the basis for making such calculations does not exist.”
While
reflecting these insights, imperfect-knowledge economics
still sees a role for economic theory in forecasting.
According to Frydman and Goldberg, to be useful,
economic-forecasting models should be based on
qualitative regularities in the way that market
participants respond to new information, which is
patterns of behavior that are observable and somewhat
predictable. Though this is not perfect, these will
often give a better clue to the future than no model at
all, or models based on rational expectations.
Coming
back to our discussion on a more fundamental economic
explanation for subprime problem: Many economists
observed that in the past, it has usually been assumed
that mortgage defaults occurred due to cash-flow
problems.
Previously in the
United States,
the property market has softened during times of
recession and rising unemployment. But this time, house
prices have fallen even though unemployment has not
risen. Mortgage delinquencies started two years ago, as
soon as house prices stopped rising and then started to
fall. That might be because overstretched households
with unsuitable loans were no longer able to refinance
their way out of trouble, when house prices stopped
rising.
Another
explanation is that people with high loan-to-value
mortgages no longer felt a strong incentive to maintain
payments when house prices started to fall—even if they
were able to. This is because of what we call negative
equity phenomenon, where house prices have fallen below
the value of the loan or will soon do so. Thus, faced
with a choice between keeping their car and maintaining
payments on a house in which they had no equity stake,
some households apparently chose to keep their car. In
fact, International Monetary Fund data show delinquency
rates on prime loans made in 2006 and 2007 are rising
more quickly than delinquency on prime loans made in
2003 or 2004.
Since
the mortgage market is still in flux, of course, it is
very hard to tell which explanation is the most
accurate. However, if consumer behavior is shifted, it
has potentially crucial policy implications. Another
implication is that debt holders may need to rethink
their reliance on ratings. In earlier credit cycles,
bankers would have probably spotted micro-level changes
in household behavior at an early stage, particularly if
they had personal knowledge of their clients. But
because banks have securitized mortgages in recent
years, this contact between lender and borrower has
diminished. As billionaire investor George Soros has
summarized: “Securitization had the effect of
transferring risk from people who are supposed to know
the risk and know the borrowers to people who don’t.”
Investors of the securitized mortgages have tried to
fill this information gap by turning to ratings
agencies. But these agencies have typically predicted
defaults by using macroeconomic models that essentially
extrapolated past trends into the future. Thus, they are
not well-equipped to spot shifts in the fundamental
economic drivers of delinquency or that loan
underwriting standards had been collapsing to a very
loose standard.
In the
securitization industry, the mortgage lenders and
investors have been filtering the loans in recent years
on the basis of FICO scores, which measure cash-flow
management, rather than loan-to-value ratios, which
denote exposure to house prices. As mentioned by
Frydman and Goldberg, the ratings agencies have
generally been better at rating corporate bonds than
rating asset-backed collateralized debt obligations. One
reason for this is that the ratings agencies used both a
mathematical model and judgment of their in-house
specialist when forecasting default probabilities of
corporate bonds; while for subprime-related securities,
they could only use mathematical models, not least
because the instruments were so new. While some
officials inside the ratings agencies have tried to
point out these shortcomings, the sheer volume of
business that has engulfed these agencies has given them
little opportunity to rethink their approach.
Nevertheless, one thing is clear: empirical proof that
relying on backward-looking models alone is not wise.
The forecaster is like an entrepreneur. He uses
quantitative methods, but he also studies history and
relies on intuition and judgment.
****
Gracia S. Ugut, Ph.D. is a full professor at the Asian
Institute of Management. She is also the associate dean
of AIM Executive Education. She teaches financial
engineering and risk management. |
